Examples from the Ejina Basin and Orog - RWTH Publications

Paleoclimatic implications from late Quaternary terrestrial

archives in the Gobi Desert: Examples from the Ejina Basin and

Orog Nuur Basin

Von der Fakultät für Georessourcen und Materialtechnik der

Rheinisch - Westfälischen Technischen Hochschule Aachen

zur Erlangung des akademischen Grades eines

Doktors der Naturwissenschaften

genehmigte Dissertation

vorgelegt von M.Sc.

Kaifeng Yu

aus Jiangsu, China

Berichter: Univ.-Prof. Dr. rer. nat. Frank Lehmkuhl

apl. Prof. Dr. rer. nat. Bernhard Diekmann

Tag der mündlichen Prüfung: 19. August 2016

Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfügbar

Dedicated to my parents and

Cathrin, without whose support,

assistance and patience, this

endeavor would not have been

possible.

谨将此博士学位论文献给我的父

亲俞忠祥，母亲张兰英，以及凯

特琳，是他们无微不至的支持和

帮助使我能够继续追求地学梦想

并完成三年的博士工作。

Gewidmet meinen Eltern und

Cathrin, ohne deren Unterstützung,

Hilfe und Geduld diese Arbeit

nicht möglich gewesen wäre.

i

Preface

It was in 2006, during the journey to the campus of Lanzhou University, the author got the

first impression of topographic and geomorphological characteristics of the arid / semi-arid

central Asia. In 2010, after four years training in the major of physical geography, he wrote

the Bachelor thesis entitled: “Environmental changes in Sangshuyuan region of the Chinese

Loess Plateau over the last ~150 ka” supervised by Prof. Xiuming Liu, employing

paleomagnetism studies comprising anhysteretic remnant magnetization (ARM), isothermal

remnant magnetization (IRM) and coercive force (Bcr) on sediments from a ~9 m loess

sequence on the western flank of the Chines Loess Plateau (CLP). During the Master study

from 2010 to 2013, at Nanjing University, the author reviewed semi-quantified paleoclimate

signals of the Last Glacial Maximum (LGM) and Holocene Optimum (HO) in the arid

northern China under the supervision of Prof. Huayu Lu. The Master thesis is entitled: “A

preliminary quantitative paleoclimate reconstruction of the dune fields of northern China

during the Last Glacial Maximum and Holocene Optimum”.

Aforementioned field, lab, and publication experiences have prompted and enabled the author

to continue with works regarding paleoclimate changes in the arid realm. Since September

12th

, 2013, the author start with the PhD study supervised by Prof. Frank Lehmkuhl, at RWTH

Aachen, in the framework of German Federal Ministry of Education and Research project

(BMBF-CAME: 03G0814A): “Supra-regional signal pathways and long-time archives:

Quaternary monsoon dynamics at the northern margin of the Tibetan Plateau”, German

Research Foundation project (DFG: LE 730/16-1): “Late Pleistocene, Holocene and ongoing

geomorpho-dynamics in the Gobi Desert, South Mongolia” and the financial support from the

China Scholarship Council (CSC: 201306190112). The research focus is using

sedimentological and geochemical signatures incorporated in diversified sediment archives to

disentangle Quaternary paleoclimate change and sedimentary processes in the Ejina Basin and

Gobi Desert of Mongolia. To briefly outline the structure of this dissertation, the chapters are

outlined and described as follows: 1, introduction to the status of the paleoclimate research in

the research area and motivations to carry out this study; 2, geologic, climate, vegetation and

paleoglacial settings of the study area; 3, field investigations, sampling, lab and mathematic

methodologies employed in this study; 4, geochemical characteristics of different terrestrial

archives (first paper); 5, multidisciplinary study to infer the paleoenvironmental change in the

Orog Nuur catchment in the Gobi Desert (second manuscript); 6, bulk-geochemistry of the

lacustrine sediments and potential elements and ratios to decipher integrated

paleoenvironmental and provenance signals (third manuscript); 7, a synthesis concerning the

paleoenvironmental reconstruction and proxy studies; brief abstract of the dissertation in

English and in German language; references, and appendix are added at the end of this

dissertation.

俞凯峰

(Kaifeng Yu)

January 25th

, 2017

Aachen

ii

Table of contents

Page

Preface ................................................................................................................................... i

List of tables ....................................................................................................................... vii

List of figures ...................................................................................................................... ix

List of units, symbols and abbreviations ............................................................................ xv

1. Introduction .................................................................................................................. 1

1.1. Research framework and outline ................................................................................... 2

1.2. Terrestrial archives in the Gobi Desert.......................................................................... 3

1.2.1. Lacustrine sediments ............................................................................................ 3

1.2.2. Aeolian,fluvial and alluvial sediments ................................................................. 4

1.3. State of art of the paleoclimate research in the Gobi Desert and adjacent regions ....... 5

1.3.1. The Ejina Basin and arid northern China ............................................................. 5

1.3.2. The Orog Nuur catchment and Mongolian Plateau .............................................. 6

1.4. Research motivations .................................................................................................... 7

2. Regional setting .......................................................................................................... 11

2.1. Geologic and geomorphic setting ................................................................................ 11

2.1.1. The Ejina Basin .................................................................................................. 11

2.1.2. The Orog Nuur catchment .................................................................................. 13

2.2. Climate regime ............................................................................................................ 14

2.2.1. Climate of the Ejina Basin ................................................................................. 14

2.2.2. Climate of the Orog Nuur catchment ................................................................. 15

2.3. Soils and vegetation cover ........................................................................................... 16

2.4. Paleo- and modern glacial setting ............................................................................... 17

2.5. Anthropo-zoogenic setting .......................................................................................... 17

3. Materials and methods ............................................................................................... 19

3.1. Field work and sampling strategy ............................................................................... 19

3.2. Laboratory methods ..................................................................................................... 20

3.2.1. Chronological methods ....................................................................................... 20

3.2.2. Grain size analysis .............................................................................................. 21

3.2.3. X-ray fluorescence analysis and X-ray diffraction ............................................. 22

3.2.4. Electrical conductivity and magnetic susceptibility measurements ................... 23

3.2.5. Carbon, nitrogen, sulfur, TOC and CaCO3 analyses .......................................... 23

3.2.6. Pollen, non-pollen palynomorphs and palynofacies analyses ............................ 23

3.2.7. Ostracod analysis ................................................................................................ 24

3.3.Mathematical methods: Correlations and multivariate statistics .................................. 24

4. Discriminating sediment archives and sedimentary processes in the arid

endorheic Ejina Basin, NW China using a robust geochemical approach ........... 26

iii

4.1. Introduction ................................................................................................................. 27

4.2. Regional Setting .......................................................................................................... 30

4.3. Methods ....................................................................................................................... 31

4.3.1. Sampling strategy ............................................................................................... 31

4.3.2. Chronological methods ....................................................................................... 32


4.3.4. X-ray fluorescence analysis and X-ray diffraction ............................................. 36

4.3.5. Correlations and multivariate statistics .............................................................. 37

4.4. Results ......................................................................................................................... 38

4.4.1. Lithology and chronology .................................................................................. 38

4.4.2. Granulometric composition and pre-assessment of archives ............................. 40

4.4.3. Elemental and mineralogical composition of archives ....................................... 43

4.4.4. Element assemblages and loadings as derived from conventional and robust

factor analysis ............................................................................................................... 45

4.5. Discussion ................................................................................................................... 47

4.5.1. Late Quaternary lithostratigraphy at the northern margin of the Ejina Basin .... 47

4.5.2. Elemental characteristics of sediment archives .................................................. 48

4.5.2.1. Elemental characteristics compared to granulometric composition .... 48

4.5.2.2. Elemental characteristics of sediment archives ................................... 49

4.5.3. Geochemical interpretation of sedimentary processes ....................................... 51

4.6. Conclusions ................................................................................................................. 53

5. Lake sediments documented late Quaternary humid pulses in the Gobi Desert of

Mongolia: Vegetation, hydrologic and geomorphic implications .......................... 55

5.1. Introduction ................................................................................................................. 56

5.2. Study Area ................................................................................................................... 59

5.2.1. Geologic and geomorphic setting ....................................................................... 59

5.2.2. Climatic and vegetation setting .......................................................................... 60

5.2.3. Paleo- and modern glacial setting ...................................................................... 62

5.3. Material and methods .................................................................................................. 62

5.3.1. Cores retrieval and sampling strategy ................................................................ 62

5.3.2. Lithology and chronology .................................................................................. 63


5.3.4. Electrical conductivity and magnetic susceptibility measurements ................... 65

5.3.5. Carbon, nitrogen, sulfur, TOC and CaCO3 analyses .......................................... 66

5.3.6. X-ray fluorescence analysis, color analysis and multivariate statistics ............. 66

5.3.7. Palynology .......................................................................................................... 67

5.3.8. Ostracod analysis ................................................................................................ 68

5.4. Results ......................................................................................................................... 68

5.4.1. Bayesian chronology and lithostratigraphic units .............................................. 68

5.4.2. Bulk-geochemistry of the lacustrine sediments ................................................. 71

5.4.3. Palynological results .......................................................................................... 71

5.4.3.1. Unit 1 (LPZ 1; 1,335-1,235 cm; ~50-~43 cal ka BP) .......................... 72


iv


5.4.3.4. Unit 4 (LPZ 4; 1,057-660 cm; ~35-~24 cal ka BP) ............................. 73

5.4.3.5. Unit 5 (LPZ 5; 660-390 cm; ~24-~19 cal ka BP) ................................ 73




5.4.3.9. Unit 9 (LPZ 9; 205-25 cm; ~11-? cal ka BP) ...................................... 74

5.4.4. Ostracod species and valve abundances ............................................................. 75

5.5. Discussion ................................................................................................................... 76

5.5.1. Vegetation dynamics and hydrologic history of the Orog Nuur ........................ 76

5.5.1.1. Unit 1 (~50-~43 cal ka BP): relatively humid Artemisia steppe around

a shallow lake ................................................................................................... 77

5.5.1.2. Unit 2 (~43-~37 cal ka BP): relatively dry and fluctuating Artemisia

steppe and water surplus to the lake ................................................................. 78

5.5.1.3. Unit 3 (~37-~35 cal ka BP): relatively dry Artemisia steppe and lake

stabilization ....................................................................................................... 79

5.5.1.4. Unit 4 (~35-~24 cal ka BP): relatively humid Artemisia steppe and

transition from maximum high to quite low lake levels ................................... 81

5.5.1.5. Unit 5 (~24-~19 cal ka BP): Artemisia steppe and playa to shallow lake

.......................................................................................................................... 81

5.5.1.6. Unit 6 (~19-~18 cal ka BP): Artemisia steppe and sand accumulation 82

5.5.1.7. Unit 7 (~18-~13 cal ka BP): Artemisia steppe, higher lake level and

decreasing river influence ................................................................................. 82

5.5.1.8. Unit 8 (~13-~11 cal ka BP): Establishment of deserts around a strongly

shrinking lake .................................................................................................... 83

5.5.1.9. Unit 9 (~11-? cal ka BP): desert steppe and a shallow lake of high

salinity and alkalinity ........................................................................................ 83

5.5.2. Geomorphic evidence of the past environmental change in the Orog Nuur

catchment ..................................................................................................................... 86

5.5.2.1. Glacial advances in the higher reaches of the Orog Nuur catchment .. 86

5.5.2.2. Ancient shorelines of the Orog Nuur ................................................... 88

5.5.3. Regional synthesis - humid pulses inferred from different archives in the

vicinity of the Gobi Desert ........................................................................................... 88

5.5.3.1. Marine Isotope Stage 3 ........................................................................ 89

5.5.3.2. Marine Isotope Stage 2 ........................................................................ 90

5.5.3.3. The Holocene ....................................................................................... 90

5.5.4. Potential driving mechanisms for the late glacial climate change in the Gobi

Desert ........................................................................................................................... 91

5.5.4.1. Humid pulses in the Gobi Desert ......................................................... 92

5.5.4.2. Playa phases in the Gobi Desert .......................................................... 93

5.6. Conclusions ................................................................................................................. 94

6. Geochemical imprints of coupled paleoenvironmental and provenance change in

the lacustrine sequence of Orog Nuur, Gobi Desert of Mongolia .......................... 97

6.1. Introduction ................................................................................................................. 98

v

6.2. Regional setting ......................................................................................................... 101

6.3. Material and methods ................................................................................................ 101

6.3.1. Core retrieval and Bayesian age-depth modelling ........................................... 101

6.3.2. Grain size analysis ............................................................................................ 103

6.3.3. TOC, CaCO3 and C/Natomic analyses ................................................................ 104

6.3.4. X-ray fluorescence scanning and X-ray powder diffraction analysis .............. 105

6.3.5. Robust multivariate statistics ........................................................................... 105

6.4. Results ....................................................................................................................... 106

6.4.1. Lithostratigraphic units and Bayesian chronology ........................................... 106

6.4.2. Elements/minerals downcore variance and multi-proxy assemblages revealed by

multivariate statistics .................................................................................................. 108

6.5. Discussion ................................................................................................................. 110

6.5.1. Evaluating the quality of the X-ray scanning data ........................................... 110

6.5.2. Governing factors for sediment bulk-geochemistry ......................................... 111

6.5.2.1. Factor 1 (26.4%): Grain-size variation of the lacustrine sediments .. 111

6.5.2.2. Factor 2 (15.5%): Authigenic productivity of the saline/alkaline

lacustrine environment .................................................................................... 112

6.5.2.3. Factor 3 (8.4%): Organic material content ........................................ 113

6.5.2.4. Factor 4 (25.9%): Terrigenous components via riverine input and

quasi-constant aeolian input ........................................................................... 113

6.5.3. Biplots of elements and scores plot as indicators of a distinct provenance signal

and high authigenic productivity of the lake system in the Holocene ....................... 117

6.5.3.1. The disparate provenance signatures and authigenic productivity of the

Holocene sediments ........................................................................................ 118

6.5.4. Summary of coupled paleoenvironmental, weathering, catchment land surface

processes, and provenance signals over the past ~50 ka BP in the Orog Nuur

catchment ................................................................................................................... 121

6.5.4.1. Unit I (~50-~19 cal ka BP): shallow to lentic water body, moderate

catchment weathering ..................................................................................... 121

6.5.4.2. Unit II (~19-~18 cal ka BP): playa phase and minor riverine inflow 121

6.5.4.3. Unit III (~18-~13 cal ka BP): shallow saline lake, moderate weathering

in higher catchment ......................................................................................... 122

6.5.4.4. Unit IV (~13-~11 cal ka BP): playa phase with slightly more riverine

inflows relative to LGM ................................................................................. 122

6.5.4.5. Unit V (<~11 cal ka BP): lentic hypersaline lake with abruptly

increased terrigenous inputs and authigenic productivity .............................. 123

6.6. Conclusions ............................................................................................................... 124

7. Synthesis: Paleoclimatic implications from late Quaternary terrestrial archives in

the Gobi Desert ......................................................................................................... 127

7.1. Bulk-geochemistry of different terrestrial archives in the arid realm and their

corresponding implications for the paleoenvironmental, pedogenesis and diagenesis

processes ..................................................................................................................... 127

7.2. Sedimentological, palynological and ostracod studies of the lacustrine sediments and

references to the paleohydrological and vegetation history ....................................... 128

vi

7.3. Geochemical imprints of the lacustrine sediments and integrated paleoenvironmental

and provenance implications ..................................................................................... 129

7.4. Outlook ...................................................................................................................... 130

References ........................................................................................................................ 132

Appendix .......................................................................................................................... 159

R script employed in the dissertation ............................................................................... 160

Acknowledgements .......................................................................................................... 163

Abstract ........................................................................................................................... 167

Zusammenfassung .......................................................................................................... 169

vii

List of Tables

Tab. 2-1: Inventory of the plutonic rocks, volcanic/subvolcanic rocks and Quaternary deposits

in the Orog Nuur catchments. For geologic mapping refer to Fig. 1-3. This set of sources

rocks is of vital importance concerning the bulk-geochemistry discussed in Chapter 6 ......... 16

Tab. 4-1: Results of the AMS 14

C-dating. Radiocarbon ages were calibrated to calendar years

with 2 σ standard deviation using the CALIB rev.7.1 (Stuiver et al., 1998) with the latest

calibration dataset (Intcal13.14c, Reimer et al., 2013) ............................................................. 38

Tab. 4-2: Summary of water content, equivalent dose value (De), dose rate, number of used

aliquots and luminescence ages of samples from study sections. 238

U, 232

Th and 40

K

concentrations were determined by Neutron Activation Analysis (NAA). Absolute uncertainty

of the water content was assumed to be < ±5%. Due to saturation of the equivalent dose, brief

constrained luminescence ages are presented here. More details concerning the method see

Chapter 4.3.2 ............................................................................................................................ 39

Tab. 4-3: Results of the oxides of ten major elements expressed as normalized wt % and other

trace elements concentrations (ppm) of sediment archives as determined by X-ray

fluorescence analysis. Decimal precision of the reported concentrations were revised

according to the formats of their counterparts in the UCC data set (Rudnick and Gao, 2003).

Root mean square errors (RMS) and lower limit of detection (LOD) of each element see

Spectro Xepos (2007) ............................................................................................................... 43

Tab. 4-4: Standard deviations, variance / cumulative proportions, element loadings for each

factor, and communalities for each element. High positive or negative (>0.75) loadings are

shown in boldface, proportions in %. ....................................................................................... 46

Tab. 5-1: The AMS 14

C ages were calibrated to calendar years with 2 σ standard deviation

using the CALIB rev. 7.1 (Stuiver et al., 1998) with the latest IntCal13.14c (Reimer et al.,

2013). According to the estimated real age based on the lithostratigraphy, raw radiocarbon

ages are subtracted by a reservoir effect of ~3,400±2,100 a BP before computing to calibrated

ages and generating the Bayesian age-depth model as given in Fig. 5-4 ................................. 63

Tab. 5-2: Synthesis of lacustrine, pelagic, aeolian and glacial sequences (spanning at least the

MIS 3) in the vicinity of the Gobi Desert ................................................................................. 85

Tab. 6-1: The AMS 14




ages are subtracted by a reservoir effect of ~3,400±2,100 a BP before computing to calibrated

ages and generating the Bayesian age-depth model as given in Fig. 6-2 ............................... 103

Tab. 6-2: Standard deviation, variance/cumulative proportion, multi-proxies loading for each

factor, and communalities for each proxy generated by robust multivariate statistics. High

positive or negative (> 0.75) loadings are marked in boldface. Blank values represent those

loadings which are under significance level in robust multivariate statistics. Communalities

for each element are computed by taking sum of the seared loadings, it can be referred to the

length of the element arrows in loadings plot (Fig. 6-4A), a value greater than 0.90 suggests a

well representative of this proxy in the statistical model. Factors 1 (26.4%) and 4 (25.9%)

explain more than have of the variance, they are therefore illustrated in the biplots of Fig. 6-4

for further implications. More details concerning the statistics refer to Yu et al. (2016) ...... 106

viii

Tab. 6-S1: Synoptic description of the coupled paleoenvironmental and provenance signatures

during the past ~50 ka BP as inferred from the lacustrine sequence of Orog Nuur. A schematic

model comprising these factors is proposed in Fig. 6-6 in the main text ............................... 125

Tab. A1: Ratios from major elements to oxides and corresponding functions ...................... 159

ix

List of Figures

Fig. 1-1: Atmospheric circulations over the central and eastern Asia, with panel A denoting

the summer pattern and panel B representing winter pattern. Solid lines indicate ca. 3,000 m

a.s.l. airflow while dashed lines indicate earth surface flow at ca. 600 m (adapted from Benn

and Owen, 1998; Lehmkuhl and Haselein, 2000; Barry and Chorley, 2009; An et al., 2014;

Loibl, 2015). Study areas of Ejina Basin and Orog Nuur Basin are marked as white rectangle

in the panels. More detailed analyzed sections in this dissertation refer to following Figs. 2-2

and 2-3. ....................................................................................................................................... 2

Fig. 2-1: Photographs of representative contexts in the study area: (A) Gobi surface and

Precambrian intrusive bedrocks with background of Mountains in Mongolia. (B) Megadunes

on the northern margin of Badain Jaran Desert. (C) Common observed Haloxylon

ammodendron on the gobi surface. (D) Unparallelled sand ripples on the dune surface. (E)

Playa surface of western Juyanze Basin with particularly high contend of evaporite minerals.

(F) The ancient city Khara Khoto (synonyms: Black City, abandoned in 1,372 A.D.) and

nearby accumulated dunes. ...................................................................................................... 12

Fig. 2-2: Geologic map and section locations in the northern Ejina Basin: Section HC8 is an

escarpment between the western and eastern Juyanze sub-basin. Sections KK1 and KK2 are

located by the braided river channel near ancient of Khara Khoto. Sections WT1 and WT2 are

yardangs located on the margin of Wentugaole Sub-basin and Badain Jaran Desert. Black

(dash-) lines denote strike-slip neotectonics compiled from Wünnemann et al. (2007), Hölz et

al. (2007), Hartmann et al. (2011), Cunningham (2013) and Rudersdorf et al. (2015). D100

correspond to the drilling core from Wünnemann (1999) and Wünnemann et al. (2007) ....... 13

Fig. 2-3: Present-day distribution of geologic terranes in the Orog Nuur catchment (marked

with dashed line). Northern region of Orog Nuur Basin is dominated with Devonian-Triassic

felsic quartz syenite, granite and plagiogranite, accompanied by intermediate mafic fractions.

Southern region of Orog Nuur exhibit increasing fraction of mafic and ultramafic basalt and

andesite-basalt rocks. Alkaline syenite occurs sporadically in the lower Orog Nuur

catchments. The depositional basin is bounded by eastern Khangai in the north and the Gobi

Altai Mountains in the south. More details concerning representative minerals see Tab. 2-1.

Concerning the geomorphological and topographic settings refer to Fig. 5-2 of the same

region ........................................................................................................................................ 15

Fig. 2-4: Schematic north-south oriented topographic transect and corresponding vegetation

patterns. The eastern parts of Khangai Mountains (higher reach of the Orog Nuur catchment)

with altitudes between 3,500 m and 3,700 m a.s.l. have no modern glaciers, while small ice-

fields may reach down to 3,200 m a.s.l. in the western Khangai Mountains (Lehmkuhl and

Lang, 2001) .............................................................................................................................. 18

Fig. 4-1: Simplified geologic map and section locations in the northern Ejina Basin: (A)

Section HC8 is an escarpment between the western and eastern Juyanze sub-basin. Sections

KK1 and KK2 are located by the braided river channel near ancient of Khara Khoto. Sections

WT1 and WT2 are yardangs located on the margin of Wentugaole Sub-basin and Badain

Jaran Desert. Black (dash-) lines denote strike-slip neotectonics compiled from Wünnemann

et al. (2007), Hölz et al. (2007), Hartmann et al. (2011), Cunningham (2013) and Rudersdorf

et al. (2015). Black bold lines denote schematic elevation cross-section in panel B. D100

correspond to the drilling core from Wünnemann, 1999 and Wünnemann et al., 2007. (B)

Cross-section from Khara Khoto to Juyanze (playa) and Wentugaole (sub-basin) ................. 29

x

Fig. 4-2: Photographs of representative geomorphological contexts from the northern Ejina

Basin: (A) Inverted channels near Khara Khoto. (B) Gobi surface and Precambrian intrusive

bedrocks in Wentugaole sub-basin. (C) Khara Khoto and nearby accumulated dunes. (D)

Yardangs in southern Wentugaole sub-basin. (E) Playa surface of western Juyanze Basin. (F)

Megadunes on the northern margin of Badain Jaran Desert .................................................... 31

Fig. 4-3: Lithostratigraphy and generally constrained chronology framework. Chronology

framework is labeled, and 2-sigma ranges for calculated radiocarbon ages are reported in Tab.

4-1. U1 is the lowermost aeolian sands (except for WT1), U2 is lacustrine stratum, and U3 is

fluvial sands (mixed with gravels) unit (e.g., Fig. 4-2A). U3 of HC8 is especially thick

mixture of poorly-sorted coarse alluvial sands and gravels ..................................................... 32

Fig. 4-4: Example of wave-length dispersive spectrums of X-ray diffraction (XRD). Panels A

to E denote aeolian sands, lacustrine clay, fluvial sands, sandy loess, alluvial gravels,

respectively from section HC8 (with depth). The powdered samples were scanned from 3° to

40° at °2θ, angle spectrums are conducted by Xpert Highscore software, the classification of

impulses following Hartmann & Wünnemann (2009). The number of counts is illustrated as

reference for e.g., aeolian sands ............................................................................................... 35

Fig. 4-5: Conceptual modal of the robust geochemical approach to discriminate archives and

sequence sedimentary processes .............................................................................................. 36

Fig. 4-6: Ternary diagrams of granulometric properties and A-CN-K (molar proportions): (A)

Granulometric ternary diagram. (B) A-CN-K ternary diagram, CaO is corrected to silicate

carbonate following the procedure of McLennan (1993). The Chemical Index of Alteration

(CIA, Nesbitt & Young, 1982) is related on the right side of the A-CN-K triangle.

Abbreviations: A=Al2O3, C=CaO, N=Na2O, K=K2O, Ka=Kaolinite, Gi=Gibbsite,

Ch=Chlorite, Il=illite, Mu=muscovite, Pl=plagioclase, Bi=Biotite, Ks=alkali feldspar,

Ch=chlorite, Cp=clinopyroxene, Hb=hornblende .................................................................... 40

Fig. 4-7: Granulometric properties of archives: (A) aeolian sands (n=51), (B) lacustrine clay

(n=19), (C) fluvial sands (n=18), (D) sandy loess (n=2), (E) alluvial gravels (n=2), different

axis scales are used ................................................................................................................... 41

Fig. 4-8: Spearman’s rank correlation coefficients between geochemical properties and grain

size parameters (all scales of red to blue units have passed 0.05 significant level). Not all

elements exhibit grain-size dependence (e.g., As, Nd, S and Al). Red pixels denote positively

correlated and blue ones depict negative correlations. These features can also shed light on

elemental characteristics of archives. C, fSi, mSi, cSi, fS, mS and cS denote the particle size

of clay, fine silt, medium silt, coarse silt, fine sand, medium sand and coarse sand,

respectively ............................................................................................................................... 42

Fig. 4-9: Loading and score plot of geochemical data set: (A) Loading plot of conventional

factor analysis, with thirty elements. (B) Score plot of conventional factor analysis, with thirty

elements. (C) Loading plot of robust factor analysis, with sixteen elements. (D) Score plot of

the robust factor analysis, with sixteen elements. Originally, the maximum direction of

variance will be regarded as axes, however, rotation is performed to get varimax rotated axis

(i.e., X-and Y-axis here) during the factor analysis to provide a higher loading for each

elements (Tab. 4-4). Cosine of the angle between the elements arrow and the component

(varimax rotated axes) are factor loadings, length of the arrows are corresponding to the

element communalities. Compared with conventional factor analysis (Panels A and B), panels

C and D exhibit higher cumulative proportions for RFs 1 and 2. Dashed line circles in panels

xi

A and C denote communality estimates (Tab. 4-4) and thereby assess the quality for each

element (> 0.9 can be regarded as good quality) in this factor modal. Panels B and D illustrate

that multivariate statistic of elemental fingerprints appears to be a practical method to

discriminate the archive types, of which mistakes may be yielded from field investigations

and GSD observations alone .................................................................................................... 44

Fig. 4-10: Principle component scores, main mineral compositions and granulometric

properties of section HC8, RF1: weathering intensity, RF2: silicate-bearing mineral

abundance (different within lacustrine and aeolian phase), RF3: saline magnitude of

sediments, and RF4: quasi-constantly aeolian input. Lithologic patterns refer to Fig. 4-3. Clay

mineral group includes smectite, illite, kaolinite and chlorite. C, fSi, mSi, cSi, fS, mS and cS

denote the particle size of clay, fine silt, medium silt, coarse silt, fine sand, medium sand and

coarse sand, respectively .......................................................................................................... 51

Fig. 5-1: Location of the study area and paleoclimate records mentioned in the text (1, Kanas

Basin; 2, Darhad Basin; 3, Hovsgol Nuur; 4, Shaamar section; 5, Baikal Lake; 6, Kotokel

Lake; 7, Wulagai Lake; 8, Manas Lake; 9, Luanhaizi Lake; 10, Wuliangsu Lake; 11, Japan

Sea; 12, Lake Biwa; 13, Khangai Glacier; 14, Altai Glacier. More details refer to Tab. 4-3).

Panel A delineates schematic vegetation cover of Mongolia. Large-scale air mass controls are

marked with arrows, comprising East Asian Summer Monsoon (EASM), Indian Summer

Monsoon (ISM) and Westerlies. Ulaan Nuur (Lee et al., 2011, 2013) and Bayan Tohomin

Nuur (Felauer et al, 2012) are the only two reported lacustrine archives in the central to

southern Mongolia. Panel B exhibits north-south oriented topographic transect and

corresponding vegetation patterns. The eastern parts of Khangai Mountains (higher reach of

the Orog Nuur catchment) with altitudes between 3,500 m and 3,700 m a.s.l. have no modern

glaciers, while small ice-fields may reach down to 3,200 m a.s.l. in the western Khangai

Mountains (Lehmkuhl and Lang, 2001) ................................................................................... 57

Fig. 5-2: Geomorphological map of the Orog Nuur Basin, three sets of lake beaches were

synthesized based on previous investigations. Section A1 is located at the third (highest) set of

sand ridge, more details see Chapter 5.5.2.2. General content is adapted after J. Grunert and F.

Lehmkuhl, and original cartography by T. Bartsch ................................................................. 59

Fig. 5-3: Schematic lithologic column, photographs of sliced drilling core ONW II along with

the reference logs. Abbreviations: cl, clay; fsi, fine silt; msi, medium silt; csi, coarse silt; fs,

fine sand; ms, medium sand, cs, coarse sand. Mudstone or siltstone is represented in

schematic colors according to core description. In general, greyish calcareous clays (upper

fraction), siltstone and intercalated sandstones constitute the entire lacustrine sequence ....... 61

Fig. 5-4: Polylinear regression age-depth model of core ONW II generated from the

radiocarbon ages. Reservoir effect is corrected by (i) late Pleistocene/Holocene transition

(Termination I) as inferred from the suite of granulometric, palynological, ostracod and

geochemical studies (marked as black triangle), and (ii) the Younger Dryas related sand layer

(Ulaan Nuur, Lee et al., 2013). Frequency-dependent magnetic susceptibility (χFD), electrical

conductivity (EC), CaCO3, TOC and C/Natomic ratios of parallel core ONW I are employed to

cross-validate the age model of core ONW II. More details with respect to the

lithostratigraphy and chronological framework see text in Chapter 5.4.1 ............................... 64

Fig. 5-5: Lithologic units of the core ONW II, and corresponding granulometric, geochemical,

and color variances. The grey bars depict the primary sand layers inferred from the

granulometric and lithostratigraphic characteristics of the core .............................................. 65

xii

Fig. 5-6: Diagram of analyzed taxa of pollen, fungal spores, aquatic algae and non-pollen

palynomorphs through the core ONW II plotted against the age and depth. All taxa were

aggregated into groups consist of arboreal plants (trees), shrubs, subshrubs/herbs/grasses,

aquatics, as well as coprophilous fungi and Glomus ................................................................ 67

Fig. 5-7: Scanning electron micrographs of ostracod valves derived from core ONW II

(external views). 1, juvenile Candona neglecta left valve (LV); 2-3, Leucocythere

dorsotuberosa, 2, female right valve (RV), 3, male LV; 4-11, Limnocythere inopinata, 4,

female LV, 5, female RV, 6, noded female LV, 7, noded female RV, 8, male LV, 9, male RV,

10, juvenile LV with spines along posteroventral margin, 11, juvenile RV with spines along

posteroventral margin; 12, juvenile Pseudocandona sp. LV; 13, Ilyocypris cf. inermis RV, 14,

Ilyocypris manasensis var. cornae LV, 15, Ilyocypris cf. bradyi LV ...................................... 69

Fig. 5-8: Absolute abundances of determined ostracod species, number of taxa and number of

valves plotted against lithologic units ...................................................................................... 72

Fig. 5-9: Semi-quantified thermal, moisture and hydrogeochemical history as inferred from

the palynological and geochemical studies. Palynological indices (curves a and b) are inferred

from pollen concentration and Artemisia/Chenopodiaceae ratios, respectively, whilst the

geochemical indices (curves c and d) i.e., factor scores, are entrained from multivariate

statistics based on the bulk-geochemistry of the sediments (Yu et al., 2016). Fossil pollen

concentration is linked to the thermal anomalies (e.g., Conroy et al., 2009). Considering the

authigenic productivity, the Holocene may experience a distinct lacustrine phase with higher

content of calcite and carbonates. In this study, the vertical blue bars depict humid pulses,

while yellow bars denote the playa phase inferred from the granulometric and

lithostratigraphic characteristics of the core. Furthermore, as stated in Chapter 5.4.3, the

pollen assemblages preserved in the lacustrine archive may indicate merely local vegetation

signals surrounding the lower catchment of the Orog Nuur. Therefore, inconsistency between

the reconstructed effective moisture and lake level fluctuations may occur. Northern

hemisphere temperature (e): Johnsen et al. (2001), H events: Bond et al. (1993), 45 °N

summer insolation (f): Whitlock and Bartlein (1997), west subtropical pacific SST (g) and sea

level changes (h): Tudhope et al. (2001). In addition, the MIS 3/MIS 2 transition in this study

is some 24 cal ka BP, which is largely in agreement with that in Martinson et al. (1987) ...... 78

Fig. 5-10: Synthesis and comparison of paleoenvironmental variability in Mongolia and

adjacent regions since the Marine Isotope Stage 3 as developed in this study and other

selected records from literatures: 1, Kanas Glacial (Xu et al., 2009; Zhao et al., 2013); 2,

Manas Lake (Rhodes et al., 1996); 3, Darhad Glacial (Gillespie et al., 2008); 4, Hovsgol Nuur

(Murakami et al., 2010); 5, Luanhaizi Lake core LH2 (Mischke et al., 2005b); 6, Baikal Lake

(Williams et al., 1997; Prokopenko et al., 2001); 7, Kotokel Lake (Shichi et al., 2009); 8,

Retreat of southern boundary of the Gobi Desert (Feng et al., 2007); 9, Shaamar Section (Ma

et al., 2013); 10, Wuliangsu Lake (Selvaraj et al., 2012); 11, Wulagai Lake (Yu et al., 2014);

12, Biwa Lake (Xiao et al., 1997); 13, Japan Sea (Nagashima et al., 2007); 14, Dust model in

central Asian desert (Nilson and Lehmkuhl, 2001). Here, the vertical blue bars depict humid

pulses, while yellow bars denote the playa phase inferred from the granulometric and

lithostratigraphic characteristics of the core. Calibrated radiocarbon ages were labeled as

black triangles near the bottom axis. Climate intervals including the Younger Dryas (YD) and

Last Glacial Maximum (LGM) were demarcated near the uppermost axis. More details

concerning the proxies of each archive refer to Tab. 5-2 and locations refer to Fig. 5-1A ..... 80

Fig. 5-11: Synoptic compilation of hydrologic, vegetation and geomorphic history of the Orog

Nuur catchment over the last ~50 ka. Playa & dune phase is recovered from the granulometric

xiii

and lithostratigraphic characteristics of the core ONW II. More details see Chapter 5.5.1.

Glacial advance in Khangai Mountains refer to Rother et al. (2014) and Lehmkuhl et al.

(2016). As mentioned in Chapter 5.5.3.3, in the Holocene, the high termpertature induced

higher evapouration might have triggered lower annual moisture in the catchment (see A/C

ratios in Fig. 5-9), it has however less impacts on the better hydrological condition of the

water body as revealed by higher authigenic productivity (Fig. 5-9) in the more alkaline water

column (see Chapter 6.5.2.2, and Fig. 6-6). ............................................................................. 92

Fig. 5-S1: Diagram of analyzed taxa of pollen, fungal spores, aquatic algae and non-pollen


aggregated into groups consist of panel (A) arboreal plants, humidity indicators, aquatic algae,

and panel (B) aridity indicators and grazing indicators ........................................................... 96

Fig. 6-1: Present-day distribution of geologic terranes in the Orog Nuur catchment. Northern

region of Orog Nuur Basin is dominated with Devonian-Triassic felsic quartz syenite, granite

and plagiogranite, accompanied by intermediate mafic fractions. Southern region of Orog

Nuur exhibit increasing fraction of mafic and ultramafic basalt and andesite-basalt rocks.

Alkaline syenite occurs sporadically in the lower Orog Nuur catchments. The depositional

basin is bounded by eastern Khangai in the north and the Gobi Altai Mountains in the south

.................................................................................................................................................. 99

Fig. 6-2: Lithostratigraphy and Bayesian age-depth modelling for core ONW II. R package

Bchron (Parnell et al., 2008) was employed. Gray shaded area represents maximum and

minimum age-depth estimates based on 100,000 smooth spline fits through the calibrated age

distributions (95% highest density region). The Bayesian age-depth model was further cross-

validated by several independent approaches: (i) two ages from the parallel core ONW I

(marked as black rectangles), (ii) sand layer and luminescence-determined age model of

Ulaan Nuur (Lee et al., 2011), (iii) downcore variance of CaCO3 concentration .................. 102

Fig. 6-3: Downcore variance of abundance of the major and trace elements expressed by

integrated peak areas (total counts per second), accompanied by the C/Natomic ratios, CaCO3

content, Rb/Sr ratio (lower values= higher catchment weathering; Davies et al., 2015), Al/Si

ratio (lower values=higher coarse-grained sands; Davies et al., 2015), and the XRD

determined minerals in sparser intervals. Silicon and K share similar patterns as Al, whilst Mn

and Co share similar patterns as Fe. Further details are presented in the supplementary

material of this paper. The grey bars depict the sand layers inferred from the granulometric

and lithostratigraphic characteristics of the core, and the dashed line mark the transition of

Termination I. The consistency between the grey bars and the elements spikes implies that the

sand fractions (Factor 1 as determined by robust statistics; Fig. 6-4) have played a pivotal role

in the alteration of bulk-geochemistry of the lacustrine sediments. ....................................... 104

Fig. 6-4: Loadings and scores plot of the multi-proxy data set. Panel A: loading plot of the

robust Factor 1 (grain-size) vs. 4 (terrigenous input). Panel B: score plot of the robust Factor 1

vs. 4. In multivariate statistics, the maximum direction of variance is regarded as axes.

Rotation is performed to get varimax rotated axis (x- and y-axis in Panel A) during the

analysis attempting to archive a higher loading for each elements (Tab. 6-2). Cosine of the

angle between the proxies and varimax rotated axes are factor loadings, length of the arrows

are corresponding to the proxy communalities. In panel B, all units of samples are defined

based on their lithostratigraphic units .................................................................................... 109

Fig. 6-5: Bivariate plots of major and trace elements in peaks areas (y-axis) against Al/Si (x-

axis) according to their stratigraphic units. Al/Si ratios indicate generally the grain-size

xiv

compositions of the bulk sediments, which display lower values/spikes according to the

coarse-grained sand layers (Fig. 6-3). The biplot of Ca vs. Al/Si seems like a useful tool to

properly discriminate the instinct (i) source rocks changes, and (ii) authigenic productivity in

comparison with the late Pleistocene units. This distinct Holocene samples characteristics are

more clearly elaborated in the scores plot of Fig. 6-4B ......................................................... 115

Fig. 6-6: Schematic model of the integrated past environmental change, land surface

processes and provenance signals (carbonate geochemistry) of the Orog Nuur catchment over

the last ~50 ka. The eastern parts of Khangai Mountains (higher reach of the Orog Nuur

catchment) with altitudes between 3,500 m and 3,700 m asl. have no modern glaciers. More

details regarding glacial and vegetation developments in the catchment are given in Lehmkuhl

and Lang (2001), Rother et al. (2014), and Lehmkuhl et al. (2016). Distinct geochemical

signals for each phase also refer to the table in supplementary material of this paper .......... 120

Fig. 6-S1: Dispersion plots of the major and trace element concentrations (peak areas)

according to their stratigraphic units. Numbers of observations for each unit are marked on the

upmost axis. The median values are linked with dash lines, while the upper and lower

quartiles are marked as error bars .......................................................................................... 126

Fig. 7-1: Synoptic compilation of hydrologic, vegetation and glacial history of the Orog Nuur

catchment over the last ~50 ka. Playa & dune phase is recovered from the granulometric and

lithostratigraphic characteristics of the core ONW II. Likewise, this late Glacial high lake

levels are also recorded in the Ejina Basin as lacustrine facies.............................................. 127

xv

List of Units, Symbols and Abbreviations

Units of the International System (SI) have been guidelines in this dissertation. Unless

otherwise specified, the abundances of trace element are based on the total content by weight

of those in oven-dried material.

Units:

A.D. anno Domini, used to label or number years in Julian and Gregorian calendars

a.s.l. above sea level

cal ka BP calculated thousand years before present (1950 A.D.)

ppm parts per million; 1 ppm = 1 µg/g

Symbols:

AMS Accelerator mass spectrometry

ELA Equilibrium Line Altitude

pH Minus logarithm, base 10, of H+ concentration

Abbreviations:

AMOC Atlantic Meridional Overturning Circulation

CNS Analysis of total carbon, nitrogen and sulfur

ECORD European Consortium for Ocean research Drilling

EGU European Geosciences Union

ESR Electron spin resonance

gLGM global Late Glacial Maximum

GSD Grain size distribution

INQUA International Union for Quaternary Research

IODP Integrated Ocean Drilling Program

ITCZ Intertropical Convergence Zone

MIS Marine Isotope Stage

OSL Optically stimulated luminescence

REE Rare earth element

XRD X-ray Diffraction

XRF X-ray fluorescence

YD Younger Dryas

1

“Arid regions are areas of concern, because of

population growth, the impacts of desertification and of

natural phenomena, particularly droughts” (David S.G.

Thomas, 1996)

1. Introduction

Being a fragile ecological system, environmental change in arid central Asia exerts

tremendous influence on a significant percentage of population concerning nomadic migration

and alternation of political regimes (Pederson et al., 2014). The arid central Asia (Fig. 1-1) is

sensitive to paleoclimate changes. The comprehensive framework of alluvial fans, terraces,

periglacial mass movement, aeolian and lacustrine sediments has documented the manner of

the paleoclimate changes as well as the markedly anthropological influences since the late

Holocene (Lehmkuhl and Haselein, 2000). Lehmkuhl et al. (2003) reported a successive

geomorphological section in the arid central Asia from higher Khangai Mountain to the Gobi

Desert, Badain Jaran Desert, Qilian Mountain, Qaidam Basin and eventually to the NE Tibet

(Fig. 1-1). According to climatic and topographic settings, this section covers a holistic array

of geomorphic units including glacial and periglacial landforms, pediments/bajadas, steppe

and desert gorges, aeolian landforms and floodplain valleys with meandering rivers. In

conjunction with high resolution sampling, well constrained chronological framework and

robust interpretations of physical, geochemical and geobiological proxies, all this set of

archives has the potential to serve as pivotal repository for paleoclimate (thermal and

moisture) signals, which in turn will benefit the climate modelling and archaeological studies.

The Gobi Desert in arid northern China and southern Mongolia was recognized as one of the

main dust sources for Chinese Loess Plateau and pelagic sediments in the western Pacific

(Pye and Zhou, 1989; Xiao et al., 1997; Lu and Sun, 2000; Vandenberghe et al., 2006).

Previously, broad pattern of late Quaternary (in particular Holocene) paleoclimate signatures

of arid central Asia were defined and reviewed by Pachur and Wünneman (1995), Yang et al.

(2004, 2011), Feng et al. (2005), Herzschuh (2006), Chen et al. (2008), An et al. (2008),

Wang et al. (2010) and Wang and Feng (2013). Ample compilations have suggested a distinct

moisture regime of middle and southern Mongolia in comparison to that of higher latitude

Baikal catchments and East Asian Summer Monsoon (EASM) domain of northern and eastern

China (Yang et al., 2004; An et al., 2008; Wang et al., 2010; Wang and Feng, 2013).

However, it remains elusive with respect to how and to what extent have EASM and

2

Westerlies affected the thermal and moisture signals in this spectacularly arid region. In

addition, substantially temporal and spatial heterogeneity exist in this arid realm, namely,

most of the records have poured attention into temporally the Holocene period and spatially,

the arid northern China and northwestern Mongolia.

Fig. 1-1: Atmospheric circulations over the central and eastern Asia, with panel A denoting

the summer pattern and panel B representing winter pattern. Solid lines indicate ca. 3,000 m

a.s.l. airflow while dashed lines indicate earth surface flow at ca. 600 m (adapted from Benn

and Owen, 1998; Lehmkuhl and Haselein, 2000; Barry and Chorley, 2009; An et al., 2014;

Loibl, 2015). Study areas of the Ejina Basin and Orog Nuur Basin are marked as white

rectangle in the panels. More detailed analyzed sections in this dissertation refer to following

Figs. 2-2 and 2-3.

In this dissertation, terrestrial archives including aeolian, fluvial/alluvial and lacustrine

sediments are selected and retrieved from the Ejina Basin and Orog Nuur (nuur = lake)

catchment, respectively (Fig. 1-1), in trying to decipher the late Quaternary

paleoenvironmental signals based on a robust multidisciplinary study. In addition, this

dissertation is assembled primarily from three manuscripts (i.e., Chapters 4, 5 and 6). Several

fractions in Chapter 1 may therefore overlap with paragraphs in those chapters.

1.1. Research framework and outline

This dissertation work is carried out in the framework of (i) German Federal Ministry of

Education and Research project (BMBF-CAME: 03G0814A): “Supra-regional signal

pathways and long-time archives: Quaternary monsoon dynamics at the northern margin of

the Tibetan Plateau”, (ii) German Research Foundation project (DFG: LE 730/16-1): “Late

Pleistocene, Holocene and ongoing geomorpho-dynamics in the Gobi Desert, South

3

Mongolia” and (iii) the financial support from the China Scholarship Council (CSC:

201306190112). The focus of this dissertation is using sedimentological and geochemical

signatures incorporated in diversified terrestrial archives to decipher late Quaternary

paleoclimate change and sedimentary processes in the Ejina Basin and Gobi Desert of

Mongolia. On the other hand, additional studies including geobiological proxy analysis (i.e.,

palynology and ostracod) and geomorphological investigations were also conducted

(Lehmkuhl and Lang, 2001; Hempelmann, 2010; Felauer, 2011; Hülle, 2011; Murad, 2011)

with the intention to gain a sound and compelling paleoenvironmental implications. To briefly

outline the structure of this dissertation, the main chapters are outlined as follows: 1,

introduction to the status of the paleoclimate research in the research area and motivations to

carry out this study; 2, geologic, geomorphological, climate, vegetation, paleoglacial and

anthropo-zoogenic settings of the study area; 3, field investigations, sampling, lab and

mathematic methodologies employed in the study; 4, geochemical characteristics of different

terrestrial archives (the Ejina Basin; first manuscript); 5, multidisciplinary study to infer the

paleoenvironmental change in the Orog Nuur catchment in the Gobi Desert (the Orog Nuur

catchment; second manuscript); 6, bulk-geochemistry of the lacustrine sediments and

potential elements and ratios to decipher integrated paleoenvironmental and provenance

signals (the Orog Nuur core; third manuscript); 7, a synthesis concerning the

paleoenvironmental reconstruction and proxy studies; brief abstract of the dissertation in

English and German language; references, and appendix are added at the end of this

dissertation.

1.2. Terrestrial archives in the Gobi Desert

To address those issues rose at the very beginning of the chapter, careful selection and full

understanding of the geological repository is a prerequisite and pivotal step.

1.2.1. Lacustrine sediments

In the context of the Gobi Desert, lacustrine sequence is a critical archive for climate and

provenance signatures. In particular in the arid central Asia, due to scarcity of abundant

continuous loess-paleosol or oceanic archives, lacustrine records were more often employed

as the Quaternary paleoenvironmental repository (Boyle, 2001; Mischke et al., 2005; Jin et

al., 2015). Southern Mongolia is located on the margin of the Westerlies dominated arid

central Asia and East Asian Summer Monsoon dominated costal eastern Asia (Chen et al.,

2008). Acquiring of a continuous and chronological reliable past environment record in this

4

region is of exceedingly importance concerning the refinement of the understanding of the

comprehensive climate pattern and possible corresponding driving mechanisms. However,

there exist hitherto numerous spatial and temporal heterogeneities concerning the archives. In

the Gobi Desert of Mongolia, only two continuous records were previously reported i.e.,

Bayan Tohomin Nuur (Felauer et al., 2012) and Ulaan Nuur (Lee et al., 2011; 2013).

Nonetheless, none of these two cores provide paleoenvironment records that trace back to

ages before ~15 ka. A record spanning longer time period is therefore indispensable to

address aforementioned issues.

1.2.2. Aeolian, fluvial and alluvial sediments

Apart from the lacustrine sequence, it is also indispensable to gain a better understanding

concerning other terrestrial archives embracing aeolian, fluvial and alluvial sediments to

figure out the robust paleoenvironmental implications. Previously, the main factors

controlling geochemical compositions of sediments are the nature of parent rocks, chemical

weathering, diagenesis, mechanical disaggregation and abrasion, authigenic / allothigenic

input, and metamorphism in case of relatively old e.g., early Cenozoic and Mesozoic

sediments (Bhatia, 1983; Johnsson, 1993; Fralick and Kronberg, 1997; Young et al., 1998;

McLennan, 2001; S. Yang et al., 2004; Xu et al., 2011). Analyzing geochemical

characteristics along with other characteristics of sediments is therefore a powerful tool to

decipher the nature of aforementioned processes. Previously, elemental (isotopic) and mineral

fingerprints have been employed to evaluate the provenance of the American loess /

mudstones / sandstones (Gallet et al., 1998; Zhang et al., 2014b), European loess (Amorosi et

al., 2002; Buggle et al., 2008; Újvári et al., 2014), Eastern / Central Asian loess / desert sands

(Liu et al., 1993; Honda and Shimizu, 1998; Jahn et al., 2001; Honda et al., 2004; Yang et al.,

2007; Maher et al., 2009; Stevens et al., 2010; Chen and Li, 2011; Xu et al., 2011; Xiao et al.,

2012; Wang et al., 2012) and Greenland dust (Újvári et al., 2015). Nesbitt and Young (1982)

reported the major element content to represent a quantitative chemical weathering proxy,

whereas in absence of strong chemical weathering (e.g., in arid and semi-arid environments,

Zhu et al., 2014), the bulk-composition and mineralogy can be explained by physical

abrasion, comminution and sorting (Nesbitt and Young, 1996; von Eynatten et al., 2012).

Elemental characteristics of lacustrine or aeolian deposits have been employed to reconstruct

the paleoenvironmental change in the northern Tibetan Plateau and the Mongolian Plateau

during the late Quaternary (e.g., Hartmann and Wünnemann, 2009; Schwanghart et al., 2009;

Lu et al., 2010; IJmker et al., 2012a; Felauer et al., 2012; Guo et al., 2013). Apart from

5

provenance and paleoenvironmental studies, the nature of bulk-compositions of different

sediment archives is hitherto incompletely understood. On the other hand, whether and how

geochemical composition can be assigned to sedimentary processes in arid region is seldom

studied. To further decipher the past processes and controlling factors, a better understanding

of the geochemical fingerprints recorded in these archives is needed.

1.3. State of art of the paleoclimate research in the Gobi Desert and adjacent regions

1.3.1. The Ejina Basin and arid northern China

In the arid northern China, i.e., where Ejina Basin locates, from west to east, either deserts or

dunefields embracing Badain Jaran Desert, Tengger Desert, Mu Us Desert, Horqin, Songnen

and Hunlunbuir Sandfield have dominated the region (Lu et al., 2013; Yu et al., 2013). In this

arid realm, as compiled from peat bog, loess-paleosol sequences, lacustrine sediments, a

sensitive alteration of the documented moisture and thermal signals in response to the

paleoclimate changes are concluded (Yu et al., 2013). In particular, compared with modern

times, the temperature has decreased by 5-11 °C and the precipitation decreased by 180-350

mm during the Last Glacial Maximum (LGM, ~26-~16 ka in global scale), whilst the

temperature has increased by 1-3 °C and the precipitation increased by 30-400 mm in the

Mid-Holocene Optimum (~9-~5 ka in global scale) (Yu et al., 2013). In the arid and semi-arid

regions of northern China, particularly, in basin systems, lacustrine / playa, aeolian and

alluvial / fluvial sediments are common. Spatial and temporal characteristics of these

sediments are correlated to littoral and pelagic deposition systems of paleolake expansion and

desiccation (Wünnemann, 1999), which can be ascribed to climatic variations. The endorheic

Ejina Basin is regarded as a fundamental dust provenance of the Chinese Loess Plateau and

adjacent regions (Pye and Zhou, 1989; Lu and Sun, 2000; Vandenberghe et al., 2006; Sun et

al., 2008). For the Ejina Basin and adjacent areas, the land relief, geomorphological

backgrounds, late Pleistocene deposition characteristics, neotectonic constrains and

hydrological changes were conducted by Wünnemann et al. (1999, 2007), Mischke et al.

(2003), Hartmann et al. (2009, 2011), Wang et al. (2011) and Zhu et al. (2014). Deciphering

the geochemical characteristics of archives in this basin system would have been one crucial

step towards a better understanding of the paleoenvironmental changes and sedimentary

processes that are related to dust activities (see Chapter 4).

6

1.3.2. The Orog Nuur catchment and Mongolian Plateau

Further to the north, more attentions have been dedicated to the northern and western

Mongolia as well as the Baikal catchments in Russia, albeit most of the lacustrine sequence

listed here spans no longer than the Holocene period: Baikal Lake (Horiuchi et al., 2000;

Karabanov et al., 2000; Prokopenko et al., 2001; Murakami et al., 2012; Kostrova et al.,

2014), Kotokel Lake (Shichi et al., 2009), Karakul Lake (Komatsu et al., 2015), Hovsgol

Nuur (Tarasov et al., 1999; Fedotov et al., 2004; Karabanov et al., 2004; Prokopenko et al.,

2002, 2007; Hovsgol Drilling Project Members, 2009; Murakami et al., 2010; Fumiko

Watanabe et al., 2014), Khuvsgul Nuur (Orkhonselange et al., 2013); Bayan Nuur (Dorofeyuk

and Tarasov, 1998; Nauman, 1999), Telmen Nuur (Peck et al., 2002; Fowell et al., 2003),

Khuisiin Nuur (Tian et al., 2013; 2014), Gun Nuur (Feng et al., 2005; Zhai, 2008; Zhang et

al., 2012), Ugii Nuur (Schwanghart et al., 2008, 2009; Wang et al., 2009), Uvs Nuur (Walther,

1999; Grunert et al., 2000), Hoton Nuur (Tarasov et al., 2000; Rudaya et al., 2009), Lake

Tuolekule (An et al., 2011); Tsetseg Nuur (Klinge and Lehmkuhl, 2013); Dada Nuur (Gunin,

1999), Yamant Nuur (Gunin, 1999), Dood Nuur (Dorofeyuk and Tarasov, 1998), Achit Nuur

(Gunin, 1999), Ulaan Nuur (Lee et al., 2011; 2013) and Bayan Tohomin Nuur (Felauer et al.,

2012). In addition, concerning other terrestrial archives, several loess paleosol sequences e.g.

Karakorum (Lehmkuhl et al., 2011), Shaamar (Feng et al., 2005) and Khyaranny (Feng et al.,

2001) were reported. The well dated interstadial Marine Isotope Stage 3 (MIS 3) paleoglacial

advance in Altai Mountain, Kanas River valley, Western Khangai, and Darhad Basin were

previously published by Reuther et al. (2006), Gillespie et al. (2008), Xu et al. (2009),

Lehmkuhl et al. (2011; 2015), Zhao et al. (2013), and Rother et al. (2014), providing

promising marks of the paleoclimate signals in this arid/semi-arid region. In general, the high

pedogenesis i.e., soil layers and large scale glacial expansions are intimately correlated to the

higher moisture supplies. However, especially in the central to southern Mongolia, all this

suite of studies has presented an urgent requirement to retrieve and reverse a reliable

paleoclimate sequence spanning longer time periods than the Holocene.

As aforementioned, on the Mongolian Plateau and adjacent arid central Asia, continuous

lacustrine sequences are widely regarded as fundamental repository for paleoenvironmental

signatures. A wealth of works has been reported concerning the late Quaternary moisture and

thermal history. However, an array of paramount aspects still needs to be tackled. They are

summarized as follows: (i) biased or even contradictory conclusions may occur due to the

interpretations of different proxies. For instance, Schwanghart et al. (2009) advocated a

7

warmer and wetter mid-Holocene in Ugii Nuur (in central-northern Mongolia) as inferred

from the mineral composition and bulk-geochemistry, while an arid mid-Holocene was

reconstructed from palynology studies (Wang and Feng, 2013). (ii) Most of the works poured

attention into the Holocene period, while only few records addressed the climate history

during the Pleistocene (Pachur and Wünnemann, 1995; Feng et al., 1998; Grunert et al., 2000;

Lehmkuhl and Haselein, 2000). (iii) Substantial spatial heterogeneity is also noteworthy.

Amount of studies on lacustrine deposits were carried out in the Baikal catchments in Russia,

northern and western Mongolia and northern China. Only two lacustrine sequences were

hitherto conducted in southern Mongolia, namely, Bayan Tohomin Nuur (Felauer et al., 2012)

and Ulaan Nuur (Lee et al., 2013) (cf., Fig. 5-1A). Furthermore, above problems might have

been reinforced by uncertainties in different radiocarbon age models due to a varying hard

water effect (Mischke et al., 2013).

On the other hand, among the suite of the multidisciplinary studies on argillaceous sediments,

geochemistry appears most likely the promising tool to decipher the interplay between the

environmental change, source lithotype and sediment bulk-composition (Boyle, 2001).

Considering the late Quaternary lacustrine sediments, the bulk-geochemistry may be

controlled by source terranes, authigenic or allothigenic input, which may be modulated by

the past environmental conditions, while pedogenesis and diagenesis might exert only limited

overprints (Yu et al., 2016). Knowledge of the major and minor element abundance downcore

variance along with the field investigation and carefully examined geologic mapping will

thereby enable us to gain a better understanding of the climate-induced environment and

provenance changes throughout the deposition process. Furthermore, surveys considering the

major and minor element abundance and corresponding environmental interpretations in the

pelagic realm have been systematically conducted and reviewed (Calvert and Pedersen, 2007),

while their counterpart elucidations in the lacustrine sediments still await more investigations

(cf., Roser and Korsch, 1988; Norman and De Deckker, 1990; Roy et al., 2012; Doberschütz

et al., 2013; Davies et al., 2015; Liang and Jiang, 2015).

1.4. Research Motivations

As reviewed above, an array of paramount aspects has inhibited our complete understanding

of the broad pattern of Quaternary moisture and thermal history of the arid central Asia and

their underlying mechanisms. Main knowledge gaps are summarized as follows:

8

(1) It remains elusive regarding how and to what extent have EASM and Westerlies affected

the thermal and moisture signals in this spectacularly arid region;

(2) Most of the geologic records have poured attention into the Holocene period, whereas

only few records can extend back to earlier marine isotope stages;

(3) Substantially spatial heterogeneity is noteworthy in the area. Exceeding amounts of

studies were carried out in the Lake Baikal catchment, northern and western Mongolia,

while only two continuous lacustrine sequences (spanning ~ 15 ka) were hitherto

conducted in southern Mongolia (Lehmkuhl and Lang, 2001; Felauer et al., 2012; Lee et

al., 2013);

(4) Concerning identical archive sequences, biased or even contradictory conclusions may

occur due to the interpretations of palynological and bulk-geochemical signals (e.g.,

Schwanghart et al., 2009; Wang and Feng, 2013), multidisciplinary study is therefore

indispensable to reconstruct sound paleoenvironmental signals;

(5) Apart from provenance and paleoenvironmental studies, the nature of bulk-compositions

of different sediment archives and different granulometric proportions is hitherto

incompletely understood. On the other hand, whether and how geochemical composition

can be assigned to sedimentary processes in arid region is seldom studied. To further

decipher the past processes and controlling factors, a better understanding of the

geochemical fingerprints incorporated in this set of terrestrial archives is needed;

(6) Surveys considering the bulk-geochemistry and corresponding environmental

interpretations in the marine realm have been systematically conducted and reviewed,

while their counterpart explanations in the lacustrine sediments still await more

investigations (cf., Boyle, 2001; Calvert and Pedersen, 2007).

Based on those knowledge gaps, the core fraction of the dissertation is outlined and

demarcated into three chapters, from which Chapter 4 is dedicated to the Ejina Basin while

Chapters 5 and 6 represent the multidisciplinary study on two lacstrine drilling cores ONW I

and ONW II recovered in the Orog Nuur catchment:

Manuscript 1: Discriminating sediment archives and sedimentary processes in the arid

endorheic Ejina Basin, NW China using a robust geochemical approach (Chapter 4):

In this study, a geochemical approach in conjunction with a broad field of granulometric,

mineralogical and generally constrained chronological data is presented in order to:

Identify the broad pattern of lithologic units at the northern margin of the Ejina Basin;

9

Differentiate the distinct nature of bulk-compositions in different granulometric

compositions and the array of archives obtained from these units;

Employ a multivariate statistic approach to test which assemblages of elements are

suitable, sensitive and robust to reconstruct sedimentary processes and thereby to evaluate

the controlling factors for the bulk-composition of sediments in (semi-)arid regions.

Hence, ultimately, the aim is to improve our understanding of the intriguing sedimentary

processes in this arid terrestrial endorheic context.

Manuscript 2: Lake sediments documented late Quaternary humid pulses in the Gobi Desert

of Mongolia: Vegetation, hydrologic and geomorphic implications (Chapter 5):

A multidisciplinary study is performed on a critically selected lacustrine sequence in

attempting to gain a robust and unambiguous knowledge of the past environmental changes.

In this study, two carefully selected parallel cores were recovered from Orog Nuur, in

southern Mongolia. Based on previous investigations, a multidisciplinary study involving

sedimentology, bulk-geochemistry, palynology, and ostracod analyses were carried out on

lacustrine cores ONW I and ONW II, intending to:

Gain a reliable knowledge of moisture history in the Orog Nuur catchment over the last

~50 ka, and its governing atmospheric circulation pattern;

Elucidate broad pattern of interrelated vegetation and hydrologic history of the Orog Nuur

catchment;

Generate and refine understandings on the spatial and temporal traits of moisture and

thermal signals in Gobi Desert of Mongolia, with aid of comparison with adjacent aeolian,

pelagic and mountainous glacial expansion records.

Manuscript 3: Geochemical imprints of coupled paleoenvironmental and provenance change

in the lacustrine sequence of Orog Nuur, Gobi Desert of Mongolia (Chapter 6):

In this study, high resolution major and trace element contents accompanied by calcium

carbonate concentration and C/Natomic ratios of the lacustrine sequence are examined in

attempts to:

Test the possible element ratios that can be employed to extract the broad pattern of the

diversified source terranes throughout the depositional period;

In terms of certain elements e.g., Co, Mn and Zr, examine the underlying driven

mechanism and environmental interpretation in the lacustrine realm;

10

In conjunction with briefly constrained mineral abundance, an array of major and trace

elements’ downcore variance were illustrated to indicate the interplay in the integrated

paleoenvironmental and provenance change in the depositional system.

11

“Mongolia is situated where contrasting geological

structures intersect, within the zone of interaction of

different systems of global atmospheric circulation.

Here, at the junction of the Siberian taiga forest,

Dahurian steppes, and Gobi desert is a crossroad of

plant and animal distribution” (P.D. Gunin et al, 1999)

2. Regional setting

The Gobi Desert covers a broad area including both northern China as well as the central to

southern Mongolia (Fig. 2-1). In this dissertation, two representative study areas comprising

the Ejina Basin in the south and the Orog Nuur catchment in the north are selected.

2.1. Geologic and geomorphic setting

2.1.1. The Ejina Basin

The Ejina Basin (synonym: Gaxun Nur Basin) lies in the lower reaches of the endorheic Hei

River catchment (with an area of ca. 130,000 km2, Fig. 2-2), which contains one of the

world’s largest continental alluvial fan systems (Hartmann et al., 2011). The basin covers an

area of about 28,000 km2

and is fed by the Hei River discharging from the Qilian Mountains.

The elevation ranges from 880 to 1,300 m above sea level (a.s.l.) (Hartmann et al., 2011) and

late Quaternary sediments are up to ~300 m thick (Zhang et al., 2006; Hartmann and

Wünnemann, 2009). The infilling of the basin started probably at earliest 250 ka ago, as

retrieved from the D100 drilling core (Fig. 2-2, 42.1°N 100.85°E, 940 m a.s.l., Wünnemann,

1999, Wünnemann et al., 2007a). Paleozoic to Mesozoic intrusive bedrock can be observed in

the northern Juyanze and Wentugaole basins (Fig. 2-2, Becken et al., 2007; Rudersdorf et al.,

2015). The basin is bound by the Heli Mountains in the south, the Bei Mountains to the west,

the Badain Jaran Desert / Alashan Plateau to the east and the Gobi Altai Mountains to the

north (Fig. 2-2). Temporal and spatial patterns of fluvial and lacustrine deposition are

influenced by neotectonics (e.g., the western margin has a subsidence rate of ca. 0.8-1.1 m/ka,

cf., Hartmann et al., 2011). Nowadays, only the Ejina oasis near Juyanze paleolake receives

ephemeral water input (Hartmann et al., 2011). Typical geomorphological features are gravel

plains, yardangs, playas, sand fields and sporadically distributed mobile linear and barchan

dunes (Zhu et al., 2014; Fig. 2-1).

12

Fig. 2-1: Photographs of representative contexts in the study area: (A) Gobi surface and

Precambrian intrusive bedrocks with background of Mountains in Mongolia. (B) Megadunes

on the northern margin of Badain Jaran Desert. (C) Common observed Haloxylon

ammodendron on the gobi surface. (D) Unparallelled sand ripples on the dune surface. (E)

Playa surface of western Juyanze Basin with particularly high contend of evaporite minerals.

(F) The ancient city Khara Khoto (synonyms: Black City, abandoned in 1,372 A.D.) and

nearby accumulated dunes.

13

Fig. 2-2: Geologic map and section locations in the northern Ejina Basin: Section HC8 is an

escarpment between the western and eastern Juyanze sub-basin. Sections KK1 and KK2 are

located by the braided river channel near ancient of Khara Khoto. Sections WT1 and WT2

are yardangs located on the margin of Wentugaole Sub-basin and Badain Jaran Desert.

Black (dash-) lines denote strike-slip neotectonics compiled from Wünnemann et al. (2007),

Hölz et al. (2007), Hartmann et al. (2011), Cunningham (2013) and Rudersdorf et al. (2015).

D100 correspond to the drilling core from Wünnemann (1999) and Wünnemann et al. (2007).

2.1.2. The Orog Nuur catchment

The Orog Nuur catchment lies in the Valley of Gobi Lakes in southern Mongolia (Fig. 2-3;

Tab. 2-1). The Orog Nuur is a brackish habitat that is subjected to desiccation and it

occasionally turns into a playa environment. Tuyn Gol (gol = river) is the main inflow that

originates from the southern slopes of Khangai Mountains around 220 km north of the lake.

The lake has no conspicuous outflow, resulting in a hydrologically closed basin system. The

northwest-southeast orientated Valley of Gobi Lakes (ca. 1,400 to 1,800 m above sea level,

a.s.l.) is an elongated intramontane depression between the Siberian Craton and the Tarim and

Sino-Korean Cratons (Baljinnyam et al., 1993; Windley and Allen, 1993; Cunningham et al.,

1996, 1997; Owen et al., 1999; Bardarch et al., 2000; Webb and Johnson, 2006; Cunningham,

14

2013; Guy et al., 2014). The Valley of Gobi Lakes is bounded by the Khangai Mountains in

the north and Gobi Altai Mountains in the south (Fig. 2-3B). The maximum elevation of the

Khangai Mountains is around 4,000 m a.s.l. at Otgon Tenger Uul in the west (4,008 m a.s.l.,

47°36’30”N, 97°33’09”E) and ca. 3,000 m a.s.l. at Erhet Hayrhan Uul in the east (3,037 m

a.s.l., 46°39’18”N, 101°13’56”E), whilst the Gobi Altai has a summit at 3,957 m a.s.l. (Ikh

Bogd Uul: 44°59’42”N, 100°13’51”E; uul = mountain). A left lateral strike-slip fault is

located at the southern side of the Khangai Mountains (Tapponnier et al., 1979; Walker et al.,

2007; Cunningham, 2013), while the southern part of the Orog Nuur watershed exhibits the

tectonic escarpment of the Gobi Altai and gently inclined alluvial fans (Baljinnyam et al.,

1993; Johnson, 2004; Vassallo et al., 2007, 2011). The elongated forebergs of the Gobi Altai

are markedly faulted, warped and partially uplifted by thrust faults (Owen et al., 1997;

Bayasgalan et al., 1999). For instance, in the Gobi Altai, due to the 1957 earthquake with a

magnitude of 8.3, left-lateral displacements reached 6 m and vertical displacements

approached 1.5 m (Baljinnyam et al., 1993), which may lead to misinterpretations regarding

the shoreline ridges or terraces. Paleozoic volcanic and plutonic rocks are exposed in the basin

and some places are overlain by conglomerates. Previously, a much larger lake extent in the

basin was reconstructed based on three sets of shorelines (Lehmkuhl and Lang, 2001;

Komatsu et al., 2001; Grunert et al., 2009). This suite of recessional beaches are determined

to be ~60 m, ~23 m and ~16 m, respectively, higher than the modern lake-level. East-west

oriented mobile dunes and cuspate bars lie in the vicinity of the Orog Nuur (Stolz et al.,

2012).

2.2. Climate regime

2.2.1. Climate of the Ejina Basin

The Ejina basin as part of the Gobi Desert is located beyond the limits of the East Asian

Summer Monsoon (cf., Chen et al., 2008). The winter Siberian Anticyclone dominates the

climate regime (Chen et al., 2008; Mölg et al., 2013). The mean January and July air

temperatures are ca. -11°C and 26 °C, respectively. More than 90% of the ca. 38 mm annual

precipitation occur between July and early September, albeit the annual precipitation in the

upper reaches of the Hei River catchment in the Qilian Mountains can reach 300-500 mm

(Wang and Cheng, 1999; Wünnemann et al., 2007a). The theoretical potential evaporation is

around 3,700 mm.

15

Fig. 2-3: Present-day distribution of geologic terranes in the Orog Nuur catchment (marked

with dashed line). Northern region of Orog Nuur Basin is dominated with Devonian-Triassic

felsic quartz syenite, granite and plagiogranite, accompanied by intermediate mafic fractions.

Southern region of Orog Nuur exhibit increasing fraction of mafic and ultramafic basalt and

andesite-basalt rocks. Alkaline syenite occurs sporadically in the lower Orog Nuur

catchments. The depositional basin is bounded by eastern Khangai in the north and the Gobi

Altai Mountains in the south. More details concerning representative minerals see Tab. 2-1.

Concerning the geomorphological and topographic settings refer to Fig. 5-2 of the same

region.

2.2.2. Climate of the Orog Nuur catchment

Further to the north, Mongolia is dominated by a continental climate characterized by dry and

cold winter and a precipitation maximum in summer (An et al., 2008). The mean annual

precipitation along a north-south transect ranges from ~500 mm to <100 mm. 65-75 % of the

precipitation occurs in summer due to the enhanced south-eastern cyclonic activity along the

polar front (Kostrova et al., 2014). Only ~100 mm precipitation is delivered in the lower Gobi

Desert of southern Mongolia, while the higher elevated Khangai Mountains can reach ~400

mm. According to records at Bayankhongor weather station near the Orog Nuur, the mean

16

July and January temperature in the Gobi Desert are ~16 °C and ~-20 °C, respectively, while

the mean annual temperature is about -2 °C to 5 °C (Felauer, 2012). In spring, dust- or

sandstorms frequently occur with maximum wind speeds exceeding 20 m/s (Hempelmann,

2010; Felauer, 2012). There is generally no snow cover because of the immense winter

dryness related to the predominant Siberian Anticyclone (Kostrova et al., 2014).

Tab. 2-1: Inventory of the plutonic rocks, volcanic / subvolcanic rocks and Quaternary

deposits in the Orog Nuur catchments. For geologic mapping refer to Fig. 1-3. This set of

sources rocks is of vital importance concerning the bulk-geochemistry discussed in Chapter 6.

Plutonic rocks and complexes Volcanic and subvolcanic rocks Genetic types of

Quaternary deposits

Ultramafic Mafic a alluvial

σ dunite, peridotite, serpentinite melange β basalt l lake

Mafic μβ diabase, dolerite al alluvial-lake

συ ultrabasite-basitic layered series τβ trachybasalty g glacier

η anorthosite Intermediate d talus

ση gabbro-troctolite-anorthositic complex αβ andesitebasalt p proluvial

υ gabbro, gabbronorite ταβ trachyandesite basalt dp proluvial-talus

υδ gabbrodiorite α andesite v aeolian

Intermediate τα trachyandesite

δ diorite Felsic

qδ quartz diorite λ dacite, rhyolite, quartz porphyry

Felsic μλ rhyodacite, trachrhyodacite

δ-γ diorite-granitic complex τλ trachydacite, trachyrhyolite

γδ granodiorite, tonalite, plagiogranite Bimodal

γ granite, leucogranite βλ spilite-keratophyric and

γξ granosyenite, quartz syenite, subalkaline granite basalt-rhyolitic complexes

γ-γξ granite-granosyenitic

γπ granite-porphyry

Alkaline

ευ alkaline gabbro, ijolite, urtite

εξ nepheline syenite, alkaline syenite

εγ alkaline granite

2.3. Soils and vegetation cover

Primary soil type in central to southern Mongolia is dry-steppe Kastanozems. Other common

soil types are Chernozems/dark Kastanozem in the forest-steppe zone as well as brown

semidesert soils (Tamura et al., 2012). In modern context, arable soil is relatively scarce in

both the Ejina Basin and Orog Nuur catchment. The modern vegetation follows a north-south

transect from alpine tundra, mountain taiga, mountain/forest steppe, grass steppe, desert

steppe (Orog Nuur catchment) and finally to desert communities such as the Ejina Basin (Fig.

1-1; Hilbig, 1995; Gunin et al., 1999). In Khangai and Altai Mountains, the dominant tree

species is Larix sibirica (accompanied by patches of Betula and Populus) on the northern and

eastern slopes, while steppe prevail the sun-exposed southern and western slopes (Miehe et

al., 2007). The uppermost boundary of the alpine meadows is ~ 2,900 m a.s.l. (Lehmkuhl and

Lang, 2001). Most of the lowland regions in the basin system are characterized by open desert

17

steppes and bare gobi surface. Dwarf bushes and reed-halophytes are the dominant

communities in lower reaches of the Orog Nuur catchment (Murad, 2011), while, as outlined

in Fig. 5-1A, the higher elevated catchment of the Orog Nuur is covered by forest steppe and /

or alpine steppe.

2.4. Paleo- and modern glacial setting

As illustrated in Fig. 2-4, there is no modern glacier or periglacial activity in the Ejina Basin,

albeit glaciers exist in the further southern higher elevated Qilian Mountains. On the other

hand, as the depositional basin of Orog Nuur is linked to the higher Khangai Mountains via

Tuyn Gol, Quaternary glacial variance may play a pivotal role in terms of water supply and

the clastic delivery. In modern times, only the western Khangai has small ice caps and the

average equilibrium line altitude (ELA) was estimated in a range from 3,900 to 4,000 m a.s.l.

(Fig. 2-4; Rother et al., 2014). The modern periglacial zone reaches down to 2,700 m a.s.l.

(Lehmkuhl and Lang, 2001; Klinge et al., 2003). A set of Last Glacial Maximum (LGM)

moraines at an altitude of ~2900 m. a.s.l. in the southern Khangai Mountains was dated

thereafter to >21 ka (Lehmkuhl and Lang, 2001). In addition, Ritz et al. (2006) determined

ages of alluvial sediments of the forelands of Gurvan Bogd (bogd = massif) by 10

Be

cosmogenic nuclides exposure dating and supposed an aggradation of sediments in the

context of the global glacial terminations. In the Gobi Altai, permafrost developed at ca. 22-

15 ka and degraded since ~13 ka, and was thereafter diminished since the early Holocene

(Owen et al., 1998).

2.5. Anthropo-zoogenic setting

In Neolithic period, first signs of nomadic presence on the NE Tibetan Plateau were reported

to be as early as 7,200 cal a BP, when temperature was up to 2 °C warmer than today (Schlütz

and Lehmkuhl, 2009). The descendant nomadic migration routes were established at least at

~2,200 years BP (Schlütz and Lehmkuhl, 2009). The “Nomadic Anthropocene” was

postulated by Schlütz and Lehmkuhl (2009) and Miehe et al. (2014) for the last 6,000 years

when nomads transformed the natural steppe-like vegetation into K. pygmaea pasture.

Landforms become more heterogeneous due to fragmentation of the rangelands (Yu et al.,

2017). Furthermore, in this region, agriculture facilitated permanent settlements were

suggested since 3,600 a BP (Chen et al., 2015). In modern times, human activities play an

important role in regulating the desert-steppe biomes in northern China and central to

southern Mongolia (Lehmkuhl, 1997). However, in our context, the time spanning of ~50 ka

18

to the early Holocene has ruled out intensively anthropo-zoogenic impacts, the reversed

thermal and moisture data indicate therefore exclusively paleoenvironmental signals.

Fig. 2-4: Schematic north-south oriented topographic transect and corresponding vegetation

patterns. The eastern parts of Khangai Mountains (higher reach of the Orog Nuur catchment)

with altitudes between 3,500 m and 3,700 m a.s.l. have no modern glaciers, while small ice-

fields may reach down to 3,200 m a.s.l. in the western Khangai Mountains (Lehmkuhl and

Lang, 2001).

19

“Paleoclimatic research provides the essential

understanding of climate system variability, and its

relationship to both forcing mechanisms and feedbacks,

which may amplify or reduce the direct consequences of

particular forcings…the proxy material has acted as a

filter, transforming climatic conditions at a point in time,

or over a period, into a more or less permanent record,

but the record is complex and incorporates other signals

that may be irrelevant to the paleoclimatologist”

(Raymond S. Bradley, 1999)

3. Materials and methods

3.1. Field work and sampling strategy

In the Ejina Basin, three field investigations were carried out in 2011, 2012 and 2014.

Selected profile locations are distributed along three geomorphological systems (Figs. 2-2),

i.e., following inverted channels (sections KK1 and KK2) to the lower Juyanze playa (section

HC8) and then toward the higher Wentugaole / Badain Jaran Desert margin (sections WT1

and WT2, see cross-section Fig. 4-1B). Ninety-three samples from five sections with different

geomorphological contexts were collected for further sedimentological and geochemical

analysis. Samples were acquired at 10-50 cm intervals based on stratum thickness and

properties. Fewer samples were acquired within the generally homogeneous lithological units.

Concerning the Orog Nuur, two parallel cores (ONW I: 6.00 m, 45°03’48”N, 100°34’39”E;

ONW II: 13.35 m, 45°04’28”N, 100°35’08”E, 1,219 m a.s.l.) are retrieved from the western

part of Orog Nuur during the playa phase in 2007 (Fig. 2-3). The core was opened, described

and documented in the field. Samples were separated and transported to University of

Göttingen, Free University of Berlin and the Alfred Wegener Institute in Bremerhaven for

palynological, ostracod, X-ray fluorescence and X-ray powder diffraction analyses. The

remaining material of the cores is preserved at RWTH Aachen University for grain size and

TOC analysis. The paleoenvironmental reconstruction is carried out primarily based on the

longer core ONW II.

20

3.2. Laboratory methods

3.2.1. Chronological methods

For the Ejina sequences, in order to acquire and assess a generally constrained chronological

framework which can be compared with previously established lacustrine or beach bar

records, AMS radiocarbon dating was applied to six samples in the Poznan Radiocarbon

Laboratory, Queen’s University Belfast and the Beta Analytic Radiocarbon Dating

Laboratory. Dated materials were bulk organic matter and carbonates. Radiocarbon ages were

calibrated to calendar years using the CALIB rev.7.1 (Stuiver et al., 1998) with the latest

calibration dataset (Intcal13.14c, Reimer et al., 2013). Radiocarbon ages are presented as

calibrated years (cal BP) unless otherwise noted. Optically Stimulated Luminescence (OSL)

dating was applied to six samples at Nanjing University. The samples were treated with 30 %

hydrogen peroxide and 10 % hydrochloric acid. The fine sand-sized (63-90 µm) quartz grains

were extracted by wet sieving. 2.58 g/cm3 sodium polytungstate was used to separate the

quartz and K-feldspar. Treatments with 40 % hydrogen fluoride for 40 min and 10 %

hydrochloric acid for 20 min were applied to purify the quartz. After washing and drying, the

grains were mounted as monolayer on the 9.8 mm diameter stainless steel discs using silicon

spray. All luminescence measurements were carried out using a Risø TL/OSL-DA-20 C/D

reader equipped with blue LEDs (470±30 nm) and an IR laser diode emitting at 830 nm

(Bøtter-Jensen et al., 2010). Irradiations were carried out with 90

Sr/90

Y beta sources mounted

on the readers and calibrated for discs. The quartz OSL signals were collected through 7.5

mm of Schott U-340 (UV) glass filter. Equivalent dose (De) was determined by single-aliquot

regenerative does (SAR) protocol following Murray and Wintle (2003). For samples with

feldspar contamination after a maximum of three etching treatments with hydrofluoric acid,

infrared IR-stimulation is inferred prior to the optical (blue-light) stimulation to both measure

and deplete the feldspar signal prior to the OSL measurement. A preheat of 260 °C for 10 s

and cut heat of 220 °C combination was used for routine equivalent dose measurements. The

238U,

232Th and

40K concentrations were detected using Neutron Activation Analysis (NAA) at

the China Institute of Atomic Energy. As the in-situ water content is extremely low, 25 % of

saturation water content was used in dose rate calculation and the absolute uncertainty on the

water content was assumed to be ±5%. Using the revised dose rate conversion factors of

Guérin et al. (2011) and water content attenuation factors (Aitken, 1998), the elemental

concentrations were converted into effective dose rate. The cosmic ray dose rate was

calculated following Prescott and Hutton (1994).

21

Concerning Orog Nuur drilling cores (ONW I; ONW II), ten radiocarbon ages were

determined from bulk materials (Tab. 5-1) using accelerator mass spectrometry (AMS) at the

University of Erlangen-Nürnberg and subsequently calibrated to calendar years using CALIB

rev. 7.1 (Stuiver et al., 1998) with the latest IntCal 13.14c calibration curve (Reimer et al.,

2013). All radiocarbon ages are presented as calibrated years (cal a BP). Apart from the

unrecovered uppermost sediments, a polylinearly regressed age-depth model of core ONW II

(25-1,335 cm) was computed. Reservoir effects at different depth are assessed by (i)

calculating the surface age from the age model, (ii) identifying Termination I in bulk-

geochemical and palynological sequences, and (iii) comparing the Younger Dryas (YD,

Dansgaard et al., 1989)-related sand layer in Orog Nuur and Ulaan Nuur (Fig. 5-1A, ~200 km

northeast of Orog Nuur, Lee et al., 2011, 2013). Two radiocarbon ages from core ONW I (at

386 cm and 588 cm, respectively) are employed to compare timing of the stratigraphic units

relative to ages of counterpart depths in core ONW II (Tab. 5-1 and Fig. 5-4).

3.2.2. Grain size analysis

The grain size distributions (GSDs) of all samples from the Ejina Basin and Orog Nuur Basin

were measured using a Beckman Coulter LS13320 Laser Diffraction Particle Size Analyzer

with a detection range of 0.04-2,000 µm. Portions larger than 2,000 µm were previously

removed by sieving. As previously proposed, the GSD, in particular the finer proportions (i.e.,

< 2 µm) in sand and loess sediments, may be potentially altered by different pretreatment

methods (e.g., Lu et al., 2001; Kovács, 2008; Mason et al., 2011). Therefore, before

measuring, a parallel test (HCl/H2O2- and H2O2-only-treatment) was done to test the impact of

possible incorporated calcium carbonate on the resulting GSD, revealing however no

pronounced differences (cf. Schulte et al., 2016). Thus, all measurements were performed

uniformly without hydrochloric acid treatment. 700 µl of 35 % H2O2 was added to remove the

organic components. Afterwards samples were dried at 70 °C for two days. Sodium

pyrophosphate was added to avoid coagulation. Optical properties (fluid RI = 1.33, sample RI

= 1.55, imaginary RI = 0.1) were used as described in Özer et al. (2010) for quartz dominated

samples. The final GSD represents the average of two subsamples. Each of them was

measured twice to ensure the reliability of the results. In case of conspicuous offset existing

between subsamples, the aforementioned procedure was repeated for the sample.

22

3.2.3. X-ray fluorescence analysis and X-ray diffraction

For those samples from the Ejina Basin, X-ray fluorescence (XRF) analysis was conducted at

RWTH Aachen University on all ninety-three samples using an Ametek X-ray Fluorescence

Spectrometer (Spectro Xepos, 2007). The whole sample material was dried, sieved through a

2,000 µm mesh sieve and then ground to < 63 µm using a Pulverisette 7 before mixing 8 g

sample with 2 g wax. The mixed powder is pressed into steel rings for measuring with a

pressure of 15 tons for 20 s. Following calculation procedure, root mean square errors of

calibration and lower limit of detection can be inferred from Spectro Xepos (2007). Certified

reference materials involving No. 5358-90, Nos 2504-2506-83, Nos 2498-2500-83, NCS DC

73375, JF-1, UG-QLO-1 and UG-SDC-1 were referred to during the analyzing. Thirty

elements are selected for further analysis. However, several values of the samples were below

detection limit (BDL). X-ray diffraction (XRD) was applied on fifty-three samples from

section HC8 to determine mineral abundances (Hartmann and Wünnemann, 2009). The XRD

was measured at Freie Universität Berlin using a Philips PW1710 and graphite

monochromator with a Cu-K-alpha at 36 kv, 24 mA. The powdered samples were scanned

from 3° to 40° at °2θ, angle spectrums are conducted by Xpert Highscore software, the

classification of impulses rates following Hartmann and Wünnemann (2009). Examples of

dispersive spectrums of XRD are illustrated in Fig. 4-4.

Concerning the Orog Nuur core sequences, the X-ray fluorescence (XRF) analysis was carried

out in Alfred Wegner Institute (AWI) for Polar and Marine Research in Bremerhaven,

Germany, using an Avaatech XRF Core Scanner configured with Canberra X-Pips 1500-1.5

detector (slit 10 mm, 10, 30 keV, 300 and 700 µA). The scanning interval is 1 cm, resulting in

1,335 measurements for the core ONW II. Processes comprising subtracting a background

curve, applying statistical corrections for physical processes in the detector and deconvolution

of overlapping peaks were performed using scanner equipped WinAxil package (NIOZ and

Avaatech, 2007). To gain a direct knowledge about the depositional environment of former

sediment-water interface, two factors were computed based on the element dataset using

factor analysis given in Yu et al. (2016). Thereby, high loading elements for each factor in

conjunction with the factor scores provide intriguing information about the sedimentary

processes (Yu et al., 2016). In addition, X-ray diffraction (XRD) for 14 samples at different

depths was performed to assess the general mineral composition and provide additional

information for the interpretation of bulk elemental compositions. The sediment color (RGB

model) was scanned using a CCD line scan camera.

23

3.2.4. Electrical conductivity and magnetic susceptibility measurements

For the core ONW I from the Orog Nuur, electrical conductivity (EC, here equals specific

conductivity) and magnetic susceptibility (MS) were carried out as an environmental and

salinity proxies. 50 ml water and 0.01 g KCl were added for each sample, the EC with 10 cm

interval was measured after 30 min using a WTW LF 196 probe. MS at 1 cm interval was

measured using a Barrington MS2 with a MS2 E sensor.

3.2.5. Carbon, nitrogen, sulfur, TOC and CaCO3 analyses

For the core ONW II from the Orog Nuur, Carbon (TC), nitrogen (TN), sulfur (TS)

concentrations, and the Total Organic Carbon (TOC) content were determined, in order to

gain a better understanding of the geochemical habitats of the former water sediment

interface. Samples were dried at 36 °C for 24 hours. The samples in zinc capsules were

subsequently analyzed in CHNS Analyzer EA3000. The Scheibler gas method (DIN 19684)

was employed to calculate the CaCO3 content, and 1-5 g samples were utilized respectively.

Calculation equation is as follows:

𝐶𝑎𝐶𝑂3 (%) = 𝑉 ∗ 𝑝 ∗ 0.1602/(𝑡 + 273) ∗ 𝐸 ∗ 𝐾

Where V is the volume of CO2 (cm3), p represents the atmospheric pressure (mm Hg), t

denotes room temperature (°C), E is the loss of weight (g) and K exhibits the correction factor

(here 2 is used). TOC was subsequently computed following Scheffer and Schachtschabel

(2002):

𝑇𝑂𝐶 = (𝐶 − (𝐶𝑎𝐶𝑂3 ∗ 0.12)) ∗ 1.72

The C/Natomic ratios throughout the ONW II core were computed and examined for further

interpretations (Meyers and Teranes, 2001).

3.2.6. Pollen, non-pollen palynomorphs and palynofacies analyses

Concerning the core ONW II, palynological studies were processed at University of

Göttingen. Sediment subsamples of 1 or 2 cm3 were acquired at 5-10 cm intervals through the

core ONW II. In total 123 samples were extracted and treated with standard laboratory

procedures including KOH, HCl, HF treatments and subsequent acetolysis (Erdtman, 1960;

Moore et al., 1991). Lycopodium spores were added as marker before calculation of the

24

pollen, spores and charcoal particle concentration (grains/cm3). Except for few samples, at

least 400 grains were counted for statistics. A total of 50 pollen and 40 spore taxa were

identified through the core ONW II. Clump of pollen that denotes immature grains sticking

together was calculated as paleoecological signals (Schlütz and Lehmkuhl, 2007). Available

references (Moore et al., 1991; Vanky, 1994; Van Geel et al., 2003) were employed to

identify the taxonomic nomenclature of pollen and spores. All taxa were subsequently

aggregated into several groups, namely, arboreal plants, humidity indicators, aquatic algae,

aridity indicators and grazing indicators. Two ratios were employed to compute the annual

temperature (T) and moisture (M) following Ma et al. (2008) (flora groups and equations see

Tab. 5-2). The moisture and thermal signals are further compared with the A/C ratios as well

as the pollen concentration, respectively.

3.2.7. Ostracod analysis

A total of 131 sub-samples from core ONW II were collected at 10 cm intervals for ostracod

analysis in order to examine and verify the paleohydrological conditions. The sediment was

treated with 3 % H2O2 for 48 hours and washed through 0.25 and 0.1 mm meshes. All

ostracod valves were picked from the dried sieve residues and identified according to Meisch

(2000) and Mischke and Zhang (2011). The species-specific ecological inferences are mainly

based on Hiller (1972), Hammer (1986), Anadon et al. (1994), Meisch (2000) and Mischke et

al. (2005).

3.3. Mathematical methods: Correlations and multivariate statistics

Data mining is an indispensable and vital step in trying to better understand intriguing

infomation in the bulk-geochemistry of the different sediment archives. Here, correlation

analysis and multivariate statistics (factor analysis) are computed. Multivariate statistics are

employed in both Chapters 4 and 5, while correlations analysis was only performed in

Chapter 4.

To provide an aid to illustrate the elemental characteristics of each archive type (see

conceptual model in Fig. 4-5), Spearman’s rank correlation was performed between GSD

proportions and XRF dataset (as a data set is not normally distributed, Spearman’s rank is

preferred to be conducted, see Spearman, 1904). In order to simplify interpretation of the

geochemical data (Fig. 4-5), a robust Factor Analysis (RFA) (Filzmoser et al., 2009; Zhang et

al., 2014a) was conducted on the whole multivariate dataset of ninety-three samples with

25

sixteen elements. Elements showing virtually all concentrations below the respective

detection limit or depicting low loadings in the conventional factor analysis were excluded.

Rarely occurred values lying below the detection limit are regarded as missing values (or

N/A) and will hinder the FA analysis (no zero value is allowed). They are therefore

substituted with ½ the lower detection limit (cf., IJmker et al., 2012b). Since the geochemical

dataset is sum-constrained and however not normally distributed, isometric log-ratio

transformation (Egozcue et al., 2003, and literature therein; Filzmoser et al., 2009) has been

applied, and the axis is rotated to the maximum direction of variance (i.e., varimax rotated

axis; Kaiser, 1958). Subsequently, the number of factors for interpretations needs to be

determined. Parallel analysis (Franklin et al., 1995) was performed and shows the first four

factors are higher than the parallel confidence. Thus, they are regarded as confident factors.

To further evaluate the overall quality of the analysis, communalities are computed by

summing up the squared loadings for each element (> 0.9 represent good representation for an

element).

Hereby, most of the original data characteristics are summarized in fewer decorrelated factors.

Cosine of the angle between the elements and the attribute- and factor-axis denote the loading

of this element on the factor (e.g., 45° represent the loading of 0.7), thus, the factor-loading-

matrix (eigenvector matrix) can be regarded as a correlation matrix for partial correlation of

input variables to factor ensemble (eigenspace bases). Sedimentary interpretations for each

factor can be inferred from corresponding higher loadings of specific elements. Factor scores

of each sample were obtained for detailed discussions of sedimentary processes (Fig. 4-5).

26

“The stratigraphic record of sedimentary rocks is the

fundamental database for understanding the evolution of

life, plate tectonics through time and global climate

change” (Gary Nichols, 2009)

4. Discriminating sediment archives and sedimentary processes in the arid endorheic

Ejina Basin, NW China using a robust geochemical approach (Published in 2016, Journal

of Asian Earth Science 119, 128-144)

Kaifeng Yu 1*

, Kai Hartmann 2, Veit Nottebaum

1, Georg Stauch

1, Huayu Lu

3, Christian

Zeeden 1, Shuangwen Yi

3, Bernd Wünnemann

2,3, Frank Lehmkuhl

1

1 Department of Geography, RWTH Aachen University, Templergraben 55, 52056 Aachen,

Germany

2 Department of Earth Sciences, Freie Universität Berlin, Malteserstraße 74-100, 12249

Berlin, Germany

3 School of Geographic and Oceanographic Sciences, Nanjing University, Xianlin Avenue

163, 210023 Nanjing, China

Geochemical characteristics have been intensively used to assign sediment properties to

paleoclimate and provenance. Nonetheless, in particular concerning the arid context, bulk

geochemistry of different sediment archives and corresponding process interpretations are

hitherto elusive. The Ejina Basin, with its suite of different sediment archives, is known as

one of the main sources for the loess accumulation on the Chinese Loess Plateau. In order to

understand mechanisms along this supra-regional sediment cascade, it is crucial to decipher

the archive characteristics and formation processes. To address these issues, five profiles in

different geomorphological contexts were selected. Analyses of X-ray fluorescence and

diffraction, grain size, optically stimulated luminescence and radiocarbon dating were

performed. Robust factor analysis was applied to reduce the attribute space to the process

space of sedimentation history. Five sediment archives from three lithologic units exhibit

geochemical characteristics as follows: (i) aeolian sands have high contents of Zr and Hf,

whereas only Hf can be regarded as a valuable indicator to discriminate the coarse sand

proportion; (ii) sandy loess has high Ca and Sr contents which both exhibit broad correlations

with the medium to coarse silt proportions; (iii) lacustrine clays have high contents of felsic,

ferromagnesian and mica source elements e.g., K, Fe, Ti, V, and Ni; (iv) fluvial sands have

27

high contents of Mg, Cl and Na which may be enriched in evaporite minerals; (v) alluvial

gravels have high contents of Cr which may originate from nearby Cr-rich bedrock. Temporal

variations can be illustrated by four robust factors: weathering intensity, silicate-bearing

mineral abundance, saline / alkaline magnitude and quasi-constant aeolian input. In summary,

the bulk-composition of the late Quaternary sediments in this arid context is governed by the

nature of the source terrain, weak chemical weathering, authigenic minerals, aeolian sand

input, whereas pedogenesis and diagenesis exert only limited influences. Hence, this study

demonstrates a practical geochemical strategy supplemented by grain size and mineralogical

data, to discriminate sediment archives and thereafter enhance our ability to offer more

intriguing information about the sedimentary processes in the arid central Asia.

4.1. Introduction

The main factors controlling geochemical compositions of sediments are the nature of parent

rocks, chemical weathering, diagenesis, mechanical disaggregation and abrasion, authigenic /

allothigenic input, and metamorphism in case of relatively old e.g., early Cenozoic and

Mesozoic sediments (Bhatia, 1983; Johnsson, 1993; Fralick and Kronberg, 1997; Young et

al., 1998; McLennan, 2001; S. Yang et al., 2004; Xu et al., 2011). Analyzing geochemical

characteristics along with other characteristics of sediments is therefore a powerful tool to

better decipher the nature of aforementioned processes. Previously, elemental (isotopic) and

mineral fingerprints have been employed to evaluate the provenance of the American loess /

mudstones / sandstones (Gallet et al., 1998; Zhang et al., 2014b), European loess (Amorosi et

al., 2002; Buggle et al., 2008; Újvári et al., 2014), Eastern / Central Asian loess / desert sands

(Liu et al., 1993; Honda and Shimizu, 1998; Jahn et al., 2001; Honda et al., 2004; Yang et al.,

2007; Maher et al., 2009; Stevens et al., 2010; Chen and Li, 2011; Xu et al., 2011; Xiao et al.,

2012; Wang et al., 2012) and Greenland dust (Újvári et al., 2015). Nesbitt and Young (1982)

reported the major element content to represent a quantitative chemical weathering proxy,

whereas in absence of strong chemical weathering (e.g., in arid and semi-arid environments,

Zhu et al., 2014), the bulk-composition and mineralogy can be explained by physical

abrasion, comminution and sorting (Nesbitt and Young, 1996; von Eynatten et al., 2012).

Elemental characteristics of lacustrine or aeolian deposits have been employed to reconstruct

the paleoenvironmental change in the northern Tibetan Plateau and the Mongolian Plateau

during the late Quaternary (e.g., Hartmann and Wünnemann, 2009; Schwanghart et al., 2009;

Lu et al., 2010; IJmker et al., 2012a; Felauer et al., 2012; Guo et al., 2013). Apart from

provenance and paleoenvironmental studies, the nature of bulk-compositions of different

28

sediment archives is hitherto incompletely understood. On the other hand, whether and how

geochemical composition can be assigned to sedimentary processes in arid region is seldom

studied. To further disentangle the past processes and controlling factors, a better

understanding of the geochemical fingerprints recorded in these archives is needed.

In arid and semi-arid areas, particularly, in basin systems, lacustrine, aeolian and alluvial /

fluvial sediments are common. Spatial and temporal characteristics of these sediments are

correlated to littoral and pelagic deposition systems of paleolake expansion and desiccation

(e.g., Wünnemann, 1999; Shanahan et al., 2013), which can be ascribed to climatic variations.

The endorheic Ejina Basin is regarded as a fundamental dust provenance of the Chinese Loess

Plateau and adjacent regions (Pye and Zhou, 1989; Lu and Sun, 2000; Vandenberghe et al.,

2006; Sun et al., 2008). Deciphering the geochemical characteristics of archives in the basin

system is one crucial step towards a better understanding of the sedimentary processes related

to dust activity.

In this study, a geochemical approach in conjunction with a broad field of granulometric,

mineralogical and generally constrained chronological data is presented in order to:

(1) Identify the broad pattern of lithologic units at the northern margin of the Ejina Basin.

(2) Differentiate the distinct nature of bulk-compositions in different granulometric

compositions and the array of archives obtained from these units.

(3) Employ a multivariate statistic approach to test which assemblages of elements are

suitable, sensitive and robust to reconstruct sedimentary processes and thereby to evaluate

the controlling factors for the bulk-composition of sediments in (semi-)arid regions.

Hence, ultimately, the aim is to improve our understanding of the intriguing sedimentary

processes in this arid terrestrial endorheic context.

29

Fig. 4-1: Simplified geologic map and section locations in the northern Ejina Basin: (A)

Section HC8 is an escarpment between the western and eastern Juyanze sub-basin. Sections

KK1 and KK2 are located by the braided river channel near ancient of Khara Khoto. Sections

WT1 and WT2 are yardangs located on the margin of Wentugaole Sub-basin and Badain

Jaran Desert. Black (dash-) lines denote strike-slip neotectonics compiled from Wünnemann

et al. (2007), Hölz et al. (2007), Hartmann et al. (2011), Cunningham (2013) and Rudersdorf

et al. (2015). Black bold lines denote schematic elevation cross-section in panel B. D100

correspond to the drilling core from Wünnemann, 1999 and Wünnemann et al., 2007. (B)

Cross-section from Khara Khoto to Juyanze (playa) and Wentugaole (sub-basin).

30

4.2. Regional Setting

The Ejina Basin (synonym: Gaxun Nur Basin) is located in the lower reaches of the endorheic

Hei River catchment (with an area of ca. 130,000 km2, Fig. 4-1), which contains one of the

world’s largest continental alluvial fan systems (Hartmann et al., 2011). The basin covers an

area of about 28,000 km2

and is fed by the Hei River discharging from the Qilian Mountains.

The elevation ranges from 880 to 1,300 m above sea level (a.s.l.) (Hartmann et al., 2011) and

late Quaternary sediments are up to ~300 m thick (Zhang et al., 2006; Hartmann and

Wünnemann, 2009). The infilling of the basin started probably at earliest 250 ka ago, as

retrieved from the D100 drilling core (Fig. 4-1, 42.1°N 100.85°E, 940 m a.s.l., Wünnemann,

1999, Wünnemann et al., 2007a). Paleozoic to Mesozoic intrusive bedrock can be observed in

the northern Juyanze and Wentugaole basins (Fig. 4-1, Becken et al., 2007; Rudersdorf et al.,

2015).

The basin as part of the Gobi Desert is located beyond the limits of the East Asian Summer

Monsoon (cf., Chen et al., 2008). The winter Siberian Anticyclone dominates the climate

regime (Chen et al., 2008; Mölg et al., 2013). The mean January and July air temperatures are

ca. -11°C and 26 °C, respectively. More than 90% of the ca. 38 mm annual precipitation occur

between July and early September, albeit the annual precipitation in the upper reaches of the

Hei River catchment in the Qilian Mountains can reach 300-500 mm (Wang and Cheng, 1999;

Wünnemann et al., 2007a). The theoretical potential evaporation is around 3,700 mm.

The basin is bound by the Heli Mountains in the south, the Bei Mountains to the west, the

Badain Jaran Desert / Alashan Plateau to the east and the Gobi Altai Mountains to the north

(Fig. 4-1). Temporal and spatial patterns of fluvial and lacustrine deposition are influenced by

neotectonics (e.g., the western margin has a subsidence rate of ca. 0.8-1.1 m/ka, cf., Hartmann

et al., 2011). Nowadays, only the Ejina oasis near Juyanze paleolake receives ephemeral water

input (Hartmann et al., 2011). Typical geomorphological features are gravel plains, yardangs,

playas, sand fields and sporadically distributed mobile linear and barchan dunes (Zhu et al.,

2014, Fig. 4-2).

31

Fig. 4-2: Photographs of representative geomorphological contexts from the northern Ejina

Basin: (A) Inverted channels near Khara Khoto. (B) Gobi surface and Precambrian intrusive

bedrocks in Wentugaole sub-basin. (C) Khara Khoto and nearby accumulated dunes. (D)

Yardangs in southern Wentugaole sub-basin. (E) Playa surface of western Juyanze Basin. (F)

Megadunes on the northern margin of Badain Jaran Desert.

4.3. Methods

4.3.1. Sampling strategy

Three field investigations were carried out in 2011, 2012 and 2014. Selected profile locations

are distributed along three geomorphological systems (Figs. 4-1B and 4-3), i.e., following

inverted channels (sections KK1 and KK2) to the lower Juyanze playa (section HC8) and then

toward the higher Wentugaole / Badain Jaran Desert margin (sections WT1 and WT2, see

32

cross-section Fig. 4-1B). Ninety-three samples from five sections with different

geomorphological contexts were collected for further sedimentological and geochemical

analysis. Samples were acquired at 10-50 cm intervals based on stratum thickness and

properties. Fewer samples were acquired within the generally homogeneous lithological units.

Fig. 4-3: Lithostratigraphy and generally constrained chronology framework. Chronology

framework is labeled, and 2-sigma ranges for calculated radiocarbon ages are reported in

Table 1. U1 is the lowermost aeolian sands (except for WT1), U2 is lacustrine stratum, and

U3 is fluvial sands (mixed with gravels) unit (e.g., Fig. 2A). U3 of HC8 is especially thick

mixture of poorly-sorted coarse alluvial sands and gravels.

4.3.2. Chronological methods

In order to acquire and assess a generally constrained chronological framework which can be

compared with previously established lacustrine or beach bar records, AMS radiocarbon

dating was applied to six samples in the Poznan Radiocarbon Laboratory, Queen’s University

Belfast and the Beta Analytic Radiocarbon Dating Laboratory. Dated materials were bulk

33

organic matter and carbonates. Radiocarbon ages were calibrated to calendar years using the

CALIB rev.7.1 (Stuiver et al., 1998) with the latest calibration dataset (Intcal13.14c, Reimer

et al., 2013). Radiocarbon ages are presented as calibrated years (cal BP) unless otherwise

noted.

Optically Stimulated Luminescence (OSL) dating was applied to six samples at Nanjing

University. The samples were treated with 30 % hydrogen peroxide and 10 % hydrochloric

acid. The fine sand-sized (63-90 µm) quartz grains were extracted by wet sieving. 2.58 g/cm3

sodium polytungstate was used to separate the quartz and K-feldspar. Treatments with 40 %

hydrogen fluoride for 40 min and 10 % hydrochloric acid for 20 min were applied to purify

the quartz. After washing and drying, the grains were mounted as monolayer on the 9.8 mm

diameter stainless steel discs using silicon spray. All luminescence measurements were

carried out using a Risø TL/OSL-DA-20 C/D reader equipped with blue LEDs (470±30 nm)

and an IR laser diode emitting at 830 nm (Bøtter-Jensen et al., 2010). Irradiations were

carried out with 90

Sr/90

Y beta sources mounted on the readers and calibrated for discs. The

quartz OSL signals were collected through 7.5 mm of Schott U-340 (UV) glass filter.

Equivalent dose (De) was determined by single-aliquot regenerative does (SAR) protocol

following Murray and Wintle (2003). For samples with feldspar contamination after a

maximum of three etching treatments with hydrofluoric acid, infrared IR-stimulation is

inferred prior to the optical (blue-light) stimulation to both measure and deplete the feldspar

signal prior to the OSL measurement. A preheat of 260 °C for 10 s and cut heat of 220 °C

combination was used for routine equivalent dose measurements. The 238

U, 232

Th and 40

K

concentrations were detected using Neutron Activation Analysis (NAA) at the China Institute

of Atomic Energy. As the in-situ water content is extremely low, 25 % of saturation water

content was used in dose rate calculation and the absolute uncertainty on the water content

was assumed to be ±5%. Using the revised dose rate conversion factors of Guérin et al. (2011)

and water content attenuation factors (Aitken, 1998), the elemental concentrations were

converted into effective dose rate. The cosmic ray dose rate was calculated following Prescott

and Hutton (1994).


The grain size distributions (GSDs) of all samples were measured using a Beckman Coulter

LS13320 Laser Diffraction Particle Size Analyzer with a detection range of 0.04-2,000 µm.

Portions larger than 2,000 µm were previously removed by sieving. As previously proposed,

the GSD, in particular the finer proportions (i.e., < 2 µm) in sand and loess sediments, may be

34

potentially altered by different pretreatment methods (e.g., Lu et al., 2001; Kovács, 2008;

Mason et al., 2011). Therefore, before measuring, a parallel test (HCl/H2O2- and H2O2-only-

treatment) was done to test the impact of possible incorporated calcium carbonate on the

resulting GSD, revealing however no pronounced differences (cf. Schulte et al., 2016). Thus,

all measurements were performed uniformly without hydrochloric acid treatment. 700 µl of

35 % H2O2 was added to remove the organic components. Afterwards samples were dried at

70 °C for two days. Sodium pyrophosphate was added to avoid coagulation. The Mie Theory,

instead of previously used Fraunhofer approximation (cf., Özer et al., 2010), is applied for

GSD calculations. Optical properties (fluid RI = 1.33, sample RI = 1.55, imaginary RI = 0.1)

were used as described in Özer et al. (2010) for quartz dominated samples. The final GSD

represents the average of two subsamples. Each of them was measured twice to ensure the

reliability of the results. In case of conspicuous offset existing between subsamples, the

aforementioned procedure was repeated for the sample.

35

Fig. 4-4: Example of wave-length dispersive spectrums of X-ray diffraction (XRD). Panels A

to E denote aeolian sands, lacustrine clay, fluvial sands, sandy loess, alluvial gravels,

respectively from section HC8 (with depth). The powdered samples were scanned from 3° to

40° at °2θ, angle spectrums are conducted by Xpert Highscore software, the classification of

36

impulses following Hartmann & Wünnemann (2009). The number of counts is illustrated as

reference for e.g., aeolian sands.

4.3.4. X-ray fluorescence analysis and X-ray diffraction

X-ray fluorescence (XRF) analysis was conducted at RWTH Aachen University on all ninety-

three samples using an Ametek X-ray Fluorescence Spectrometer (Spectro Xepos, 2007). The

whole sample material was dried, sieved through a 2,000 µm mesh sieve and then ground to <

63 µm using a Pulverisette 7 before mixing 8 g sample with 2 g wax. The mixed powder is

pressed into steel rings for measuring with a pressure of 15 tons for 20 s. Following

calculation procedure, root mean square errors of calibration and lower limit of detection can

be inferred from Spectro Xepos (2007). Certified reference materials involving No. 5358-90,

Nos 2504-2506-83, Nos 2498-2500-83, NCS DC 73375, JF-1, UG-QLO-1 and UG-SDC-1

were referred to during the analyzing. Thirty elements are selected for further analysis.

However, several values of the samples were below detection limit (BDL).

Fig. 4-5: Conceptual modal of the robust geochemical approach to discriminate archives and

sequence sedimentary processes.

X-ray diffraction (XRD) was applied on fifty-three samples from section HC8 to determine

mineral abundances (Hartmann and Wünnemann, 2009). The XRD was measured at Freie

Universität Berlin using a Philips PW1710 and graphite monochromator with a Cu-K-alpha at

36 kv, 24 mA. The powdered samples were scanned from 3° to 40° at °2θ, angle spectrums

are conducted by Xpert Highscore software, the classification of impulses rates following

37

Hartmann and Wünnemann (2009). Examples of dispersive spectrums of XRD are illustrated

in Fig. 4-4.

4.3.5. Correlations and multivariate statistics

To provide an aid to illustrate the elemental characteristics of each archive type (see

conceptual model in Fig. 4-5), Spearman’s rank correlation was performed between GSD

proportions and XRF dataset (as a data set is not normally distributed, Spearman’s rank is

preferred to be conducted, see Spearman, 1904).

In order to simplify interpretation of the geochemical data (Fig. 4-5), a robust Factor Analysis

(RFA) (Filzmoser et al., 2009; Zhang et al., 2014a) was conducted on the whole multivariate

dataset of ninety-three samples with sixteen elements. Elements showing virtually all

concentrations below the respective detection limit or depicting low loadings in the

conventional factor analysis were excluded. Rarely occurred values lying below the detection

limit are regarded as missing values (or N/A) and will hinder the FA analysis (no zero value is

allowed). They are therefore substituted with ½ the lower detection limit (cf., IJmker et al.,

2012b). Since the geochemical dataset is sum-constrained and however not normally

distributed, isometric log-ratio transformation (Egozcue et al., 2003, and literature therein;

Filzmoser et al., 2009) has been applied, and the axis is rotated to the maximum direction of

variance (i.e., varimax rotated axis; Kaiser, 1958). Subsequently, the number of factors for

interpretations needs to be determined. Parallel analysis (Franklin et al., 1995) was performed

and shows the first four factors are higher than the parallel confidence. Thus, they are

regarded as confident factors. To further evaluate the overall quality of the analysis,

communalities are computed by summing up the squared loadings for each element (> 0.9

represent good representation for an element).

Hereby, most of the original data characteristics are summarized in fewer decorrelated factors.

Cosine of the angle between the elements and the attribute- and factor-axis denote the loading

of this element on the factor (e.g., 45° represent the loading of 0.7), thus, the factor-loading-

matrix (eigenvector matrix) can be regarded as a correlation matrix for partial correlation of

input variables to factor ensemble (eigenspace bases). Sedimentary interpretations for each

factor can be inferred from corresponding higher loadings of specific elements. Factor scores

of each sample were obtained for detailed discussions of sedimentary processes (Fig. 4-5).

38

4.4. Results

4.4.1. Lithology and chronology

Radiocarbon (Tab. 4-1) and OSL ages (Tab. 4-2) provide a chronological framework for the

studied sections. In conjunction with field investigations and visual observations, three

lithologic units were recognized (Figs. 4-3 and 4-5): (i) U1, the lowermost facies of the

aeolian sands (except WT1) are dated to older than 200 ka BP according to OSL ages. (ii) U2,

the clay to silty-clay stratum are dated to start at around the early global LGM ca. 26 and ca.

32 ka BP (for stratigraphy and error ranges see Fig. 4-3 and Tab. 4-1, respectively). (iii) U3,

the uppermost poorly-sorted mixture of coarse sands and angular gravel strata are dated to

start at around ca. 14 to 25 ka BP. The purported age difference resulting from the

chronological methods falls out of the scope of this study (Zhang et al., 2006; Long and Shen,

2015). In addition, two sand samples from yardangs (corresponding to U1) in Wentugaole

were obtained to test the feldspar dating performance. They yielded saturated equivalent doses

as well. The section with the probably most complete sediment stratum, HC8 (Figs. 4-1 and 4-

3), will be discussed as the key profile to assess the capability to reconstruct the sedimentary

processes. Brief section descriptions are provided subsequently.

Tab. 4-1: Results of the AMS 14

C-dating. Radiocarbon ages were calibrated to calendar years

with 2 σ standard deviation using the CALIB rev.7.1 (Stuiver et al., 1998) with the latest

calibration dataset (Intcal13.14c, Reimer et al., 2013).

Location Section

name

Lab.-No. Depth/cm Dating

material

Δ13C TOC/mg Age /ka BP 2 σ/cal ka BP

(95.4 %)

Khara Khoto KK2 Poz-53749 50 Bulk material - 0.15 11.82±0.13 13.73±0.29

Khara Khoto KK2 Poz-53750 120 Bulk material - 0.19 20.36±0.22 24.56±0.59

Khara Khoto KK1 UBA-19486 U. surface Bulk material -23.6 - 23.87±0.14 28.09±0.17

Wentugaole WT1 Poz-53751 470 Bulk material - 0.70 28.41±0.20 32.30±0.67

Wentugaole WT1 Poz-53752 350 Bulk material - 0.20 21.26±0.21 25.42±0.27

Juyanze HC8* Beta-190785 502 Bulk organic carbon

-22.8 - 15.06±0.80 18.29±0.24

* HC8 radiocarbon age was published in Hartmann & Wünnemann, 2009.

HC8 (41.9023°N, 101.7578°E, 934 m a.s.l.) represents a 13 m escarpment. Below 9 m, the

section exhibits loose fine sand with increasing silt contents up to 8 m. 8-2.5 m (base of U2)

show slightly indurated silty clay material. Sandy loess-like material was observed in depths

of ca. 3.5-3.0 m. 3.0-0 m is represented by a poorly-sorted mixture of coarse sands and

gravels.

KK1 (41.7494°N, 101.1582°E, 952 m a.s.l.) is located in an outcropping inverted channel

near the ancient city of Khara Khoto / Black City (Fig. 4-1). Coarse sand prevails below 3.1

39

m, overlain by a cemented mixture of coarse sand and pebbles (3.1-2.5 m). Layers of fine to

medium sand were preserved from 2.5 to 1.7 m, sand ripple structures and root channels were

sporadically exhibited. Deposits from 1.7 to 0.1 m (U2) were weakly indurated clayey to silty-

clayey sediments. The surface (0.1-0 m, U3) is covered by set of gravels with largest

diameters of ca. 70 mm and mixed with coarse sand.

KK2 (41.7563°N, 101.1255°E, 953 m a.s.l.) is in the vicinity of section KK1 (see Fig. 4-1).

U1 is represented by heterogeneous laminae of sand layers (3.2-1.25 m). U2 (1.25-0.2 m) is a

slightly reddish clayey to silty-clayey stratum. The surface, representing U3 (0.3-0 m),

consists of slightly indurated sand.

WT1 (41.3559°N 102.2468°E, 967 m a.s.l.) reflects a ca. 6 m yardang feature located on the

northern margin of the Badain Jaran Desert (Wentugaole Basin). Indurated clayey to silty-

clayey sediments were recovered below 5.4 m. 5.4-4.9 m is represented by diagonally cross-

bedded silt to sand with interbedding gypsum crusts. It is comparably thin in regard of the

other sections’ U1 unit. U2 (4.9-0.2 m) is clayey sediment while the surface (U3, 0.2-0 m) is

composed of strongly cemented coarse sand.

WT2 (41.3593°N 102.2456°E, 960 m a.s.l.) is a yardang record adjacent to WT1. Between

7.75 and 7.70 m horizontally layered sand was recovered, which is interrupted by a 5 cm thick

whitish-grayish laminated clay layer (7.75-7.70 m). The base of U2 is represented by a 5 cm

thick laminated clay layer (7.05-7.00 m), while up to 0.10 m the deposit is homogeneous clay.

The overlying U3 is an indurated white to greenish sandstone.

Tab. 4-2: Summary of water content, equivalent dose value (De), dose rate, number of used

aliquots and luminescence ages of samples from study sections. 238

U, 232

Th and 40

K

concentrations were determined by Neutron Activation Analysis (NAA). Absolute uncertainty

of the water content was assumed to be < ±5%. Due to saturation of the equivalent dose, brief

constrained luminescence ages are presented here. More details concerning the method see

Chapter 4.3.2.

Locati

on

Sectio

n

Lab.-No. Dept

h/cm

Water

/%

238U/pp

m

232Th/p

pm

40K/% De/Gy Dose rate/Gy

ka-1

Aliquots

number

Age/ka

Khara

Khoto

KK2 NJU1365 145 6 0.99±0.05

3.97±0.05

1.43±0.05

˃310 2.06±0.21 8 ˃150

Khara

Khoto

KK2 NJU1364 307 6 0.93±0.

05

3.70±0.

05

1.32±0.

05

˃400 1.90±0.20 3 ˃200

Wentu

gaole

WT1 NJU1366 500 4 0.76±0.

05

2.73±0.

05

1.31±0.

05

˃280 1.80±0.20 5 ˃158

Wentu

gaole

WT2 NJU1367 765 7 0.81±0.05

3.03±0.05

1.44±0.05

- - 6 Saturated

40

4.4.2. Granulometric composition and pre-assessment of archives

Grains size distributions (GSDs) are divided into six fractions as described in Blott and Pye

(2012): Clay (0.04-2 µm), fine silt (2-6.3 µm), medium silt (6.3-20 µm), coarse silt (20-63

µm), fine sand (63-200 µm), medium sand (200-630 µm) and coarse sand (630-2,000 µm).

Granulometric composition is illustrated in a ternary diagram (Fig. 4-6A).

Fig. 4-6: Ternary diagrams of granulometric properties and A-CN-K (molar proportions):

(A) Granulometric ternary diagram. (B) A-CN-K ternary diagram, CaO is corrected to

silicate carbonate following the procedure of McLennan (1993). The Chemical Index of

Alteration (CIA, Nesbitt & Young, 1982) is related on the right side of the A-CN-K triangle.

Abbreviations: A=Al2O3, C=CaO, N=Na2O, K=K2O, Ka=Kaolinite, Gi=Gibbsite,

Ch=Chlorite, Il=illite, Mu=muscovite, Pl=plagioclase, Bi=Biotite, Ks=alkali feldspar,

Ch=chlorite, Cp=clinopyroxene, Hb=hornblende.

For section HC8 (cf., Fig. 4-10): (i) most samples in lower U1 are characterized by unimodal

GSD curves and bimodal GSDs in the upper part. Clay proportion increases from 1.4 % to

18.9 % accompanied by an increasing silt proportion from 1.5 % to 24.5 %. Several coarse

sand input intervals can be detected in these layers. Clay and silt fractions share quite similar

depth patterns. Moreover, the medium sand proportion has the highest percentage (ca. 80 %)

at the bottom (ca. 13 m), whereas the upper part shows broadly less than 30 % with low

variance. (ii) U2 mostly consists of clay interbedded with coarse sand input. The clay

proportion varies between 15-40 %, silt proportion varies between 30-70 % and sand

proportion remains between 15-40 %. Generally, silt and sand proportions are increasing from

basal U2 to the upper part. (iii) The surface samples from U3 are predominantly unsorted silt

41

to sands mixed with angular gravels. In contrast, the other sections exhibit thinner U3 strata.

Regarding the other sections, broadly resembling GSD patterns throughout the profile can be

observed (Fig. 4-3; detail GSDs not illustrated).

Fig. 4-7: Granulometric properties of archives: (A) aeolian sands (n=51), (B) lacustrine clay

(n=19), (C) fluvial sands (n=18), (D) sandy loess (n=2), (E) alluvial gravels (n=2), different

axis scales are used.

42

Fig. 4-8: Spearman’s rank correlation coefficients between geochemical properties and grain

size parameters (all scales of red to blue units have passed 0.05 significant level). Not all

elements exhibit grain-size dependence (e.g., As, Nd, S and Al). Red pixels denote positively

correlated and blue ones depict negative correlations. These features can also shed light on

elemental characteristics of archives. C, fSi, mSi, cSi, fS, mS and cS denote the particle size of

clay, fine silt, medium silt, coarse silt, fine sand, medium sand and coarse sand, respectively.

According to the GSD patterns and the field investigation, all samples are differentiated into

five archive categories: aeolian sands, lacustrine clays, fluvial sands, sandy loess and alluvial

gravels (Fig. 4-5). The clusters of granulometric patterns are acquired as depicted in Fig. 4-7.

Combined with the criterion reported by Lu et al. (2001), Blott and Pye (2012), Vandenberghe

(2013) and Nottebaum et al. (2014), differentiation strategies are as follows: (i) aeolian sand

usually has a unimodal or sometimes bimodal distribution with a mode at around 50-250 µm

or even coarser (Fig. 4-7A); (ii) lacustrine clay usually shows bimodal or trimodal grain size

distribution including well sorted suspension proportions (8-9 µm) as the main mode (Fig. 4-

7B); (iii) fluvial sand’s GSDs are often polymodal, including a dominant well sorted saltation

portion (usually 250-400 µm) as the main mode and poorly sorted suspension proportions

(<63 µm) (Fig. 4-7C); (iv) sandy loess has a unimodal leptokurtic distribution and has a mode

43

around 60 µm (cf., end member 1 in Vriend and Prins, 2005; Prins et al., 2007; Nottebaum et

al., 2014, 2015b) and a minor population at around 2-15 µm (Fig. 4-7D); (v) alluvial gravels

have a unimodal distribution and a mode much larger than 1,000 µm (Fig. 4-7E).

4.4.3. Elemental and mineralogical composition of archives

In total thirty major and trace elements in ninety-three samples were acquired (Tab. 4-3). The

correlation coefficients between grain size proportions and element concentrations are

illustrated in Fig. 4-8.

Tab. 4-3: Results of the oxides of ten major elements expressed as normalized wt % and other

trace elements concentrations (ppm) of sediment archives as determined by X-ray

fluorescence analysis. Decimal precision of the reported concentrations were revised

according to the formats of their counterparts in the UCC data set (Rudnick and Gao, 2003).

Root mean square errors (RMS) and lower limit of detection (LOD) of each element see

Spectro Xepos (2007).

Element UCC* Factors Aeolian

sands

Lacustrine clay Fluvial sands Sandy loess Alluvial

gravels

SiO2 66.6 RF2 71.72 55.08 47.30 69.25 36.30

Al2O3 15.4 - 9.69 16.48 13.18 13.73 6.37

Fe2O3 5.04 RF1 2.96 7.53 5.58 4.95 2.18

MnO 0.1 RF2(-) 0.07 0.16 0.20 0.06 0.17

MgO 2.48 RF2(-) 4.50 7.07 11.63 3.72 7.96

CaO 3.59 - 6.11 6.77 15.40 2.00 42.64

Na2O 3.27 RF3 2.12 2.51 3.23 2.47 2.56

K2O 2.8 RF1 2.28 3.43 2.69 2.88 1.37

TiO2 0.64 RF1 0.39 0.81 0.63 0.74 0.28

P2O5 0.15 - 0.16 0.16 0.16 0.20 0.20

TOTAL - - 100 100 100 100 100

Rb 82 RF1 57 118 85 35 85

Y 21 - 11 23 20 17 17

V 97 - 67 142 102 28 116

Nb 12 - 4.1 10.1 7.1 2.3 9.4

Ni 47 - 23 60 42 9 39

Cu 28 RF1 12 39 28 8 27

Zn 67 RF1 26 92 67 23 54

Ga 17.5 RF1 8.0 19.4 14.2 5.8 13.3

Pb 17 - 11.6 29.1 22.7 10.5 18.3

As 4.8 - 5.6 19.4 10.9 4.6 11.7

Th 10.5 RF1 5.2 15.9 11.2 5.1 10.6

Sr 320 - 138 165 184 338 140

Cl 294 RF3 4052 6685 8513 1309 4615

S 621 - 785 2155 2194 935 2634

Cr 92 - 266 124 122 99 399

Zr 193 RF4 344 156 143 175 269

Hf 5.3 RF4 6.6 4.2 3.1 4.4 3.7

Nd 27 - 52 99 91 63 72

Ba 628 - 580 515 439 545 467

W 1.9 - 2.8 2.8 3.1 2.0 3.7

* UCC=Upper Crust Composition (Rudnick and Gao, 2003). Total iron expressed as Fe2O3.Ten major elements are expressed as oxides (wt

%), while other trace elements are expressed as ppm. The description of factors analysis refers to Tab. 4-4.

44

Fig. 4-9: Loading and score plot of geochemical data set: (A) Loading plot of conventional

factor analysis, with thirty elements. (B) Score plot of conventional factor analysis, with thirty

elements. (C) Loading plot of robust factor analysis, with sixteen elements. (D) Score plot of

the robust factor analysis, with sixteen elements. Originally, the maximum direction of

variance will be regarded as axes, however, rotation is performed to get varimax rotated axis

(i.e., X-and Y-axis here) during the factor analysis to provide a higher loading for each

elements (Tab. 4-4). Cosine of the angle between the elements arrow and the component

(varimax rotated axes) are factor loadings, length of the arrows are corresponding to the

element communalities. Compared with conventional factor analysis (Panels A and B), panels

C and D exhibit higher cumulative proportions for RFs 1 and 2. Dashed line circles in panels

A and C denote communality estimates (Tab. 4-4) and thereby assess the quality for each

45

element (> 0.9 can be regarded as good quality) in this factor modal. Panels B and D

illustrate that multivariate statistic of elemental fingerprints appears to be a practical method

to discriminate the archive types, of which mistakes may be yielded from field investigations

and GSD observations alone.

For samples from section HC8, minerals comprising quartz, feldspar, amphibole, clay

minerals, calcite, aragonite, dolomite, gypsum and halite abundances were determined.

Generally, the U1 stratum has higher contents of quartz and feldspar, the U2 stratum has

pronounced higher contents of clay minerals, gypsum and halite, and the U3 stratum has

especially high content of calcite. They therefore shed light on the mineralogical compositions

of archives, which will also aid the further interpretation of the XRF dataset.

4.4.4. Element assemblages and loadings as derived from conventional and robust factor

analysis

To gain an overview of the distribution of all thirty elements, conventional factor analysis was

done for all samples (Fig. 4-9A and 9B). Major and trace elements can be categorized into

several assemblages as follows: (i) cluster of elements such as K, Rb, Ti, V, Nb, Fe, Ni, Cu,

Zn, Al, Ga, As, Y, Nd, Pb and Th behave relatively similar with respect to convergence /

variation against depth and the directions generally point to factor 1 positively. In section

HC8, they have lower abundances in the basal U1 unit. Interestingly, there is one extremely

high value for all these elements at a depth between 943 and 959 cm. All these elements show

particularly low values in U3, accompanied by lower values in U1 and relatively higher

values in U2. (ii) Concerning Na, Cl and S, the former two elements have resembled variation

trends against depth. Three high value peaks within U2 can even reach 70,000 ppm for Na

and 55,000 ppm for Cl. S has a similar convergence zone as Na and Cl. Na and Cl show

slightly positive indications to both factor 1 and factor 2 (Fig. 4-9A). (iii) Si, Mg, Sr, Ca and

Mn show a similar distribution in Fig. 4-9A. Ca and Sr have relatively high values within U3,

whereas Mg exhibits extremely low contents in U3 (figures are omitted here). Mn is relatively

rare and commonly originates from highly soluble minerals (e.g., mobile Mn2+

) and occurs in

ultramafic rocks (e.g., dust transport of solid Mn3+

and Mn4+

, IJmker et al., 2012a), it exhibits

however higher concentrations in mixed fluvial sands (Tab. 4-3), the former process is thus

more likely the origin of Mn. In contrast, no dominant trends can be extracted from the

irregular variance of Si. All these elements show indications generally positive to factor 2 and

may be connected to fluvial archives (Fig. 4-9A and 9B). (iv) For Cr, Hf, Zr and Ba,

compared to U2, Hf and Zr values show remarkably higher contents in aeolian sands (U1). Cr

46

shows much more sensitive variations to the aeolian input, whenever aeolian sand input

occurs. The Cr values are approximately three times higher than the neighboring sample

values. Meanwhile, Ba values show irregular variances and no trend can be figured out. All

Cr, Hf, Zr and Ba values show negatively correlations to factor 2 (Fig. 4-9A). (v) P and W

show slightly positive loadings to both factors 1 and 2, no dominant trend can be extracted

from their irregular variance. Thus, they were excluded from further discussions.

Tab. 4-4: Standard deviations, variance / cumulative proportions, element loadings for each

factor, and communalities for each element. High positive or negative (>0.75) loadings are

shown in boldface, proportions in %.

Elements RF1, Weathering

intensity

RF2, Silicate-bearing mineral abundance

RF3, Saline

magnitude

RF4,

Aeolian

input

Communality

estimates

Na 0.12 - 0.96 - 0.94

Mg - -0.83 0.17 -0.12 0.73

Si - 0.95 - 0.22 0.95

K 0.91 0.34 - -0.13 0.97

Ca -0.17 -0.63 - -0.11 0.44

Ti 0.95 - 0.13 - 0.92

Mn 0.26 -0.79 - -0.14 0.72

Fe 0.99 - - -0.11 0.99

Cl - - 0.99 - 0.98

Cu 0.96 -0.17 - -0.13 0.98

Zn 0.97 -0.20 - -0.13 0.99

Ga 0.99 - - -0.12 0.99

Rb 0.98 - - -0.15 0.98

Zr -0.33 0.44 - 0.81 0.95

Hf -0.23 0.52 - 0.75 0.89

Th 0.98 -0.13 - -0.11 0.98

Standard

deviations

7.72 3.28 1.97 1.43 -

Variance proportion/%

0.48 0.21 0.12 0.09 -

Cumulative proportion/%

0.48 0.69 0.81 0.90 -

Blank represents those loadings which are under significant level in robust factor analysis. Communalities for each element are computed by

taking sum of the squared loadings, it can be referred to the length of the element arrows in loadings plot.

As a comparison to the aforementioned conventional factor analysis, RFA was performed to

reduce the systematic and random error of compositional data (according to Filzmoser, 2009;

Fig. 4-9C and 9D). Subsequently, the suitable and sensitive elements for further assessment of

sedimentological interpretations are compared. Element loadings (Tab. 4-4) and

communalities (cf., Figs. 4-9A and 9C) are thereby significantly higher compared with the

conventional method. Factors exhibiting eigenvalues >1 are extracted (e.g., five factors can be

determined by applying the Kaiser-Guttmann Criterion, Guttman, 1954). These factors have

variance higher than an input variable. Finally, the first four factors which are above the

parallel line are determined as confident factors (more details see Chapter 4.3.5; Franklin et

al., 1995). They represent ca. 90.00 % of the total geochemical variance of the XRF data set.

47

Relatively high (> 0.75, Tab. 4-4) positive or negative loadings of elements for each factor are

extracted. High loading elements for robust factor 1 (RF1) are Ti, Fe, K, Cu, Zn, Ga, Rb, Th.

RF2 has high loadings of Si, Mg, Mn. RF3 is represented by Na and Cl. RF4 yields high

loadings of Zr and Hf. All elements only display one relatively high loading, which implies

that they have only specific interpretation for one specific factor.

4.5. Discussion

4.5.1. Late Quaternary lithostratigraphy at the northern margin of the Ejina Basin

Similar late Quaternary lithostratigraphic facies are shared by five sections. Considering HC8,

aeolian phases dominated the sedimentary processes in U1 (beyond the upper limits of OSL

ages), which was sporadically interrupted by ephemeral riverine sediments.

Accompanied by several high salinity periods, the lacustrine sedimentation (U2) started

before 30 ka BP. The reported highest lake level of Sogo Nur occurred at around 41-33 ka BP

(i.e., MIS 3a, Lehmkuhl and Haselein, 2000; B. Yang et al., 2004 and literatures therein). The

highest shoreline of the Sogo Nur is radiocarbon dated to 33 ka BP (Wang et al., 2011). The

highest remnant beach bars near the Juyanze Basin was also radiocarbon dated to 37-29 ka BP

(Wünnemann et al., 2007; cf., Long and Shen, 2015). Molluscs from the highest shoreline of

eastern Juyanze Lake to the northwestern Badain Jaran Desert were dated to 30.40±0.87 ka

BP, indicating a huge expanded paleolake (Mischke et al., 2003). Additionally, δ18

O

reconstructions yielded high lake levels from 37-34, at 31 and 28-26 ka BP, respectively in

the Ejina Basin (Wünnemann and Hartmann, 2002). However, a far more eastern “Jilantai-

Hetao Megalake” existed between 50-60 ka BP (Chen et al., 2008a), the inception of the

lacustrine period occurred earlier compared with the stratum U2 in HC8. Furthermore, calcite

layers occur at depth of ca. 3.5 m within U2 (not captured in XRD data, Fig. 4-10). Cemented

calcareous layers in present time can be referred to an annual precipitation of 100-300 mm

(Hövermann, 1998), which indicates a much more humid period recoded by U2 compared to

today. In addition, the Sogo Nur lake basin experienced dune accumulation between 19-14 ka

BP (Lehmkuhl and Haselein, 2000), which is not observed here in section HC8. The

uppermost lacustrine sediments of U2 (Tab. 4-1; Fig. 4-3) yield calibrated 14

C ages of

13.661±0.227 and 28.757±0.500 ka BP, respectively.

48

Above U2, a mixture of fluvial / alluvial sand and gravels form a deflated gravel gobi surface.

Local ephemeral runoff occurred in U1 and the mixed fluvial sands and gravels may be

subsequently modulated by intensive deflation (cf., Nottebaum et al., 2015a).

4.5.2. Elemental characteristics of sediment archives

4.5.2.1. Elemental characteristics compared to granulometric composition

In order to properly interpret the geochemical behaviors of sediments, particular attention

should be contributed to elemental characteristics in relation to grain size parameters (Honda

and Shimizu, 1998), when assuming similar provenance. In case of absence of intensive

pedogenesis (see low Chemical Index of Alteration (CIA) in Fig. 4-6B), this heterogeneity

may originate in minerals’ re-distribution during transportation and deposition (Schettler et

al., 2009) and their deviant resistances to weathering processes (von Eynatten et al., 2012).

Most samples from U1 and U3 have higher proportions of coarse silt to sand. The clay

proportion is relatively low, indicating that the transport mode is high energy fluvial or

aeolian saltation, rather than a lacustrine or still water environment (Blott and Pye, 2012). The

clay and fine silt proportions can be considered as detrital component. Medium silt

contributes a considerable fraction to aeolian dust or loess deposits which are transported

several hundred meters above the surface, being trapped and deposited subsequently (e.g.,

Prins et al., 2007; Nottebaum et al., 2014, 2015b). Winter monsoon here is a low altitude

phenomenon (Vandenberghe, 2013) which may be a possible transporting mechanism. In

contrast, fine particles transported by Westerlies can reach up to 10,000 m (Sun et al., 2000;

Sun, 2004). Silty sand (Fig. 4-3) and fine sand proportions may indicate near surface

suspension and dust storm saltation occurring in winter and spring (Dietze et al., 2014). The

medium to coarse sand proportions may represent high energy transporting processes

(Derbyshire et al., 1998) exerted by the Hei River.

Previously, the grain size dependence of elemental (mineralogical) compositions of various

archives were tested and evaluated by a set of studies e.g., Honda and Shimizu (1998), Yang

et al. (2007), Schettler et al. (2009), Garzanti et al. (2009), Lupker et al., (2011) and von

Eynatten et al. (2003, 2012). However, few studies have investigated bulk geochemical of

granulometric proportions (cf., Fig. 4-5 and correlation matrix Fig. 4-8).

Median, mean, mode values are negatively correlated to alkali metals and ferromagnesian

elements e.g., Al, K, Ti, V, Fe, Ni, etc. In contrast, the positively correlated K and Si may

49

originate from orthoclase in sands (e.g., K-feldspar; Divis and McKenzie, 1975). Hf may

originate from weathering-resistant heavy minerals e.g., zircon, which are often enriched

during long aeolian or fluvial sediment recycling (cf., Owen, 1987). Thus, silica-rich minerals

(e.g., orthoclase) and allochthonous heavy minerals may be the main provenance for the

coarser materials in this endorheic basin. Ferromagnesian, mafic and mica source elements

encompassing Al, K, Ti, V, Fe, Ni, Cu, Zn, Ga Y, Nb, Nd, Pb and Th (cf., Nickel, 1954; von

Eynatten et al., 2012) are broadly correlated (+0.5-0.6) to clay and fine to coarse silt (Fig. 4-

8), S is only highly correlated to coarse silt (+0.5). Mg is only correlated to clay and fine silt

(+0.6 and +0.5, respectively). Ca and Sr are generally correlated to medium to coarse silt

proportions (cf., Fig. 4-8). Granitoid source elements e.g., K, As, Rb and V have negative

correlation with coarse silt. Interestingly, the heavy mineral sourced element Zr shows

slightly negative correlations with fine to medium sand (Fig. 4-8). von Eynatten et al. (2012)

reported the hump-shaped pattern of Zr (i.e., relatively low abundance in both clay and coarse

sand). In contrast, Zr has been suggested previously to be associated with coarse silt to sand in

aeolian sediments (e.g., loess) (Yang et al., 2006). Thus, whether Zr can be attributed to the

sand proportion needs to be tested in future studies. Furthermore, with similar geochemical

behavior as Zr, Hf is positively correlated to medium to coarse sand (Fig. 4-8). Arsenic and

Nd have no conspicuous correlations with any GSD fractions. This may be ascribed to their

diversified sources or presence in different minerals with different weathering resistances.

Silicon shows only intermediate correlation with medium sand (50 %). Being one of the

typical elements in all aeolian sand, clay minerals and some heavy minerals (Millot, 1970;

Pettijohn et al., 1973), Si cannot be easily applied as proxy for the coarse grain size

proportions.

4.5.2.2. Elemental characteristics of sediment archives

After deflation and disaggregation in the weak chemical weathering environment (Fig. 4-6B),

groups of minerals with similar physical and geochemical behaviors (e.g., durability) are more

likely to be enriched / depleted in specific archives or grain size fractions (von Eynatten et al.,

2012), in case no shifts of source rocks exist.

In the light of above statement, distinct enrichment / depletion of elements for certain archives

can be observed in this study: (i) Aeolian sands have higher contents of Zr, Hf, Ba and Si.

Aeolian sands transported in saltation mode during dust / sand storms have typical minerals of

quartz, feldspar, pyroxene and some heavy minerals (e.g., zircon; Fitzsimmons et al., 2009).

(ii) Lacustrine clay is composed of fine materials whose source minerals could be micas,

50

kaolinite, chlorite and some other ferromagnesian minerals (Hillier, 1993). They have a

common enrichment in K, Rb, Y, Ti, V, Nb, Fe, Cu, Zn, Al, Ga, Pb, As Th and Nd. Titanium,

Mg and Fe-bearing mafic minerals such as biotite are more often present in fine sediments

(Nesbitt and Young, 1996; Young and Nesbitt, 1998). (iii) Fluvial sands show high contents

in Mn, Mg, Cl and Na. This archive may contain more saline / evaporite minerals such as

mirabilite and glauberite formed after paleolake desiccation / saline processes (Smith and

Haines, 1964). (iv) Sandy loess has a distinctly high Sr content. Meanwhile, ferromagnesian

provenance elements e.g., Fe, Ni and V show relatively low values, which may be ascribed to

the lack of clay and fresh weathered materials in the archive. Generally, due to their similar

hydro-geochemical characteristics in pelagic and lacustrine settings, Sr behaves quite analog

as Ca (Calvert and Pedersen, 2007). Sr may engage together with Ca in carbonates in e.g.,

alkaline habitats. However, carbonates seems relatively rare in sandy loess (see low CaO and

MgO content in sandy loess (Tab. 4-3), and low aragonite and dolomite in Fig. 4-10), leading

to a distinctly different abundances between Sr and CaO (MgO). On the other hand, sandy

loess samples are corresponding to layers with spikes of quartz and feldspar concentrations in

the upper U2 (Figs. 4-4 and 4-10). And feldspar is more siliceous than most clay minerals

(Millot, 1970; Pettijohn et al., 1973). These set of evidences imply that the higher

concentration of quartz and feldspar may account for the higher Sr content (along with

relative high Si concentration) in the sandy loess. (v) The unsorted alluvial gravels have high

values of Ca, S, Cr, P and W. High content of P, S and W may result from the increasing

anthropogenic activities in the last thousand years (cf., northern Tibetan Plateau in Chen et al.,

2015). Naturally, Cr occurs in chromite together with Fe or Mg which is quite commonly

observed in nearby ultramafic intrusive rocks and ophiolite deposits (i.e., upper reach of Hei

River catchment in the western Qilian Mountains and the Bei Mountains, see Fig. 4-1 and Hu

et al., 2014). The generally high correlation between Cr and Fe, as well as Cr and Mg (Fig. 4-

8) demonstrate the chromite provenance transported not only as aeolian particles from the

west (acidic intrusive rock in Bei Mountains, Fig. 4-1) but also as fluvial material transported

by Hei River (greenschist-rich metamorphic rocks in mountain proximal reaches of the Hei

River, see Schimpf et al., 2014).

51

Fig. 4-10: Principle component scores, main mineral compositions and granulometric

properties of section HC8, RF1: weathering intensity, RF2: silicate-bearing mineral

abundance (different within lacustrine and aeolian phase), RF3: saline magnitude of

sediments, and RF4: quasi-constantly aeolian input. Lithologic patterns refer to Fig. 4-3.

Clay mineral group includes smectite, illite, kaolinite and chlorite. C, fSi, mSi, cSi, fS, mS and

cS denote the particle size of clay, fine silt, medium silt, coarse silt, fine sand, medium sand

and coarse sand, respectively.

Specifically, throughout the section HC8 and all other samples, Cl content is especially

elevated relative to Upper Crust Composition (UCC) values (Tab. 4-3), which may indicate

the generally high saline / evaporite mineral contents in late Quaternary sediments (cf., Smith

and Haines, 1964) of the dry endorheic Ejina Basin.

Furthermore, as illustrated in Fig. 4-9B and 9D, samples from the same sediment archive are

more likely to aggregate together in the score plot. Thus, multivariate statistics of

geochemical properties is a useful tool to discriminate the archive types. Outliers may be

attributed to certain subjectiveness in field discrimination of archive types and GSD

observations (Fig. 4-5).

4.5.3. Geochemical interpretation of sedimentary processes

As stated in Chapter 4.4.1, all sediments are dated to the (late) Quaternary. Although post-

depositional processes e.g., hydrodynamic sorting and diagenesis may indeed exert an

influence on the nature of geochemical compositions (Murray et al., 1992; Johnsson, 1993), in

case of late Quaternary sediments (cf., Mid- to Early-Cenozoic desert alluvium in Walker et

al., 1978 and Mesozoic sandstone and mudstone in Zhang et al., 2014b), the diagenesis

52

induced elemental alteration stays more likely in the initial stages (see Stage I in Walker et al.,

1978) and is therefore a negligible factor. On the other hand, the nature of parent rocks and

tectonic settings need to be clarified. As illustrated in the ternary A-CN-K diagram (Fig. 4-

6B), different archives exhibit generally a similar origin of mafic rocks which are close to the

CN apex (i.e., hornblende and clinopyroxene). Thus, markedly diversified terrain sources and

tectonic settings during the late Quaternary sedimentary processes can be ruled out.

Subsequently, elemental composition supplemented by XRD data from the key section HC8

can be employed to interpret the intriguing information of the sedimentary processes:

RF1 has relatively high loadings of felsic, ferromagnesian, mica and clay mineral source

elements such as Ti, Fe, K, Cu, Zn, Ga, Rb and Th (Tab. 4-4), which are all from soluble or

easily weathered minerals. These elements are commonly encompassed in clay minerals (cf.,

RF1 curve and clay mineral abundance in Fig. 4-10). Quartz is relatively stable and difficult

to decompose, while feldspars (mainly K-rich orthoclase, while Na- and Ca-rich plagioclase

are more likely to be attributed to RF2; Tab. 4-4) are more easily weathered to clay minerals

(Chen et al., 2010). Fine materials have high enrichment of least durable minerals, e.g., sheet

silicates such as micas (von Eynatten et al., 2012, cf. muscovite content in lacustrine clay and

alluvial gravels in Fig. 4-4). Hence, RF1 may indicate the varied weathering intensity, albeit

the magnitude is generally low throughout the sedimentary processes (cf., CIA in Fig. 4-10).

RF2 has high loadings of Si and negative loadings of Mg, Mn and Ca. The trend of RF2 is

generally anticorrelated (Tab. 4-4; Fig. 4-10) with dolomite / dolostone variances, which is

interpreted to precipitate within saline phase of playa basins (Last, 1990). Mg and Ca are

generally enriched together in carbonate and metamorphic silicate minerals such as

magnesite, dolomite, strontianite and pyroxene (Tucker et al., 1990). A possible origin may be

the autochthonous production of the carbonates within the lake period (IJmker et al., 2012b)

and allochthonous debris from aeolian sands (e.g., Ca in feldspathic sands, Divis and

McKenzie, 1975; cf. aeolian sands in Fig. 4) during a non-lacustrine phase. Mg was

particularly suggested as the most sensitive element to the paleolake depth (Shanahan et al.,

2013). Thus, tectosilicates (e.g., quartz and feldspar / orthoclase) are assumed to be the

predominant silicates in the aeolian U1, whereas clayey phyllosilicates (cf., RF1) e.g.,

kaolinite, illite and montmorillonite are assumed to be the prominent silicates in lacustrine

period U2 (cf., muscovite / illite content in lacustrine clay in Fig. 4-4). Hence, RF2 may

indicate still water level or in situ water budget (authigenic input) in lacustrine phases,

whereas with indication of silicate-bearing sands input during aeolian phases.

53

RF3 has high loadings of Cl and Na, which displays correlation with calcareous evaporate

minerals precipitated during the paleolake shrinkage (e.g., hydro-magnesite and halite, see

halite content in lacustrine clay (Fig. 4-4), cf. RF3 and halite pattern in Fig.4-10).

Furthermore, calcite can only be found in relatively high saline magnitude lacustrine periods,

albeit calcite determined in aeolian phases may be attributed to the mineral provenance

(intensive pedogenesis can be ruled here, see calcite and CIA in Fig. 4-10). Hence, RF3 can

be referred to the salinity component of sediments.

RF4 has generally high loadings of Zr and Hf, which are usually from dust / airborne minerals

e.g., quartz and feldspar (cf., aeolian sands in Tab. 4-3 and Chapter 4.5.2.2). Meanwhile,

factor scores do not show a pronounced downcore trend along section HC8, suggesting a

quasi-constant aeolian input into the basin systems throughout the late Quaternary.

In summary, the nature of bulk-compositions in the arid realm is governed by weak chemical

weathering, authigenic minerals and aeolian sand input. In contrast, concerning the suite of

late Quaternary sediments in this study, pedogenesis and diagenesis exert only very limited or

no influence. In addition, it needs to be noted that the generally undiversified source terrain

and tectonic settings during the sedimentary processes should be constrained as a prerequisite

before any following interpretations.

4.6. Conclusions

A practical strategy is presented to discriminate sediment archive properties and thereafter

sequence the controlling factors of sedimentary processes. The main conclusions are:

The considered records of northern Ejina Basin can be discriminated into three broad

lithological units: basal aeolian strata, lacustrine strata and the uppermost poorly-sorted

mixture of coarse sand and gravels.

Elemental characteristics in conjunction with the granulometric composition in different

sediment archives in the northern Ejina Basin are obtained. Aeolian sands have high contents

of Zr, Hf, Ba and Si. Especially Hf can be regarded as good indicator to discriminate the

coarse sand proportion. Sandy loess has high contents of Ca and Sr which both exhibit broad

correlations with medium to coarse silt proportions. Lacustrine clay has high contents of

felsic, ferromagnesian and mica source elements. Fluvial sands have distinctly high contents

of Mg, Cl and Na which are enriched in evaporate minerals. Alluvial gravels have especially

high contents of Cr which may originate from nearby Cr-rich bedrock.

54

Multivariate statistical methods applied to geochemical compositions is a useful tool to

discriminate archive types in our case. Considering the suite of late Quaternary sediments in

the arid realm, temporal variation of the bulk-geochemistry can be well explained by four

semi-quantified sediment components: (weak-) weathering intensity, silicate-bearing mineral

abundance, saline / alkaline influence of sediments, and the quasi-constant aeolian input into

the arid endorheic basin system. In contrast, pedogenesis and diagenesis exert only a limited

influence.

55

“It seems that these Heinrich events have left their

signature in the Chinese loess record. This is consistent

with simulations of the glacial climate, which imply that

the climates of the North Atlantic and China were linked

by the effect of westerly winds” (Stephen C. Porter and

Zhisheng An, 1995)

5. Lake sediments documented late Quaternary humid pulses in the Gobi Desert of

Mongolia: Vegetation, hydrologic and geomorphic implications

Kaifeng Yu 1, Frank Lehmkuhl

1, Frank Schlütz

2, Bernhard Diekmann

3, Steffen Mischke

4,

Jörg Grunert 5, Waheed Murad

6,7, Veit Nottebaum

1, Georg Stauch

1


Germany

2 Lower Saxony Institute for Historical Coastal Research, Viktoriastraße 26/28, 26382

Wilhelmshaven, Germany

3 Alfred Wegener Institute, Helmholtz Center for Polar and Marine Research, Telegrafenberg

A43, 14473 Potsdam, Germany

4 Faculty of Earth Sciences, University of Iceland, Sturlugata 7, 101 Reykjavík, Iceland

5 Department of Geography, Johannes Gutenberg-Universität Mainz, Becherweg 21, 55128

Mainz, Germany

6 Department of Palynology and Climate Dynamics, Georg-August-Universität Göttingen,

Untere Karspüle 2, 37073 Göttingen, Germany

7 Department of Botany, Kohat University of Science and Technology, 26000 Kohat, Pakistan

Considerable efforts have been devoted to decipher the late Quaternary moisture and thermal

evolution of arid central Asia. However, disparate interpretations still exist concerning

different proxies. The spatial and temporal heterogeneities have inhibited a holistic

understanding of general patterns and underlying mechanisms. To address these issues, two

parallel cores (ONW I, 6.00 m; ONW II, 13.35 m) were retrieved from lake Orog Nuur, in the

Gobi Desert of Mongolia. An array of multidisciplinary investigations of geomorphological

mapping, radiocarbon dating, geochemical and biotic studies embracing palynological and

56

ostracod valve analyses has enabled us to gain a comprehensive dataset for vegetation

development and hydrological variability over the last ~50 ka. Ample evidence reveals a

marked moisture pulse during the Marine Isotope Stage 3 (~36-~24 ka) which might have

induced the maximum last glacial expansion in the high elevated Khangai Mountains. A sharp

transition of Termination I (~11 ka) is indicated by geochemical, palynological, and ostracod

data. The lower area of the Orog Nuur catchment was dominated by Artemisia steppe in the

late Pleistocene and gradually altered to Chenopodiaceae desert steppe in the Holocene. The

early Holocene is also characterized by a relatively humid environment, albeit discordant

downcore variability of moisture and thermal signals can be derived between palynological

and bulk geochemical signals. The water body of the lake appears to be a distinct alkaline

environment during the Holocene which was subjected to frequent allogenic input and

disturbance of the anoxic states in relative to the late Pleistocene Epoch. These two humid

pulses may be the trait of a larger scale of arid central Asia that is documented in a suite of

archives. Considering kinetics, a coupled atmospheric component comprising both East Asian

Summer Monsoon and strengthened westerlies moisture supply seems to be the driving

mechanism. This coupled mode might have been modulated broadly by boreal insolation

variances. On the other hand, four harsh climatic conditions were documented in the core at

~47 ka, ~37 ka, the global Last Glacial Maximum and the Younger Dryas as playa phases. An

impeding of westerlies’ moisture conveying along with the retreated East Asian Summer

Monsoon might have contributed to these playa phases in the Gobi Desert of Mongolia.

5.1. Introduction

Being a fragile ecological system, environmental change in arid central Asia exerts

tremendous influence on a significant percentage of world population concerning nomadic

migration and alternation of political regimes (Pederson et al., 2014). The Gobi Desert in

southern Mongolia was recognized as one of the main dust sources for Chinese Loess Plateau

and pelagic sediments in the northwest Pacific (Pye and Zhou, 1989; Xiao et al., 1997; Lu and

Sun, 2000; Vandenberghe et al., 2006). Previously, general pattern of late Quaternary (in

particular Holocene) paleoclimate signatures of arid central Asia were defined and reviewed

by an array of works including Pachur and Wünnemann (1995), Yang et al. (2004, 2011),

Herzschuh (2006), Chen et al. (2008), An et al. (2008), Wang et al. (2010) as well as Wang

and Feng (2013). These compilations have suggested a distinct moisture regime of middle and

southern Mongolia in comparison to that of higher latitude Baikal catchments and East Asian

Summer Monsoon (EASM) domain of China (Yang et al., 2004; Wang and Feng, 2013).

57

Fig. 5-1: Location of the study area and paleoclimate records mentioned in the text (1, Kanas

Basin; 2, Darhad Basin; 3, Hovsgol Nuur; 4, Shaamar section; 5, Baikal Lake; 6, Kotokel

Lake; 7, Wulagai Lake; 8, Manas Lake; 9, Luanhaizi Lake; 10, Wuliangsu Lake; 11, Japan

Sea; 12, Lake Biwa; 13, Khangai Glacier; 14, Altai Glacier. More details refer to Tab. 5-3).

Panel A delineates schematic vegetation cover of Mongolia. Large-scale air mass controls

are marked with arrows, comprising East Asian Summer Monsoon (EASM), Indian Summer

Monsoon (ISM) and Westerlies. Ulaan Nuur (Lee et al., 2011, 2013) and Bayan Tohomin

Nuur (Felauer et al, 2012) are the only two reported lacustrine archives in the central to

southern Mongolia. Panel B exhibits north-south oriented topographic transect and

corresponding vegetation patterns. The eastern parts of Khangai Mountains (higher reach of

the Orog Nuur catchment) with altitudes between 3,500 m and 3,700 m a.s.l. have no modern

58

glaciers, while small ice-fields may reach down to 3,200 m a.s.l. in the western Khangai

Mountains (Lehmkuhl and Lang, 2001).

On the Mongolian Plateau and adjacent arid central Asia, continuous lacustrine sequences are

widely regarded as fundamental repository for paleoenvironmental signatures. A wealth of

works has been reported concerning the late Quaternary moisture and thermal history.

However, an array of paramount aspects still needs to be tackled. They are summarized as

follows: (i) biased or even contradictory conclusions may occur due to the interpretations of

different proxies. For instance, Schwanghart et al. (2009) advocated a warmer and wetter mid-

Holocene in Ugii Nuur (nuur = lake; in central-northern Mongolia) as inferred from the

mineral composition and bulk-geochemistry, while an arid mid-Holocene was reconstructed

from palynological studies (Wang and Feng, 2013). (ii) Most of the works poured attention

into the Holocene period, while only few records addressed the climate history during the

Pleistocene (Pachur and Wünnemann, 1995; Grunert et al., 2000; Lehmkuhl and Haselein,

2000). (iii) Substantial spatial heterogeneity is also noteworthy. An amount of studies on

lacustrine deposits were carried out in the Baikal area in Russia, in northern and western

Mongolia and in northern China (Fig. 5-1A; Tab. 5-2; Rhodes et al. 1996; Prokopenko et al.,

2001; Mischke et al., 2005; Hovsgol Drilling Project Members, 2009). Only two lacustrine

sequences were hitherto conducted in southern Mongolia, namely from the Bayan Tohomin

Nuur (Felauer et al., 2012) and the Ulaan Nuur (Lee et al., 2013) (Fig. 5-1A), albeit these two

records cover only the last ~15 ka. Furthermore, climate reconstructions might be biased by

uncertainties in the radiocarbon based age models due to lack of datable organic matter as

well as varying hard water effects (Mischke et al., 2013).

Consequently, it is indispensable to perform multidisciplinary studies on a critically selected

lacustrine sequence in attempting to gain a more unambiguous knowledge of the past

environmental changes. Here, two parallel cores (ONW I; ONW II) were recovered from the

western Orog Nuur Basin, in southern Mongolia. Based on previous investigations, a

multidisciplinary study involving sedimentology, bulk-geochemistry, palynology and

ostracod analyses were carried out, intending to:

(1) gain a compelling knowledge of moisture history in the Orog Nuur catchment over the last

~50 ka, and its governing atmospheric circulation;;

(2) elucidate the broad pattern of interrelated vegetation and hydrologic history of the Orog

Nuur catchment;

59

(3) generate and refine understandings on the spatial and temporal traits of moisture and

thermal signals in southern Mongolia, with aid of comparison with adjacent aeolian,

pelagic and mountainous glacial expansion records.

5.2. Study Area

The Orog Nuur catchment lies in the Valley of Gobi Lakes in southern Mongolia (Fig. 5-1A).

The Orog Nuur is a brackish habitat that is subjected to desiccation and it occasionally turns

into a playa environment. Tuyn Gol (gol = river) is the main inflow that originates from the

southern slopes of Khangai Mountains around 220 km north of the lake. The Orog Nuur has

no conspicuous outflow, resulting in a hydrologically closed basin system.

Fig. 5-2: Geomorphological map of the Orog Nuur Basin, three sets of lake beaches were

synthesized based on previous investigations. Section A1 is located at the third (highest) set of

sand ridge, more details see Chapter 5.5.2.2. General content is adapted after J. Grunert and

F. Lehmkuhl, and original cartography by T. Bartsch.

5.2.1. Geologic and geomorphic setting

The northwest-southeast orientated Valley of Gobi Lakes (ca. 1,400 to 1,800 m above sea

level, a.s.l.) is an elongated intramontane depression between the Siberian Craton and the

Tarim and Sino-Korean Cratons (Baljinnyam et al., 1993; Windley and Allen, 1993; Bardarch

et al., 2000; Cunningham, 2013). The Valley of Gobi Lakes is bounded by the Khangai

60

Mountains in the north and Gobi Altai Mountains in the south (Fig. 5-1B). The maximum

elevation of the Khangai Mountains is around 4,000 m a.s.l. at Otgon Tenger Uul in the west

(4,008 m a.s.l.; 47°36’30”N, 97°33’09”E) and ca. 3,000 m a.s.l. at Erhet Hayrhan Uul in the

east (3,037 m a.s.l.; 46°39’18”N, 101°13’56”E), whilst the summit of the Gobi Altai is in

3,957 m a.s.l. (Ikh Bogd Uul: 44°59’42”N, 100°13’51”E; uul = mountain). A left lateral

strike-slip fault is located at the southern side of the Khangai Mountains (Tapponnier et al.,

1979; Walker et al., 2007; Cunningham, 2013), while the southern part of the Orog Nuur

watershed exhibits the tectonic escarpment of the Gobi Altai and gently inclined alluvial fans

(Baljinnyam et al., 1993; Vassallo et al., 2011). The elongated forebergs of the Gobi Altai are

markedly faulted, warped and partially uplifted by thrust faults (Owen et al., 1997). As an

example, the 1957 earthquake (magnitude 8.3) caused a left-lateral displacement of 6 m and

vertical displacements approached 1.5 m (Baljinnyam et al., 1993), demonstrating that

tectonic movement may lead to some misinterpretations regarding shoreline ridges or terraces

(Fig. 5-2). Paleozoic volcanic and plutonic rocks are exposed in the basin and some places are

overlain by conglomerates. Previously, a much larger lake extents were reconstructed based

on three sets of shorelines (Lehmkuhl and Lang, 2001; Komatsu et al., 2001; Grunert et al.,

2009). These shorelines are situated 3 m, 10 m, 16 m (10 and 16 m = lake level 1, L.L.1) and

23 m (L.L.2) above the modern lake-level (Fig. 5-2). East-west oriented mobile dunes and

cuspate bars lie in the vicinity of the Orog Nuur (Stolz et al., 2012).

5.2.2. Climatic and vegetation setting

Mongolia is dominated by a continental climate characterized by dry and cold winter and a

precipitation maximum in summer (An et al., 2008). The mean annual precipitation along a

north-south transect ranges from ~500 mm to <100 mm (Fig. 5-1A). 65-75 % of the

precipitation occurs in summer due to the enhanced south-eastern cyclonic activity along the

polar front (Kostrova et al., 2014). Only ~100 mm precipitation is delivered in the lower Gobi

Desert of southern Mongolia, while the higher elevated Khangai Mountains can reach ~400

mm. According to records at Bayankhongor weather station near the Orog Nuur, the mean

July and January temperature in the Gobi Desert are ~16 °C and ~-20 °C, respectively, while

the mean annual temperature is about -2 °C to 5 °C (Felauer, 2012). In spring, dust- or

sandstorms frequently occur with maximum wind speeds exceeding 20 m/s (Hempelmann,

2010; Felauer, 2012). There is generally no snow cover because of the immense winter

dryness related to the predominant Siberian Anticyclone (Kostrova et al., 2014).

61

Fig. 5-3: Schematic lithologic column, photographs of sliced drilling core ONW II along with

the reference logs. Abbreviations: cl, clay; fsi, fine silt; msi, medium silt; csi, coarse silt; fs,

fine sand; ms, medium sand, cs, coarse sand. Mudstone or siltstone is represented in

schematic colors according to core description. In general, greyish calcareous clays (upper

fraction), siltstone and intercalated sandstones constitute the entire lacustrine sequence.

In general, the dominant soil types in central to southern Mongolia are dry-steppe

Kastanozems. Other common soil types are Chernozems/dark Kastanozems in the forest-

steppe zone as well as brown semi-desert soils (Tamura et al., 2012). The modern vegetation

follows a north-south transect from mountain taiga, forest steppe, grass steppe, desert steppe

62

and finally to desert communities (Fig. 5-1A; Hilbig, 1995; Gunin et al., 1999). Detailed

modern vegetation of the Valley of Gobi Lakes see Felauer (2011). In the forest steppes of the

Khangai and Altai Mountains, Larix sibirica is the dominating tree species on the northern

and eastern slopes, while steppe prevail the sun-exposed southern and western slopes (Miehe

et al., 2007). Pinus sibirica can be found in the restricted mountain taiga of the Khangai

Mountains (Hilbig, 1995). The uppermost boundary of the alpine meadows is ~2,900 m a.s.l.

(Lehmkuhl and Lang, 2001). Most of the lowland regions in the basin system are

characterized by open desert steppes and bare gobi surface. Dwarf bushes and reed-halophytes

are the dominant communities in the lower reaches of the Orog Nuur catchment (Murad,

2011).

5.2.3. Paleo- and modern glacial setting

As the depositional basin of Orog Nuur is linked to the higher Khangai Mountains via Tuyn

Gol, Quaternary glacial variance may play a pivotal role in terms of water supply and the

clastic delivery. In modern times, only the western Khangai has small ice caps and the

average equilibrium line altitude (ELA) was estimated in a range from 3,900 to 4,000 m a.s.l.

(Fig. 5-1B; Rother et al., 2014). The modern periglacial zone reaches down to 2,700 m a.s.l.

(Lehmkuhl and Lang, 2001). A set of Last Glacial Maximum (LGM) moraines at an altitude

of ~2,900 m. a.s.l. in the southern Khangai Mountains was dated thereafter to >21 ka

(Lehmkuhl and Lang, 2001). In addition, Ritz et al. (2006) determined ages of alluvial

sediments of the forelands of Gurvan Bogd (bogd = massif) by 10

Be cosmogenic nuclides

exposure dating and supposed an aggradation of sediments in the context of the global glacial

terminations. In the Gobi Altai, permafrost developed at ca. 22-15 ka and degraded since ~13

ka, and was thereafter diminished since the early Holocene (Owen et al., 1998).

5.3. Material and methods

5.3.1. Cores retrieval and sampling strategy

Two parallel sediment cores (ONW I: 6.00 m, 45°03’48”N, 100°34’39”E; ONW II: 13.35 m,

45°04’28”N, 100°35’08”E, 1,219 m a.s.l.) are retrieved from the western part of Orog Nuur

during the playa phase (Figs. 5-1A and 5-2). The core was opened, described and documented

in the field. Samples were separated and transported to University of Göttingen, Free

University of Berlin and the Alfred Wegener Institute in Bremerhaven for palynological,

ostracod, X-ray fluorescence and X-ray powder diffraction analyses. The remaining material

63

of the cores is preserved at RWTH Aachen University for grain size and TOC analysis. The

paleoenvironmental reconstruction is carried out primarily based on the longer core ONW II.

Table 5-1 The AMS 14




ages are subtracted by a reservoir effect of ~3,400±2,100 a BP before computing to

calibrated ages and generating the Bayesian age-depth model as given in Fig. 5-4.

Core

Nr.

Dept

h

(cm)

Lab Nr. Dating

material

14C ages

(a BP)

δ13C

(%)

2σ (cal a

BP)

(95.4 %)*

Estimated

real age

according to

lithostratigr

aphy

(cal a BP)*

Calibrated

Age based on

Bayesian age-

depth

modelling (cal

a BP)

Median Age

based on

Bayesian

age-depth

modelling

(cal a BP)

ONW

II

94 Erl-15622 Bulk

material

12,581±

103

-25.4 14,325-

15,192

- 5,054-16,854 10,697

ONW

II

199 Erl-15623 Bulk

material

13,242±

85

-24.2 15,691-

16,138

11,000-

13,000

5,988-17,523 11,499

ONW

II

271 Erl-15624 Bulk

material

17,733±

149

-25.4 21,042-

21,819

13,000-

18,000

10,905-22,893 17,703

ONW

II

392 Erl-15625 Bulk

material

17,713±

180

-24.5 20,945-

21,857

18,000-

19,000

10,827-22,924 17,046

ONW

II

571 Erl-15626 Bulk

material

19,738±

179

-23.8 23,344-

24,150

- 13,528-25,249 19,604

ONW

II

778 Erl-15627 Bulk

material

21,501±

217

-24.2 25,386-

26,122

- 15,408-27,018 21,685

ONW

II

1,19

2

Erl-15628 Bulk

material

46,103±

2,281

-25.4 44,908-

49,914

- 37,748-49,716 45,303

ONW

II

1,27

2

Erl-15629 Bulk

material

40,429±

1,340

-24.9 42,221-

46,226

- 34,226-47,355 40,873

ONW

I

386 Erl-12107 Bulk

material

16,020±

105

-28.1 19,049-

19,560

18,000-

19,000

- -

ONW

I

588 Erl-12108 Bulk

material

17,642±

112

-23.4 21,020-

21,653

- - -

* The reservoir effect is determined by taking average of the difference between the boldface ages in these two columns.

5.3.2. Lithology and chronology

In total ten radiocarbon ages were determined from bulk organic materials (Tab. 5-1) using

accelerator mass spectrometry (AMS) at the University of Erlangen-Nürnberg and

subsequently calibrated to calendar years using CALIB rev. 7.1 (Stuiver et al., 1998) with the

latest IntCal 13.14c calibration curve (Reimer et al., 2013). All radiocarbon ages are presented

as calibrated years (cal a BP) with a 2 σ error bar. The upper 24 cm of the core ONW II were

not retrieved. The R package “Bchron” (Parnell et al., 2008) was employed here to set up a

Bayesian age-depth model for core ONW II (Fig. 5-4) including its calibration using the

IntCal calibration curve mentioned above. Reservoir effects at different anchor points were

assessed by (i) identifying Termination I in bulk-geochemical and palynological sequences,

and (ii) comparing the Younger Dryas (YD) - related sand layer in Orog Nuur and Ulaan

64

Nuur (Fig. 5-1A; ~200 km northeast of Orog Nuur; Lee et al., 2011, 2013). All of the

radiocarbon ages were first subtracted by the reservoir effect and then computed to calibrated

calendar years and used in the establishment of the Bayesian age-depth model. Two

radiocarbon ages from core ONW I (386 cm, 588 cm) are employed to compare timing of the

stratigraphic units relative to ages of counterpart depths in core ONW II (Tab. 5-1; Fig. 5-4).

Fig. 5-4: Polylinear regression age-depth model of core ONW II generated from the

radiocarbon ages. Reservoir effect is corrected by (i) late Pleistocene / Holocene transition

(Termination I) as inferred from the suite of granulometric, palynological, ostracod and

geochemical studies (marked as black triangle), and (ii) the Younger Dryas related sand layer

(Ulaan Nuur, Lee et al., 2013). Frequency-dependent magnetic susceptibility (χFD), electrical

conductivity (EC), CaCO3, TOC and C/Natomic ratios of parallel core ONW I are employed to

cross-validate the age model of core ONW II. More details with respect to the

lithostratigraphy and chronological framework see text in Chapter 5.4.1.


Grain Size Distributions (GSD) of all samples with 5 cm intervals (114 samples for ONW I;

258 samples for ONW II) were measured using a Beckman Coulter LS13320 laser diffraction

particle size analyzer with a detection range of 0.04-2,000 µm. Preparation of samples follows

Konert and Vandenberghe (1997). Before measuring, a parallel test (HCl/H2O2- and H2O2-

only-treatment) was done to test the impact of possible incorporated calcium carbonate on the

65

resulting GSD, revealing however no pronounced differences (Schulte et al., 2016). Thus, all

measurements were performed uniformly without hydrochloric acid treatment. For clayey

sediments subsamples of 0.1 g and for coarse sands 1 g were utilized. 700 µl of 35 %

hydrogen peroxide was added to remove the organic components. Samples were then dried at

70 °C for two days. Sodium polyphosphate was added afterwards to avoid coagulation. The

Mie Theory was applied for GSD calculations (Özer et al., 2010; Schulte et al., 2016). The

final GSD represents the average of two subsamples, each of which was measured twice to

ensure reliability. In case notable offsets existed between subsamples, the measurement

procedure was repeated for the sample.

Fig. 5-5: Lithologic units of the core ONW II, and corresponding granulometric,

geochemical, and color variances. The grey bars depict the primary sand layers inferred from

the granulometric and lithostratigraphic characteristics of the core.

5.3.4. Electrical conductivity and magnetic susceptibility measurements

Electrical conductivity (EC, here equals specific conductivity) and magnetic susceptibility

(MS) were carried out exclusively on poweder sediment from core ONW I as an

environmental and salinity proxy. 50 ml water and 0.01 g KCl were added for each sample,

the EC with 10 cm interval (1 cm thick each subsample) was measured after 30 min using a

WTW LF 196 probe. MS at 1 cm interval was measured using a Barrington MS2 with a MS2

E sensor.

66

5.3.5. Carbon, nitrogen, sulfur, TOC and CaCO3 analyses

Carbon (TC), nitrogen (TN), sulfur (TS) concentrations, and the Total Organic Carbon (TOC)

content were determined, in order to gain a better understanding of the geochemical habitats

of the former water sediment interface. Samples were dried at 36 °C for 24 hours. The

samples in zinc capsules were subsequently analyzed in CHNS Analyzer EA3000. The

Scheibler gas method (DIN 19684) was employed to calculate the CaCO3 content, and 1-5 g

samples were utilized respectively. Calculation equation is as follows:

𝐶𝑎𝐶𝑂3 (%) = 𝑉 ∗ 𝑝 ∗ 0.1602/(𝑡 + 273) ∗ 𝐸 ∗ 𝐾

Where V is the volume of CO2 (cm3), p represents the atmospheric pressure (mm Hg), t

denotes room temperature (°C), E is the loss of weight (g) and K exhibits the correction factor

(here 2 is used). TOC was subsequently computed following Scheffer and Schachtschabel

(2002). The C/Natomic ratios throughout the ONW II core were computed and examined for

further interpretations (Meyers and Teranes, 2001).

5.3.6. X-ray fluorescence analysis, color analysis and multivariate statistics

The X-ray fluorescence (XRF) analysis was carried out in Alfred Wegner Institute (AWI) for

Polar and Marine Research in Bremerhaven, Germany, using an Avaatech XRF Core Scanner

configured with Canberra X-Pips 1500-1.5 detector (slit 10 mm, 10, 30 keV, 300 and 700

µA). The scanning interval is 1 cm, resulting in 1,335 measurements for the core ONW II.

Processes comprising subtracting a background curve, applying statistical corrections for

physical processes in the detector and deconvolution of overlapping peaks were performed

using scanner equipped WinAxil package (NIOZ and Avaatech, 2007). To gain a direct

knowledge about the depositional environment of former sediment-water interface, two

factors were computed based on the element dataset using factor analysis given in Yu et al.

(2016). Thereby, high loading elements for each factor in conjunction with the factor scores

provide intriguing insights into the sedimentary processes (Yu et al., 2016). In addition, X-ray

diffraction (XRD) for 14 samples at different depths was performed to assess the general

mineral composition and provide additional information for the interpretation of bulk

elemental compositions. The sediment color (RGB model) was scanned using a CCD line

scan camera.

67

Fig. 5-6: Diagram of analyzed taxa of pollen, fungal spores, aquatic algae and non-pollen


aggregated into groups consist of arboreal plants (trees), shrubs, subshrubs/herbs/grasses,

aquatics, as well as coprophilous fungi and Glomus.

5.3.7. Palynology

In total 123 sediment subsamples of 1-2 cm3 were acquired in 5-10 cm intervals from core

ONW II and treated with KOH, HCl, HF and subsequent acetolysis for palynological analyses

(Erdtman, 1960; Moore et al., 1991). Lycopodium spores were added as exotic marker to

calculate pollen concentration (n/ml). With few exceptions, at least 400 pollen of terrestrial

plant were counted per sample. Pollen grains in clumps were counted separately (Schlütz and

Lehmkuhl, 2007). A total of about 50 pollen and 40 non-pollen palynomorphs (NPP: spores,

algae, etc.) were identified. Based on a variety of specific literatures (Müller, 1970; Vanky,

1994; Wang et al., 1995; Beug, 2004; Van Geel and Aptroot, 2006) and a collection of more

than 5,000 type slides, selected taxa and groups are arranged and presented in ecological

groups. All percentages of pollen and NPPs refer to the pollen sum of terrestrial plants. Beside

presented (Fig. 5-6), the tree taxa also comprise low abundances of Quercus, Ulmus, Tilia and

Alnus. The group of coprophilous fungi covers spore types related to the genera Arnium,

Cercophora, Delitschia, Podospora, Sordaria, Sporormiella, and Trichodelitschia. In

addition, the concentration of the pollen sum and of burnt plant particles (“charcoal”) as well

the ratio of Artemisia to Chenopodiaceae (A/C ratio) are given (see supplementary material).

The average of A/C ratio is marked by a line at 5.8 (Fig. 5-6). The pollen diagram was plotted

using the software C2 version 1.5 (Juggins, 2007) and was based on changes in taxa

68

abundance divided into 9 local pollen assemblage zone (LPZ), mostly in conformity with the

stratigraphic units.

5.3.8. Ostracod analysis

A total of 131 sub-samples from core ONW II were collected at 10 cm intervals for ostracod

analysis in order to examine and verify the paleohydrological conditions. The sediment was

treated with 3 % H2O2 for 48 hours and washed through 0.25 and 0.1 mm meshes. All

ostracod valves were picked from the dried sieved residues and identified according to Meisch

(2000) and Mischke and Zhang (2011). Further inferences are mainly based on Hiller (1972),

Hammer (1986), Meisch (2000) and Mischke et al. (2005).

5.4. Results

5.4.1. Bayesian chronology and lithostratigraphic units

According to the Bayesian age-depth model, the basal age of the core ONW II was

determined to be between 38 and 52 ka (Fig. 5-4). For lacustrine deposits in the Valley of

Gobi Lakes, lack of datable organic matter and old carbon effects may hamper the discerning

of reliable chronological framework. The Termination I has manifested remarkable

consistency in a complete set of palynological, geochemical and ostracod studies (i.e., abrupt

increase of pollen concentration, CaCO3 abundance, and ostracod valves, respectively after

the Termination I; Figs. 5-4, 5-5, 5-6 and 5-8) on core ONW II, and thereby validate the onset

of the Holocene stratigraphy in our age model. Thus, the sediments in a depth of ~205 cm are

presumed to be ~11,000 a BP old), which is ~4,600 years younger than that in our model. In

order to further corroborate the Termination I, Younger Dryas-related sand deposits in Ulaan

Nuur sediments dated by optically stimulated luminescence (OSL) is applied (Fig. 5-1A). The

chronology of the 600 cm Ulaan Nuur core was determined by 12 luminescence ages and the

sand layer regarding the Pleistocene and Holocene transition was determined to ca. 13-11 ka

(Fig. 5-4). Thereby, an age of 13,000 a BP could be assigned to sediments in ~248 cm depth,

which is ca. 3,500 years younger than that in the age model. In the light of Lee et al. (2011)

and Mischke et al. (2013), high reservoir effect in a scale of greater than 3,000 years in the

Orog Nuur sediments is highly feasible. Likewise, based on the estimation of the real ages of

the lithostratigraphy, the raw radiocarbon ages were subtracted by a reservoir effect of c.

3,400±2,100 yrs (Tab. 5-1) and then computed to calibrated calendar years and employed in

the establishment of the Bayesian age-depth model (Fig. 5-4).

69

Fig. 5-7: Scanning electron micrographs of ostracod valves derived from core ONW II

(external views). 1, juvenile Candona neglecta left valve (LV); 2-3, Leucocythere

dorsotuberosa, 2, female right valve (RV), 3, male LV; 4-11, Limnocythere inopinata, 4,

female LV, 5, female RV, 6, noded female LV, 7, noded female RV, 8, male LV, 9, male RV,

10, juvenile LV with spines along posteroventral margin, 11, juvenile RV with spines along

posteroventral margin; 12, juvenile Pseudocandona sp. LV; 13, Ilyocypris cf. inermis RV, 14,

Ilyocypris manasensis var. cornae LV, 15, Ilyocypris cf. bradyi LV.

Subsequently, based on lithostratigraphic, together with granulometric, and geochemical

characteristics, the entire sequence is delimited into nine units as follows (Fig. 5-3):

(1) Unit 1 (1,335-1,235 cm; ~50-~43 cal ka BP) is a greyish silty-clay layer with more than

50 % of silt and clay fraction. U1 is interbedded by a 20 cm (1,315-1,290 cm)

predominant coarse sand layer. CaCO3, C/Natomic along with TOC portray extremely low

values with respect to the sand layer and gradually increase upward (Fig. 5-5).

70

(2) Unit 2 (1,235-1,120 cm; ~43-~37 cal ka BP) is dominated by dark greyish clayey-silty

sediments (60-80 % of silt to clay proportion) with several intercalated brownish sand

layers (Figs. 5-3 and 5-4). CaCO3 retains roughly stable within U2, while C/Natomic ratio

manifests a more varied pattern, with values at gyttja-like laminae (~1,120 cm) greater

than 20 (Fig. 5-5). The C/Natomic ratios together with S content in rest depth of the unit

delineate generally low values.

(3) Unit 3 (1,120-1,057 cm; ~37-~35 cal ka BP) is a light greyish silty-clay stratum with

several intercalated brownish sand layers (Fig. 5-3). The C/Natomic ratios and TOC

progressively attenuate upcore and increase slightly in the uppermost U3. S and N

contents depict higher abundances than preceding units (Fig. 5-5).

(4) Unit 4 (1,057-660 cm; ~35-~24 cal ka BP) is a dark greyish silty-clay stratum with

intercalated sand layers at ~940 cm and ~840 cm, as well as several sinuously intercalated

layers with light greyish color (~750-730 cm). TC and CaCO3 concentration illustrate

relative low values with slight variability. Two vigorous spikes of TOC and C/Natomic ratio

occur at ~820 cm and ~720 cm (Fig. 5-5), albeit no counterpart phenomenon is manifested

in lithofacies and granulometric composition (Figs. 5-3 and 5-4).

(5) Unit 5 (660-390 cm; ~24-~19 cal ka BP) is a greyish silty-clay layer, which is intercalated

by sand layers at ~540 cm and ~460 cm (Fig. 5-3). Unlike U4, the TOC and C/Natomic

ratios represent quasi-constant values. In particular, the basal fraction of U5 is laminae of

dark brownish gyttja-like sediment characterized by pronounced high S and TOC content

(Fig. 5-5). Considering their counterpart stratigraphy in the core ONWI, spikes of

magnetic susceptibility (MS) in conjunction with slightly higher EC values in the

uppermost part of U5 are illustrated (Fig. 5-4).

(6) Unit 6 (390-350 cm; ~19-~18 cal ka BP) is a salient brownish coarse sand layer which is

also captured by the parallel core ONW I (Fig. 5-4). The sand fraction ranges from 91.5 %

to 96.1 %. CaCO3 abundance exhibits extremely low values whilst the C/Natomic ratios

illustrate high values with oscillations (Fig. 5-5).

(7) Unit 7 (350-248 cm; ~18-~13 cal ka BP) is a light greyish clayey-silty layer similar to U1.

The sand fraction ranges from 10 % to 35 % with an abrupt decrease in the basal part of

U7 to ~35 % and a subsequent decrease to 10 % at the depth of 283 cm. The CaCO3 and

C/Natomic ratio depict similar patterns as U1.

(8) Unit 8 (248-205 cm; ~13,000-~11,000 cal a BP) is dominated by the sand fraction (50-90

%). This sand layer is also documented in the parallel core ONW I and the adjacent Ulaan

Nuur (Fig. 5-1A). CaCO3 abundance remains analogue to U7, while the C/Natomic ratio

71

shows more fluctuations similar to the global LGM phase (U6). Throughout U8, S content

exhibits remarkable low values (Fig. 5-5).

(9) Unit 9 (205-25 cm; ~11-? cal ka BP) is marked by fine sediments comprising the highest

clay fractions (30-50 %) and the lowest sand proportion (1-10 %). An abrupt increase of

the CaCO3 (i.e., calcite, unpublished XRD data) and concomitantly diminished S content

are represented in U9. C/Natomic ratios decrease from ca. 20 to <10 upward and increase

incrementally approaching top of the core.

5.4.2. Bulk-geochemistry of the lacustrine sediments

Two factors were derived based on the multivariate analysis (Yu et al., 2016) on the bulk-

geochemistry of the lacustrine sequence. Factor 1 denotes high loadings of Fe, Mn and Al,

while factor 2 manifests particularly high loadings for Ca, Sr and Cl. Scores of each factor are

given in Fig. 5-9. Four concurrent periods exhibiting low values of factors 1 and 2 are

delineated in U1 (~1,300 cm), U2 (~1,120 cm), U6 (390-350 cm) and U8 (248-205 cm) (see

yellow bars in Fig. 5-9). In the Holocene (U9), factor 1 has extremely low values while factor

2 exhibits high values. Compared with the uppermost portion, the basal part of U9 represents

lower values for factor 1 and higher values for factor 2.

5.4.3. Palynological results

With values around 0.5 % tree pollen is infrequent throughout the whole archive and never

reaching more than 3.5 % (Fig. 5-6). Findings of Hippophae shrubs are more abundant for

most of the lower part at around 1.4 % (max. ca. 8 %) and Ephedra values exceed in the upper

part over 45 %. The genus Artemisia, includes several species of subshrubs and herbs, is

absolutely predominant with values around 70 % (max. 83.5 %). Chenopodiaceae are the

second common taxon. Their values are around 12 % but increase to 71.5 % when Artemisia

decreases at the top of the profile. For most of the profile the A/C ratio fluctuates between

0.13 and 14.5, but shows a clear trend to lower values and lower variability approaching the

top. Beside Artemisia and Chenopodiaceae only Poaceae occur in every of the 123 samples,

mostly with around 2.5 % and never but not reaching much more than 9 %. The pollen of

sedges (Cyperaceae) and some groups of Asteraceae (Matricaria-type, Liguliflorae) occurs in

about 120 samples respectively, while Rheum and Caryophyllaceae are represented in nearly

110 samples, Brassicaceae and the Senecio-type (Asteraceae) occur in around 100 samples. A

maximum of more than 10 % is reached only by the Cyperaceae and the Matricaria-type. The

pollen concentration is around 5,000 grains per ml (n/ml) but reaches up to greater than

72

41,000 n/ml in the upper part of the core (Fig. 5-6). In addition, nine local pollen zones

(LPZs) are mostly in conformity with the stratigraphic units delimited in Chapter 5.4.1 (Fig.

5-6). Pollen concentration (n/ml) and A/C ratios are employed to generally constrain semi-

quantified thermal (T) and effective moisture (M) anomalies (Fig. 9; Ma et al., 2008, 2013;

Yang and Scuderi, 2010; Zhao et al., 2012).

5.4.3.1. Unit 1 (LPZ 1; 1,335-1,235 cm; ~50-~43 cal ka BP)

Artemisia dominates with pollen percentages at around 70 %, while Chenopodiaceae stay

below 10 %. The A/C ratio increases, while percentages of spores (coprophilous fungi,

Glomus-type) and aquatics (Botryococcus, Pediastrum) are low.

Fig. 5-8: Absolute abundances of determined ostracod species, number of taxa and number of

valves plotted against lithologic units.

73


Chenopodiaceae increase to 30 %, Artemisia falls below 50% and the A/C ratio to 1.6. Values

of spores and most aquatics increase considerably in the second half.


Values of Artemisia are around 70 %, Chenopodiaceae are below 20 %. Spores stay quite

common, while some green algae (Botryococcus) and rotifer resting eggs (cf. Trichocerca

cylindrica) decrease considerably. Other green algae (Pediastrum) remain on a low level and

the amount of mandibles of Chironomidae is from now on recurrent over 1 %.

5.4.3.4. Unit 4 (LPZ 4; 1,057-660 cm; ~35-~24 cal ka BP)

In general, the abundance of Artemisia increases slightly reaching values around 80 % while

percentages of Chenopodiaceae decrease weakly, leading to some high A/C ratios in the

middle of LPZ 4 (including the maximum A/C ratio of 14.5 through the core). After a short

suspend the before sporadic finds of Salix pollen become more frequent. Aquatics like rotifer

resting eggs (55 %) and Pediastrum (43 %) reach their absolute maxima before the middle of

LPZ 4 and decrease substantially thereafter. In addition, only the upper boundary of the LPZ 4

shows slightly different depth compared with that of the Unit 4 in lithostratigraphy (Figs. 5-2

and 5-6). To provide a consistent outline in the following discussions, time slices of the

lithostratigraphic units are employed given that there is no unambiguous boundary taxa-

marker through those depths.

5.4.3.5. Unit 5 (LPZ 5; 660-390 cm; ~24-~19 cal ka BP)

Percentages of Artemisia are reduced around the middle of LPZ 5 while Chenopodiaceae

values increase and therefore the A/C ratio falls to values below 4. Finds of green algae

(Botryococcus, Pediastrum) and rotifer resting eggs (cf. T. cylindrica) strongly reduce or even

disappear around the turn to LPZ 5. Short after spore values are reduced as well. The

continuous finds of Chironomidae since LPZ 3 expires abruptly at the end of LPZ 5 for the

rest of the record. The pollen grain concentration increases to values of greater than 8,000

n/ml and up to 17,000 n/ml short before the end of LPZ 5.

74


In Fig. 5-6, the LPZ 6 is represented by a single sample (372 cm) of low pollen concentration

(400 n/ml) and a low pollen sum of 200 grains instead of more than 400 in the samples below

and above. Due to insufficient pollen content, two further analyzed samples are not plotted.

The amount of Hippophae pollen seems to be quite high in and around LPZ 6 and the low

values of cf. T. cylindrica and Pediastrum continue the trend from the end of LPZ 5.


With values of some 10,000 n/ml, the pollen concentration is not only much higher than in

LPZ 6 but even considerable higher than in LPZ 5. While the curves of Artemisia and

Hippophae decline towards the end of LPZ 7 and the curve of Salix gets disrupted, pollen of

Ephedra, including the E. distachya-type and the E. fragilis-type, increases for the first time

to above 5 %. The A/C ratio drops markedly at the top of LPZ 7 below its mean level (5.8)

and coprophilous fungi vanish for nearly the rest of the record. Botryococcus and Pediastrum,

after reaching high values again, nearly disappear at the end of LPZ 7.


The pollen concentration is reduced and Artemisia and Hippophae drop to their lowest values

till then while Chenopodiaceae pollen becomes much more common. The A/C ratio falls

sharply to lowest values (0.4-1.7) so far and Ephedra increases for the first time to over 30 %.

Algae and resting eggs are present in small numbers. Pollen of the Sparganium-type, here

locally named Typha angustifolia/australis, increases drastically to its highest amounts (max.

13 %), while it occurred only very sporadically in the preceding record.

5.4.3.9. Unit 9 (LPZ 9; 205-25 cm; ~11-? cal ka BP)

After a sharp rise the Artemisia curve falls again and Chenopodiaceae become dominant

resulting in A/C ratios <1 and <0.4 lastly. While the record of Hippophae ends with the onset

of the LPZ 9, Ephedra curve reaches its highest values. Tribulus terrestris, sporadically till

then, exhibits a more closed curve in LPZ 9, but never exceeds 1.5 %. Finds of the Juniper-

type occur for the first time, but only very sporadically. After a strong increase, Botryococcus

finds disappear. T. angustifolia/australis values decrease considerably and remain below 3 %.

75

Furthermore, a salient increase of the pollen concentration from some 5,000 n/ml to >41,000

n/ml occurs at the onset of LPZ 9 (Fig. 5-6).

5.4.4. Ostracod species and valve abundances

Valves of eight ostracod species were identified from the sediments of core ONW II:

Limnocythere inopinata (unnoded valves of female and male specimens, and noded female

specimens), Ilyocypris cf. bradyi, Ilyocypris cf. inermis, Ilyocypris manasensis var. cornae,

Sarscypridopsis aculeata, Pseudocandona cf. compressa, Candona neglecta and

Leucocythere dorsotuberosa (female and male valves) (Fig. 5-7). A total of nine samples did

not contain any ostracod valves. The maximum abundance of 380 valves per sample is

recorded at 187 cm. Valves of L. inopinata represent the dominant taxon throughout the core.

The other taxa are only sporadically recorded (Fig. 5-8):

(1) Unit 1 (1,335-1,235 cm; ~50-~43 cal ka BP): L. inopinata is the predominant species in

U1. The number of valves displays an increasing tendency upcore and approaches 142 at

1,244 cm. Two carapaces of L. inopinata are preserved at 1,284 cm.

(2) Unit 2 (1,235-1,120 cm; ~43-~37 cal ka BP): The ostracods exhibit minimum valve

number at 1,158 cm (5) and a maximum at 1,124 cm (157). Eight carapaces of L.

inopinata are recorded at ~1,190 cm.

(3) Unit 3 (1,120-1,057 cm; ~37-~35 cal ka BP): U3 is marked by an incipient increase of the

valve number (generally >150). One to sixteen valves of I. cf. inermis are recorded at the

upper part of U3 (1,074-1,057 cm).

(4) Unit 4 (1,057-660 cm; ~35-~24 cal ka BP): A significantly high valve number (309) is

achieved in the basal part of U4 (1,034 cm). Specifically, C. neglecta and L.

dorsotuberosa occur at 877-827 cm (Fig. 5-8). One valve of I. cf. bradyi appears at 794

cm and one valve of I. cf. inermis is represented at 683 cm.

(5) Unit 5 (660-390 cm; ~24-~19 cal ka BP): Two carapaces of L. inopinata were recorded in

the basal part of U5 (653 cm). a high valve number (312) was recorded at 415 cm. This

layer is overlain by sediments with I. cf. bradyi at 407 cm.

(6) Unit 6 (390-350 cm; ~19-~18 cal ka BP): U6 is marked by a low abundance of valves.

One valve of I. cf. inermis is preserved at 373 cm. L. inopinata has vanished throughout

this unit.

(7) Unit 7 (350-248 cm; ~18-~13 cal ka BP): L. inopinata is recovered from the very

beginning of U7, albeit the valve number remains relatively low. One valve of I. cf.

inermis occurs at 285 cm and 257 cm, respectively.

76

(8) Unit 8 (248-205 cm; ~13-~11 cal ka BP): U8 is marked by frequent appearance of I. cf.

bradyi at 227-207 cm. A suite of species comprising I. manasensis var. cornae, S.

aculeata and P. cf. compressa are represented for the first time in core at 207 cm.

(9) Unit 9 (205-25 cm; ~11-? cal ka BP): U9 is characterized by a transient increase of total

taxa of ostracod as well as the number of valves (Fig. 5-8). The valve number per sample

reaches its maximum (380) of the core at 187 cm. I. cf. bradyi and I. manasensis var.

cornae are preserved at 202 cm. In the uppermost part of the core (31 cm), three valves of

I. cf. bradyi are represented.

5.5. Discussion

5.5.1. Vegetation dynamics and hydrologic history of the Orog Nuur

Biomes and corresponding environmental conditions can be derived from palynofacies (Figs.

5-6 and 5-9; e.g., Tarasov et al., 2011). However, beside the ostracod fauna (Fig. 5-8), in

terms of direct hydrological signals of the lacustrine sediments, bulk-geochemical fingerprints

would be a vital indicator:

(1) For the high loading elements of factor 1 (i.e., Fe, Mn and Al), aluminosilicates and micas

in clayey sediments were recognized as the main host (Millot, 1970). Considering the

behavior of Mn in the pelagic realm, insoluble oxi-hydroxides (Mn3+

and Mn4+

) are

represented in oxygenated environments, while soluble Mn2+

occurs in anaerobic

environments (Calvert and Pedersen, 2007). In terms of higher lake level i.e.,

suboxic/anoxic water bottoms, more soluble Mn will be readily engaged into carbonate

lattices and thereby will share analogue variance pattern as CaO or CaCO3, which seems

largely untenable in this study (Ca has high loading for factor 2). This is consistent with

the C/Natomic ratios of the bulk material, those values greater than 20 can be ascribed to the

frequent input of allochthonous materials (Fig. 5-5; Meyers and Lallier-Vergès, 1999). In

summary, factor 1 may point to the redox state of the depositional environment, from

which higher scores indicate reduced environments.

(2) Those elements for factor 2 (i.e., Ca, Sr and Cl) are always associated with carbonates

(Tucker and Wright, 1990). The analogue pattern between calcite and the CaCO3 along

with the absence of the dolomite and aragonite (unpublished XRD data) indicate that the

carbonates in the Orog Nuur are mostly presented as calcite. As suggested by Wünnemann

et al. (2010), calcite is practically authigenic precipitation in the lacustrine environment.

In addition, Yu et al. (2016) suggest that the aeolian activity hasn’t been significantly

77

altered. This is also demonstrated by the largely unchanged quartz and K-feldspar contents

determined by the XRD (see Chapter 6, and Fig. 6-3). Hence, factor 2 may indicate the

downcore variability of the authigenic productivity.

Based on the entire dataset, a comprehensive view of vegetation dynamics, hydrologic history

and climate developments are provided as follows:

5.5.1.1. Unit 1 (~50-~43 cal ka BP): relatively humid Artemisia steppe around a shallow

lake

As the amount of tree pollen is low and dominated by wind pollinated taxa, the wider area

around the Orog Nuur was most likely treeless. Shrubs of sea buckthorn (Hippophae) were

quite common and most probably growing on unconsolidated sand and gravels along the Tuyn

Gol and other inflows and/or the Orog Nuur shore. The buckthorn may have been

accompanied by some willows (Salix). Dry minerogenic habitats carried xeromorphic shrubs

of Ephedra and Nitraria.

As for most of the late Pleistocene, steppes were the dominating vegetation type during Unit

1. They were very rich in Artemisia, accompanied by some Chenopodiaceae. Regarding to

their potential high pollen representation due to wind transport, grasses (Poaceae, Cyperaceae)

must have been quiet sparse while insect pollinated herbs like different kinds of Asteraceae

(Liguliflorae, Matricaria-type, Senecio-type) and others (Rheum, Caryophyllaceae,

Brassicaceae) must have been more common than the grasses.

The small amounts of green algae (Botryococcus, Pediastrum), rotifer resting eggs (cf. T.

cylindrica) and chironomid mandibles (Chironomidae) suggest a lake environment. The

monospecific occurrence of L. inopinata in the lowermost unit and samples without ostracod

valves indicate a fluctuating and possibly temporary water body (Meisch, 2000; cf., grey

horizontal bars in Fig. 5-5). Increasing valve numbers from the base to the top of the unit

suggest a gradually increasing stability of the water body.

78

Fig. 5-9: Semi-quantified thermal, moisture and hydrogeochemical history as inferred from

the palynological and geochemical studies. Palynological indices (curves a and b) are

inferred from pollen concentration and Artemisia/Chenopodiaceae ratios, respectively, whilst

the geochemical indices (curves c and d) i.e., factor scores, are entrained from multivariate

statistics based on the bulk-geochemistry of the sediments (Yu et al., 2016). Fossil pollen

concentration is linked to the thermal anomalies (e.g., Conroy et al., 2009). Considering the

authigenic productivity, the Holocene may experience a distinct lacustrine phase with higher

content of calcite and carbonates. In this study, the vertical blue bars depict humid pulses,

while yellow bars denote the playa phase inferred from the granulometric and

lithostratigraphic characteristics of the core. Furthermore, as stated in Chapter 5.4.3, the

pollen assemblages preserved in the lacustrine archive may indicate merely local vegetation

signals surrounding the lower catchment of the Orog Nuur. Therefore, inconsistency between

the reconstructed effective moisture and lake level fluctuations may occur. Northern

hemisphere temperature (e): Johnsen et al. (2001), H events: Bond et al. (1993), 45 °N

summer insolation (f): Whitlock and Bartlein (1997), west subtropical pacific SST (g) and sea

level changes (h): Tudhope et al. (2001). In addition, the MIS 3/MIS 2 transition in this study

is some 24 cal ka BP, which is largely in agreement with that in Martinson et al. (1987).

5.5.1.2. Unit 2 (~43-~37 cal ka BP): relatively dry and fluctuating Artemisia steppe and

water surplus to the lake

Regarding the decreasing and strongly fluctuating Artemisia curve, the overall climatic

conditions in Unit 2 were unstable. While increases in Poaceae and Asteraceae indicate more

humid pulses, a higher abundances of Chenopodiaceae point to dryer or locally more salty

79

conditions. The increase of aquatics hints at a more stable water body of the lake. The

synchronously increase of spores of coprophilous fungi and spores of the Glomus-type

document a higher water inflow, as those fungi spores are mostly transported into lakes by

water (Ahlborn et al. 2015; Shumilovskikh et al. 2016).

The ostracod assemblage in Unit 2 is similar to those of Unit 1 apart from the general

occurrence of L. inopinata valves in all samples. Slightly more stable conditions are

suggested. Although it may be preserved in the catchment at any time in the past, the

occurrence of carapaces of L. inopinata probably results from temporarily high sediment

accumulation rates due to the sporadic delivery of large amounts of sediment during flood

events. The higher water income as documented by aquatics, fungi spores and ostracods may

indicate melt-water impulses either from the Khangai via the Tuyn Gol or by smaller streams

from the neighboring Gobi Altai. A playa phase during the upper U2 (~37 ka) is also

documented by oxidizing environment and low authigenic productivity (Fig. 5-9).

5.5.1.3. Unit 3 (~37-~35 cal ka BP): relatively dry Artemisia steppe and lake stabilization

Artemisia-steppes are again the prevailing vegetation type. Fungi spores still indicate ongoing

water inflow into the Orog Nuur. From now on, chironomids (Insecta: Diptera) became well

established in the lake for most of the late Pleistocene, possibly indicating slightly less

brackish conditions (Chen, 2009). As the A/C ratios during Unit 3 and in the prevailing time

of Unit 2 are low, the river discharge documented by the fungi spores might be more

generated by precipitation in the higher elevations of the catchment/pastures (Khangai, Gobi

Altai) than in the low lands surrounding the lake.

L. inopinata reaches higher abundances in Unit 3, indicating a more stable environment

and/or possibly a lower sediment accumulation rate. The portion of male valves of L.

inopinata increases towards the top of the unit. More alkaline conditions were apparently

established following a period with more Cl- or SO4-dominated lake waters (see Chapter 6,

and Fig. 6-2). Higher runoff to the lake might have probably caused a dilution of the more

saline lake waters, which is also indicated by the occurrence of I. cf. inermis near the top of

Unit 3. The species typically prefers spring habitats and waters flowing from springs (Victor

and Fernando, 1981; Meisch, 2000; Mischke et al., 2005). Springs are situated today at the

southern shore of the Orog Nuur fed by water from the flanking Gobi Altai.

80

Fig. 5-10: Synthesis and comparison of paleoenvironmental variability in Mongolia and

adjacent regions since the Marine Isotope Stage 3 as developed in this study and other

selected records from literatures: 1, Kanas Glacial (Xu et al., 2009; Zhao et al., 2013); 2,

Manas Lake (Rhodes et al., 1996); 3, Darhad Glacial (Gillespie et al., 2008); 4, Hovsgol

Nuur (Murakami et al., 2010); 5, Luanhaizi Lake core LH2 (Mischke et al., 2005b); 6, Baikal

Lake (Williams et al., 1997; Prokopenko et al., 2001); 7, Kotokel Lake (Shichi et al., 2009); 8,

Retreat of southern boundary of the Gobi Desert (Feng et al., 2007); 9, Shaamar Section (Ma

et al., 2013); 10, Wuliangsu Lake (Selvaraj et al., 2012); 11, Wulagai Lake (Yu et al., 2014);

12, Biwa Lake (Xiao et al., 1997); 13, Japan Sea (Nagashima et al., 2007); 14, Dust model in

central Asian desert (Nilson and Lehmkuhl, 2001). Here, the vertical blue bars depict humid

pulses, while yellow bars denote the playa phase inferred from the granulometric and

81

lithostratigraphic characteristics of the core. Calibrated radiocarbon ages were labeled as

black triangles near the bottom axis. Climate intervals including the Younger Dryas (YD) and

Last Glacial Maximum (LGM) were demarcated near the uppermost axis. More details

concerning the proxies of each archive refer to Tab. 5-2 and locations refer to Fig. 5-1A.

5.5.1.4. Unit 4 (~35-~24 cal ka BP): relatively humid Artemisia steppe and transition

from maximum high to quite low lake levels

Higher A/C ratios point to somewhat less arid conditions in the landscape around. With regard

to the spores, the water inflow into the lake underwent no substantial changes. In the first half

of Unit 4 extremely high frequencies of rotifer eggs and Pediastrum point to an amelioration

of living conditions for water organisms. This might especially include much less brackish

conditions implying a fresh to oligohaline Orog Nuur of greater water depth (Jiang et al.,

2006, Zhao et al., 2007; Kramer et al., 2010). I. cf. inermis is present at the base of the unit,

suggesting substantial inflows and lower salinity levels (Victor and Fernando, 1981; Meisch,

2000; Mischke et al., 2005). The mostly monospecific ostracod assemblage and low valve

abundance afterwards indicates a return to less stable lake conditions and a higher salinity.

The few shells of C. neglecta and L. dorsotuberosa recorded at ca. 28 ka (Fig. 5-8) were

possibly washed to the lake from neighboring habitats during floods or may represent a few

species living at the extreme ends of their tolerance ranges.

The Orog Nuur experienced a higher lake level and somewhat fresh water conditions in the

beginning of Unit 3 (~ 35 ka) until about 30 ka BP. This MIS 3 lake high stand seems to be a

characteristic for arid central Asia as well as the Tibetan Plateau (Feng et al., 2007), albeit

debate still exists as the ages may be biased as determined by different chronological methods

(Lai et al., 2014; Long and Shen, 2015), which is out of the scope of this study. As the A/C

ratios point to a relatively humid landscape, the effective moisture must have increased in

general, possibly by lower temperatures documented in the altitudinal down move of the

landscape belts (Fig. 5-11; e.g., Lehmkuhl and Haselein, 2000).

5.5.1.5. Unit 5 (~24-~19 cal ka BP): Artemisia steppe and playa to shallow lake

While the A/C ratios indicate relative humid conditions, fungi spores point to lasting river

water inflow, nevertheless low and respectively interrupted evidences of the aquatics hint at a

deterioration of the lake environment in the beginning of Unit 5. Probably glaciers as water

resource have diminished by the end of the prevailing Unit 4. The maximum peak of the

82

Glomus-type might document short-lived run-off events with fast soil erosion but overall low

water surplus. Subsequent algae and rotifers were quite common again while chironomids

seem to diminish during Unit 5 before they disappear permanently. Chironomids

disappearance and the synchronous reduction of rotifers and Pediastrum may result from

changes in the salinity. Slightly higher valve abundances in Unit 5 indicate a diminished

sediment accumulation rate and/or slightly more stable environmental conditions. Lower

abundances of male valves of L. inopinata closer to the top of Unit 5 probably result from an

increase in salinity and predominance of Cl- or SO4-dominated lake waters.

5.5.1.6. Unit 6 (~19-~18 cal ka BP): Artemisia steppe and sand accumulation

From three pollen samples in Unit 6 only one (372 cm) yielded a minimum of palynological

remains. The overall low pollen concentration coincides with sand accumulation in the lake.

The relatively abundant pollen of Rheum might came with the sand by strong winds into the

playa basin, as Rheum nanum and most other rhubarbs of Mongolia grow in desert habitats,

from where sand is blown out (Hilbig, 1995; Grubov, 2001). As chironomids are eliminated

and demanding aquatics (rotifers, Pediastrum) are low, the water quality and amount may

have been limited, possibly caused by a decreasing inflow (low fungi spores). The almost

monospecific occurrence of L. inopinata, lowest valve numbers and low numbers of male

valves of L. inopinata suggest unstable conditions and Cl- or SO4-dominated, relatively saline

lake waters.

5.5.1.7. Unit 7 (~18-~13 cal ka BP): Artemisia steppe, higher lake level and decreasing

river influence

The landscape was still dominated by Artemisia steppes. Some amelioration of the lake

ecosystem is indicated by higher Pediastrum abundance, while the stay away repopulation by

chironomids point to quite salty conditions. One could speculate that an inhospitable brackish

water body was seasonally overlaid by light fresh water, making a blooming of Pediastrum

possible. At the end of Unit 7 the decrease of Hippophae and Salix together with the offset of

coprophilous spores marks a strong decrease in river discharge into the western basin of the

Orog Nuur. The contemporaneous increase in Chenopodiaceae and Ephedra indicates the end

of Artemisia steppe dominance. The ostracod assemblage of Unit 7 is almost identical to those

of Unit 6 and conditions apparently remained similar. A few valves of I. cf. inermis in the

upper part of the unit probably result from sporadic flood events.

83

5.5.1.8. Unit 8 (~13-~11 cal ka BP): Establishment of deserts around a strongly shrinking

lake

Unit 8 stands for the displacement of Artemisia steppe by desert communities. The spreading

of Chenopodiaceae reflect the desiccation of the wider landscape possibly including the

establishment of salt tolerant Chenopodiaceae species on the salty playa enclosing the

shrinking Orog Nuur. Ephedra includes typical desert species, which are less salt tolerant and

often inhabit places with strong physical weathering like rocks and pebbles (Grubov, 2001;

Schlütz and Lehmkuhl, 2007). The desertification of the landscape came together with low

river influence (Hippophae, Salix, Glomus-type) leading to a nearly disappearing of algae and

rotifers. The sudden spreading of T. angustifolia and/or T. australis - two highly salt tolerant

cattails - and the only representatives of the Sparganium-type at the Orog Nuur (Hilbig, 1995)

indicate marshlands near the core site. The occurrences of valves of I. cf. bradyi, I.

manasensis var. cornea, S. aculeata and P. cf. compressa in the upper part of Unit 8 indicate a

decreasing salinity in the lake and apparently wetter conditions. Recorded valves of I. cf.

bradyi and S. aculeata suggest that flowing waters entered a relatively brackish lake. I. cf.

bradyi is known as typical stream dweller whilst S. aculeata prefers slightly brackish

(oligohaline to mesohaline) stagnant waters (Ganning, 1971; Meisch, 2000; Mischke et al.,

2005). In summary, the YD event correlated phase is marked by a playa to shallow swamp

habitat, influenced by sporadic pulses of fluvial inputs. This is corroborated by the low values

of redox and authigenic productivity (Fig. 5-9).

5.5.1.9. Unit 9 (~11-? cal ka BP): desert steppe and a shallow lake of high salinity and

alkalinity

The abruptly increased pollen concentration (Fig. 5-9) points to overall higher plant

productivity. This may be associated to the improved climatic conditions and mark the

inception of the Holocene period (Fig. 5-9; e.g., Mackay et al., 2012). Likewise, this is in

concordance with the somewhat higher occurrence of trees probably indicating more effective

moisture in higher elevations. While birches (Betula) may have grown in the nearby Gobi

Altai, where small relicts of Betula exist today (Fig. 5-11), pollen of conifers (Pinus, Picea,

Abies) comes most probably from remote mountain areas (e.g., Ma et al., 2008). Artemisia-

steppes recovered in the beginning, indicating higher effective moisture, but did not become

dominating again. As documented by Chenopodiaceae and Ephedra, the lower elevations

stayed arid. The reestablished steppe may have allowed a higher density of wild and/or

domestic animals, leading to an increasing grazing pressure in the dry habitats documented by

84

the expansion of the ectozoochorous grazing weed T. terrestris (Miehe et al. 2007). In

addition, Juniper shrubs (Juniperus-type) might have been favored by grazing (Miehe et al.

2009). In terms of ostracod records, the I. cf. bradyi and I. manasensis var. cornea occur only

in the lowermost part of Unit 9, indicating relatively low salinities and higher runoff to the

lake. Afterwards, the almost monospecific ostracod assemblage of L. inopinata valves, the

absence of male valves of the species and a few samples lacking ostracod valves indicate

harsh fluctuating environmental conditions and the establishment of generally Cl- or SO4-

dominated, relatively saline lake. In the modern context, note that the Tuyn Gol flows into the

eastern main basin of the Orog Nuur until today, the offset of the Hippophae and Glomus-type

finds may indicate the establishment of the recently active sediment barrier, segmenting the

small western basin including the coring sites from the direct influence of the Tuyn Gol.

Today the water in the western basin comes in addition from springs, located at the margin as

well as inside the basin, and with a small creek fed by the precipitation in the Gobi Altai. To

the end, our archive documents a strong decrease and the expansion of Chenopodiaceae

deserts, the still today predominating vegetation (Fig. 5-1B).

Likewise, in the perspective of bulk-geochemistry, sediments exhibit particularly high

authigenic productivity in the early Holocene and slightly decreased afterwards (Fig. 5-9).

The higher authigenic productivity is characterized by high abundance of Ca and Sr in the

sediments. The Holocene water body is presumably an alkaline habitat with higher

concentrations of cations that can be bound to chloride. However, this high alkalinity or

salinity in the water body may not necessarily result from high temperatures. As illustrated in

Fig. 5-9, the reconstructed thermal condition of the Holocene appears to be slightly better than

that during the MIS 3. This mild temperature typically favors the flourishing of Betula in

northeastern Asia biomes (Fig. 5-6; Heusser, 1989). Hence, rather than temperature, a

significant change of the sediment provenance might have contributed to the distinctive saline

and alkaline environment in the Holocene, albeit assessment of the provenance change is out

of the scope of this study. On the other hand, the supply of the meltwater may provide

abruptly increased riverine inflows into the sedimentary basin and thereby destroy the former

established suboxic habitats, and invoke low values of redox conditions through the Holocene

(Fig. 5-9). In summary, the relatively higher lake level in the early Holocene might be

explained by climate-induced higher meltwater flux from ice, snow and frozen ground in the

higher catchment, and subsequent drying till today as a result of diminishing inflow and

higher evaporations.

85

Tab. 5-3: Synthesis of lacustrine, pelagic, aeolian and glacial sequences (spanning at least

the MIS 3) in the vicinity of the Gobi Desert. For locations of each site refer to Fig. 5-1A.

Nr. Name Archive a

Lat.

(°N)

Long.

(°E)

Chronology b

Time

span

Analytical

techniques

and proxies c

Reference

1 Kanas

Basin

G 48°41’ 87°01’ OSL 0-50

ka

M Xu et al. (2009); Zhao

et al. (2013)

2 Darhad

Basin

G 51.3° 99.5° 14C, CRN 0-150

ka

ELAs, sh Gillespie et al., (2008)

3 Hovsgol

Nuur

L 51°53’ 100°21’ 14C 0-27

ka

ICP-MS,

AAS, TOC

Fedotov et al. (2008);

Hovsgol Drilling

Project Members

(2009); Murakami et al.

(2010)

4 Shaamar

Section

A 50.2° 105.2° 14C 0-35

ka

p, ms, gs,

CaCO3

Feng et al. (2007); Ma

et al. (2013)

5 Baikal

Lake

L 52°32’ 106°12’ 14C, mag 0-75

ka

d, BSi, Prokopenko et al.

(2001)

6 Kotokel

Lake

L 52°45’ 108°05’ 14C 0-50

ka

p, ch, BSi Shichi et al. (2009)

7 Wulagai

Lake

L 45°25’ 117°29’ 14C 0-48.5

ka

loi, gs Yu et al. (2014)

8 Manas

Lake

L 45°45’ 86°00’ 14C 0-37.8

ka

p, d, XRD,

δ13C, 18O

Rhodes et al. (1996)

9 Luanhaizi

Lake

L 37°35’ 101°12’ 14C, TIMS 0-50

ka

os, ICP-AES,

thmag

Mischke et al. (2005)

10 Wuliangsu

Lake

L 40°58’ 108°54’ 14C 0-36.5

ka

ms, gs, XRF,

CaCO3

Selvaraj et al. (2012)

11 Japan Sea O 37°04’ 134°42’ 14C 0-

144.4

ka

ESR, CI, ms Nagashima et al. (2007)

12 Biwa Lake L 35°15’ 136°01’ tephro 0-145

ka

EQF, SEM, gs Xiao et al. (1997)

13 Khangai

Glacier

G 46°-48° 96°-103° CRN 0-45

ka

M Rother et al. (2014);

Lehmkuhl et al. (2015)

14 Altai

Glacier

G 46°-52° 82°-95° - 0-

>132

ka

ELAs Lehmkuhl et al. (2011)

Notes:

a Archives: L, lacustrine; G, paleoglacial; A, aeolian loess and interbedded paleosol; O, oceanic material.

b Chronology methods: mag, paleomagnetic events/reversals; OSL, optical stimulated luminescence dating;

IRSL, infrared-stimulated luminescence dating; CRN, beryllium-10 cosmogenic nuclide dating; TIMS, thermal

ionisation mass spectrometry.

c Analytical techniques and proxies: d, diatom abundance; BSi, biogenic silica content; loi, loss on ignition; gs,

grain size distribution; p, pollen spectra; m, glacial moraines; ch, charcoal; os, ostracod; thmag, thermomagnetic

measurements to check the Curie Temperature of each sample; XRF, X-ray fluorescence analysis; XRD, X-ray

diffraction analysis; ms, mass magnetic susceptibility; AAS, atomic absorption spectrometry; ICP-MS,

inductively coupled plasma mass spectrometer; ICP-AES, inductively coupled plasma atomic emission

spectroscopy. ELAs, equilibrium-line altitudes; sh, paleolake shoreline; SEM, scanning electron microscopic

photographs; TOC, total organic carbon; ESR, electron spin resonance; CI, crystallinity of quartz; EQF, aeolian

quartz flux.

86

Notably, the palynofacies (i.e., local pollen zones/LPZs in Chapter 5.4.3) employed here may

only reveal the vegetation evolution in the lower catchment surrounding the lake. Direct

hydrologic signatures could merely be referred from algae, ostracod specimens and bulk-

geochemical characteristics of the lacustrine sediments. Hence, not surprisingly, geochemical

and palynological indices may occasionally result in inconsistent or controversial

interpretations regarding the past environmental changes (Fig. 5-9). This specious mismatch

between palynological and geochemical signals may be attributed to a nonlinear response of

the catchment biomes to the orbital scale insolation modulations (Schwanghart et al., 2009;

Wang and Feng, 2013). For instance, in the Holocene, high temperature induced higher

evapouration may reduce the annual moisture and thereby the A/C ratios. However, the

thermal characteristics may exert only limited impacts on lakes levels and accompanied

geochemical nature of the core sediments. A holistic view of different proxies as well as

geomorphological features is therefore indispensable in attempts to reconcile the

inconsistency, and interpret sound and reliable paleoenvironments. On the other hand, in

contrary to the multiproxy reconstruction for the hydrological/moisture conditions, credible

paleo-thermal reconstruction in the lacustrine or terrestrial realm still awaits novel techniques.

The testified functions as given in Ma et al. (2008, 2013) seem not applicable in this study due

to lack of pivotal taxa. Apart from the palynomorph- (Fig. 5-9), chironomid-, and diatom-

based reconstructions, pilot studies on (i) fluid inclusions (Lowenstein and Brennan, 2001)

and (ii) organic signatures of branched glycerol dialkyl glycerol tetraethers (brGDGTs;

Tierney et al., 2010) preserved in lacustrine sequences would be practical and promising in

achieving sound thermal signals, albeit the following discussions have focused more on

moisture signals in terms of the broadly arid condition in the Gobi Desert of Mongolia.

5.5.2. Geomorphic evidence of the past environmental change in the Orog Nuur

catchment

In the scenario of Orog Nuur catchment, past glacial moraines in the higher elevated Khangai

Mountains and shorelines of the Orog Nuur in lower catchment area (Figs 5-1 and 5-2) are

pivotal aid that could be employed in paleoenvironmental reconstructions.

5.5.2.1. Glacial advances in the higher reaches of the Orog Nuur catchment

Glaciology and permafrost study is an essential complement to gain a more comprehensive

knowledge of the past environment in the higher reach of Orog Nuur catchment (Fig. 5-1B).

In the Khangai Mountains (above 3,000 m a.s.l.), as determined by the cosmogenic nuclide

87

dating of the terminal moraine boulders (Lehmkuhl et al., 2016), the oldest detected glacial

advance was dated in the westernmost part of the Khangai to ~40 ka, which roughly

corresponds to the uppermost of U2 (Fig. 5-11). Thus, in the western Khangai Mountains, ca.

40 ka marked the inception of the glacial advance in MIS 3. In addition, two granite erratic

boulders (47.85°N, 97.31°E; 47.68°N, 97.21°E) from terminal moraines of Otgon Tenger in

the western Khangai were dated using 10

Be/26

Al exposure dating, yielding an age of 30.5±1.4

ka and 37±2.7 ka (Lehmkuhl et al., 2016). Moreover, the MIS 3 glacial advances were also

suggested in (i) western Khangai (Rother et al., 2014), (ii) Chinese Altai (Xu et al., 2009;

Zhao et al., 2013), and in (iii) Darhad Basin (Fig. 5-1A; Gillespie et al., 2008), implying that

these MIS 3 glacial advances may be a common phenomenon in a greater spatial scale in the

arid central Asia. Interestingly, the glacial expansion during the MIS 3 interstadial is even

larger than that during the LGM (Rother et al., 2014), albeit the MIS 3 is a relatively warm

interstadial (Dansgaard et al., 1993). During the MIS 3, the time span of the glacial advance is

in phase with the lacustrine period in this study, whereas the LGM and YD glacial advances

are accompanied by playa phases as inferred from the core and the accumulation of dune

sediments in the adjacent sections of the Orog Nuur (Section A1 in Fig. 5-2; Figs. 5-10 and 5-

11). For the Khangai Mountains and adjacent regions, a suite of the compilations from

Lehmkuhl et al. (2011, 2016) and Rother et al. (2014) have implied that nival and periglacial

features together with major glacial expansions occurred between 28 and 24 ka, and thereafter

a dry and cold phase (24-15 ka) accompanied by loess accumulations is inferred.

Consequently, the considerable humid pulse during the mid- to late-MIS 3 (~36-~24 ka as in

this study) may be the main driving mechanism for MIS 3 glacial advance, whereas it is not

pertained in case of the paucity of moisture in the LGM, when lower temperature in particular

in the high elevated eastern Khangai Mountains would have been the main forcing factor

(Rother et al., 2014; Lehmkuhl et al., 2016). On the other hand, no glacial advance was

detected in the higher eastern Khangai Mountains, where merely patches of undated

periglacial remains were observed (Fig. 5-11), which may be attributed to the unfavored

thermal amelioration and hence the elevated ELA since the early Holocene. In southern

mountain ranges (i.e., Gobi Altai) of the Orog Nuur, the degradation of permafrost since

Termination I was concluded by Owen et al. (1998), which coincides with the fewer

occurrences of periglacial features on the southern slopes of eastern Khangai Mountains (Fig.

5-11).

88

5.5.2.2. Ancient shorelines of the Orog Nuur

Although it might be slightly altered by neotectonic events, evidences of former sand beaches

of the Orog Nuur can still shed light on the paleohydrologic conditions of the lake. The oldest

remnants of beach sediments can be found 57-62 m above the present lake-level (L.L.3).

Section A1 (Fig. 5-2; 45°11’N, 100°46’E) provides two luminescence ages, namely HDS 047

and HDS 049 (at depth of ~0.5 m and ~3 m, respectively; Lehmkuhl and Lang, 2001). The

age of ~70 ka (HDS 047: 75±14 ka; HDS 049: 71±11 ka) serves as evidence for the highest

former lake level (L.L.3, 1280 m a.s.l.). Hence, the highest beach ridges may indicate high

lake levels in the early part of the last glacial or even the Eemian period, and these phases are

also supported by floodplain sediments in the Mongol Els (300 km west of the Orog Nuur;

Stolz et al., 2012). Consequently, the reconstructed MIS 3 lake level may be slightly lower

than the early last glacial paleolake. Considering the lower sand ridges in Fig. 5-2, recessional

beach lines reveal that the shrinking of the lake can be identified at 20 m (L.L.2), 10-16 m

(L.L.1) and 3 m above the modern lake-level. The L.L.2 and L.L.1 may be indicative of a

more or less stable lake corresponding to MIS 3 and early Holocene humid pulses,

respectively, albeit no detailed chronology is available here. Moreover, in the study of

Komatsu et al. (2001), based on the remote sensing investigations of the paleoshorelines in

the catchment and nearby basins, the stable paleolakes were estimated at MIS 3. In addition,

in the Valley of Gobi Lakes, Owen et al. (1997) suggest that more humid conditions have

prevailed between 40 and 23 ka. On the southern slopes of the eastern Khangai Mountains, as

concluded by Lehmkuhl and Lang (2001), high humidity-induced slope processes such as

sheet wash was stronger in the early Holocene.

5.5.3. Regional synthesis - humid pulses inferred from different archives in the vicinity

of the Gobi Desert

In the Orog Nuur, two chief humid pulses (~36-~24 ka, ~11-~? ka) and four relatively short

playa phases (~47 ka, ~37 ka, ~19 ka and ~13-~11 ka) were determined (Figs 5-9, 5-10 and 5-

11). Several more thin coarse-grained sediment layers can be identified through the MIS 3

humid pulses, they are however less significant as aforementioned playa phases (Fig. 5-5).

Furthermore, rather than pure aeolian sediments, this layers might be attributed to fluvial

materials as determined by geochemical compositions (see Chapter 6.5.2.4, and Fig. 6-3),

albeit their granolometric composition share similar chracteristics with aeolian sands (Yu et

al., 2016). In order to gain a holistic paleoenvironmental view in a greater spatial scale, an

array of studies in the proximity to the Gobi Desert is compiled and evaluated (Fig. 5-10).

89

5.5.3.1. Marine Isotope Stage 3

Humid conditions during the middle and late MIS 3 were recorded in northern Chinese

deserts and northern Mongolia by several works based on diverse archives and proxies

(Herzschuh, 2006; Yang et al., 2011). As illustrated in Fig. 5-10, the higher moisture signals

along with glacial advances during mid- to late MIS 3 are documented in the Kanas Basin (Xu

et al., 2009; Zhao et al., 2013), Manas Lake (Rhodes et al., 2006), Darhad Basin (Gillespie et

al., 2008), Wuliangsu Lake (Selvaraj et al., 2012), Wulagai Lake (Yu et al., 2014), Biwa Lake

(Xiao et al., 1997) and Shaamar loess section (Feng et al., 2007; Ma et al., 2013). This entire

set of records hints at a humid pulse during the MIS 3 in arid northern China and northwest

Pacific (Tab. 5-2; Figs. 5-1 and 5-10). Considering other records in Tengger Desert and

drylands of northern China, most of the high lake-level periods have also been dated to the

main interstadial of the last glaciation i.e., MIS 3 (Pachur and Wünnemann, 1995; Lehmkuhl

and Haselein, 2000), albeit age difference may occur as determined by different chronological

methods (Lai et al., 2014; Long and Shen, 2015). More specifically, a higher lake level at

Gaxun Nur was determined to 41-33 ka (Fig. 5-1A; Wünnemann and Pachur, 1998).

Likewise, a marked expansion of needle-leaved forest in the southern Mongolian Plateau and

northern China in the late MIS 3 were suggested by several studies (Pachur and Wünnemann,

1995; Ma et al., 2013). This MIS 3 humid pulse is also corroborated by low dust supply as

suggested by the model of Nilson and Lehmkuhl (2001) and in the northwest Pacific (Leinen,

1989) (Fig. 5-10). However, whether the high lake levels of Qinghai Lake have to be assigned

to MIS 3 or MIS 5 still remains elusive (Colman et al., 2007; Madsen et al., 2008; Liu et al.,

2010).

On the other hand, in the Baikal catchments and northern Mongolia, Baikal core GC-1

(Williams et al., 1997; Prokopenko et al., 2001), Kotokel Lake (Shichi et al., 2009) and

Hovsgol Nuur (Murakami et al., 2010), no humid pulses during the MIS 3 is displayed,

suggesting a distinct moisture pattern compared with those in northern China and northwest

Pacific (Fig. 5-10).

In summary, aquatic organisms such as Pediastrum and Botryococcus, as well as relative

haigh A/C ratios have revealed a humid pulse during the MIS 3 (Fig. 5-6; Unit 4 in the core

sequence), albeit this phase is not well presented in other plynological and geochemical

fingerprints (Fig. 5-9). Although there exist certain inconsistency between all this suites of

imprints, we need to focus on the direct signals such as aquatic organisms in order to better

discern the hydrological state of the Orog Nuur in the past.

90

5.5.3.2. Marine Isotope Stage 2

A fluctuating shallow lake together with a moisture deficit in the LGM and YD were implied

in the studies of Manas Lake (Rhodes et al., 1996), Luanhaizi Lake (Mischke et al., 2005),

Wuliangsu Lake (Selvaraj et al., 2012), Wulagai Lake (Yu et al., 2014), Biwa Lake (Xiao et

al., 1997) and Schaamar loess section (Feng et al., 2007; Ma et al., 2013) (Fig. 5-10). Arid

LGM conditions were previously suggested for the Asian interior by numerous works (Lu and

Sun, 2000; Bush, 2004). After the LGM, a stepwise climate amelioration was conceived by

Herzschuh (2006) as such can be correlated to U7 in our study (Fig. 5-9). Interestingly, the

reconstructed authigenic productivity curve (Fig. 5-9) exhibit broadly similar downcore

variability as west Subtropical sea surface temperature (SST) and sea level variances

(Tudhope et al., 2001), indicating certain links between the moisture supply in the Gobi

Desert and the northwest Pacific, both of which lie beneath the pathways of the boreal

westerlies and EASM throughout the late Quaternary period (Nagashima et al., 2007).

5.5.3.3. The Holocene

As illustrated by geochemical indices (Fig. 5-9), the higher effective moisture and authigenic

productivity in the early Holocene corroborates the lake expansion phase postulated in

Lehmkuhl and Lang (2001). This is also in line with a suite of studies in Luanhaizi Lake

(Mischke et al., 2005), Baahr Nuur (Feng et al., 2005), Khyaraany loess-paleosol sequence

(Feng, 2001), Qilian Mountain loess sequences (Nottebaum et al., 2015), Yili loess-paleosol

sequence (Li et al., 2011) and other westerlies domain signals (Arz et al., 2003; Ilyashuk and

Ilyashuk, 2007), albeit Herzschuh (2006) synthesized the palynologcal signals in this area and

conceived that the early Holocene optimum may not be a uniform phenomenon in the entire

westerlies domain. However, as reconstructed by palynofacies (Fig. 5-9), the increasing

temperature and decreasing moisture in the early Holocene is concordant with the Altai region

(Demske et al., 2005; Rudaya et al., 2009). To gain a broad view, a wealth of palynological

studies with focus on Mongolia was carried out by Tarasov et al. (2000, 2007), assembling the

vegetation history of Mongolia and Baikal catchment and reconstructing a broadly warm and

humid early Holocene and a deteriorated climate since the mid-Holocene. Likewise, Miehe et

al. (2007) investigated forest patches in the Gobi Altai and concluded a shift from dark taiga

to a steppe environment over the last 5 ka. In the Holocene, the greatest dust fluxes from Asia

into the northwest Pacific are estimated to be in the mid-Holocene (Leinen, 1989). A dry mid-

Holocene mega-thermal that may favor dust storms was also evidenced in studies in Inner

Mongolia (Chen et al., 2003). Moreover, late Holocene dune sediments were uncovered in the

91

sections around the Orog Nuur catchments during our field investigations (Fig. 5-2),

indicating a less favored environment after the mid-Holocene and in modern times. However,

due to the absence of at least the late Holocene sediments (the uppermost 24 cm of core ONW

II), interpretations for this time slice in our study are restricted.

In addition, as aforemented, the high temperature induced higher evapouration might have

reduced the annual moisture in the catchment, shifted the vegetation to desert steppe (e.g., the

increses in Ephedra, Chenopodiaceae, and Nitraria; see supplementary materials) and thereby

reduced the A/C ratios in the Holocene (Yang et al., 2010; Zhao et al., 2012). However, the

thermal characteristics may exert only limited impacts on lakes levels and accompanied

geochemical nature (higher contents of carbonate) of the core sediments (Fig. 5-9).

5.5.4. Potential driving mechanisms for the late glacial climate change in the Gobi Desert

From an arid central Asia perspective, the moisture signature is of particular importance with

respect to the dust activity in the Asian interior as well as for the northwest Pacific (Duce et

al., 1980; Leinen, 1989; Pye and Zhou, 1989; Nakai et al., 1993; Nilson and Lehmkuhl, 2001;

Nagashima et al., 2007). A general assertion is that the moisture signatures here have an out-

of-phase relationship with the EASM domain (Chen et al., 2008). At least in the western part

of arid central Asia, the moisture signal is believed to be brought by the westerlies

(Vandenberghe et al., 2006; Chen et al., 2008). Nonetheless, recently, based on the analysis of

modern precipitation signals, Huang et al. (2015) demonstrated that the Gobi Desert of

southern Mongolia is outside of the core region of westerlies-dominated arid central Asia (50-

90 °E, 36-54 °N). In their study, the southern Mongolian Plateau portrays a modern

precipitation variation pattern analogue to the EASM-influenced China, while an anti-

correlated relationship is obtained in contrast to the core region of the westerlies (50-90 °E,

36-54 °N). The early Holocene moisture optimum is at odds with the dry pattern in the

westerlies-dominated Hovsgol Nuur (Fig. 5-1; Prokopenko et al., 2007; Rudaya et al., 2009).

On the other hand, the reconstructed indices (Fig. 5-9) imply intimate connections between

hydrochemical conditions of the Orog Nuur and the west subtropical Pacific signals. Hence,

partially contrary to the envisaged assertion, it is likely that thermal and moisture signals in

the Gobi Desert may not be driven by unitary mechanisms. In other words, the Gobi Desert is

rather affected by an atmospheric mode comprising both the EASM and the westerlies, which

as a dynamically coupled system, is governed by the nature of boreal insolation variances

(Fig. 5-9).

92

Fig. 5-11: Synoptic compilation of hydrologic, vegetation and geomorphic history of the Orog

Nuur catchment over the last ~50 ka. Playa & dune phase is recovered from the

granulometric and lithostratigraphic characteristics of the core ONW II. More details see

Chapter 5.5.1. Glacial advance in Khangai Mountains refer to Rother et al. (2014) and

Lehmkuhl et al. (2016). As mentioned in Chapter 5.5.3.3, in the Holocene, the high

termpertature induced higher evapouration might have triggered lower annual moisture in

the catchment (see A/C ratios in Fig. 5-9), it has however less impacts on the better

hydrological condition of the water body as revealed by higher authigenic productivity (Fig.

5-9) in the more alkaline water column (see Chapter 6.5.2.2, and Fig. 6-6).

5.5.4.1. Humid pulses in the Gobi Desert

The asynchronous Holocene optimum of East Asia was demonstrated by An et al. (2000),

Kerr (2003) and Chen et al. (2008), concluding that the moisture from the northwest Pacific

can be transported into continental Asia. On the other hand, the westerlies were also

suggested as a feasible moisture supplier. For instance, a weakening of the Siberian winter

anticyclone coupled with a retreat of the northern hemisphere ice sheet would have acted as a

93

forcing factor that permits more moisture to penetrate into the Asian interior by both Atlantic

cyclone activity and the EASM (cf., Moros et al., 2004; Peltier and Fairbanks, 2006). Boyle

and Keigwin (1987) reported the connection between the north Atlantic thermohaline

circulation - namely Atlantic meridional overturning circulation (AMOC) - and temperature in

high-latitude regions. In the scale of the last glacial period, the massive iceberg discharges

may trigger global scale climate change via alteration of the AMOC (Bond et al., 1993;

Broecker, 1994; Porter and An, 1995; Böhm et al., 2015). The prominent decay of ice sheets

and concomitant weakening or shutdown of AMOC (Ding et al., 1995) were demonstrated to

be the conceivable driving mechanism for the global scale transition during the last

deglaciation (Rodwell et al., 1999; Rahmstorf, 2002; Liu et al., 2009; Clark et al., 2012).

Furthermore, the moisture history of the north Atlantic may exert fundamental impacts on the

EASM signals via the teleconnected interactions (Ivanova, 2009). In this study, the higher

lake level in the MIS 3 and early Holocene implies a broadly resembling pattern with the

higher values of 45°N summer insolation (Fig. 5-9; Whitlock and Bartlein, 1997). Likewise,

further western Altai depicts a similar and only slightly lagged pattern compared with the

boreal insolation curve (Eichler et al., 2009). Interestingly, this early Holocene optimum is

also documented in the Indian Monsoon domain (Sirocko et al., 1993; Fleitmann et al., 2003;

Gupta et al., 2003; Liu, 2011), leading to our assumption that, the Gobi Desert is governed by

a coupled Asian monsoon system that is largely modulated by boreal insolation variances.

Presumably, in the mid-latitudes, higher insolation forcing during the MIS 3 and early

Holocene has supplied more moisture into the Asian interior via pathways of EASM or

westerlies (Cheng et al., 2016).

5.5.4.2. Playa phases in the Gobi Desert

Considering the aridity of the Asian interior, Bush (2004) suggests that the arid conditions in

the LGM may result from significantly declined moisture transferred from the North Atlantic

Ocean due to the orographic blocking (establishment of the northern European ice sheet) of

the westerlies. Meanwhile, the weakening or shutdown of the AMOC may induce the

mitigation of the ocean-continental pressure gradients (Rodwell et al., 1999) and thereafter

diminish the moisture delivered by both the westerlies and EASM (Wu and Wang, 2002;

Ivanova, 2009). It is postulated that, during the YD event, the relationship between the

Rossby wave pattern of the upper-level westerlies and the tropical convection intensity may

exert an influence on the precipitation regime over the low latitude Asia and thereby promote

widespread droughts (Staubwasser and Weiss, 2006). Moreover, impeded moisture conveying

94

by the westerlies (Chiang et al., 2015) along with the southward displacement of summer

mean latitudinal position of Inter Tropical Convergence Zone (ITCZ) (Haug et al., 2001;

Fleitmann et al., 2003; Broccoli et al., 2006; Yancheva et al., 2007) would have exerted a

pivotal influence on the less moisture supply of the Asian monsoon system. Furthermore, in

this study, the four playa phases exhibit broadly concurrent phases of low values on northern

Hemisphere temperature and insolation curves (see yellow bars in Fig. 5-9), suggesting

moisture deficits in response to the coupled oceanic circulation and boreal summer insolation

variances.

5.6. Conclusions

A multidisciplinary investigation is conducted on a 13.35 m lacustrine core retrieved from

Orog Nuur, in the Gobi Desert of Mongolia. Methods embracing sedimentological,

geochemical, palynological and ostracod studies are combined with geomorphological

evidence (sets of dated terrace ridges and moraines) in order to illustrate the vegetation

development and hydrologic history during the last ~ 50 ka. For the first time, in the Gobi

Desert, semi-quantified moisture and thermal sequences that trace back to the MIS 3 are

presented. The main conclusions are summarized as follows:

In Gobi Desert of Mongolia, the MIS 3 experienced sufficient moisture supply in particular

between ~36 ka and ~24 ka; MIS 2 was subjected to cool temperature and moisture-deficient,

which was interrupted by two exceedingly cold and dry playa phases related to the LGM and

YD event; the early Holocene exhibited ameliorated climatic condition characterized by

higher lake levels. Moreover, the merely local vegetation imprints (palynological spectrums)

may be responsible for the inconsistency or even controversy between the palynologically and

geochemically reconstructed hydrological signals. Thus, in this arid realm, a holistic view of

different proxies in conjunction with the consideration of geomorphological features

surrounding the lake as well as in the higher catchment is necessary. Furthermore, we need to

bear in mind that palynological imprints indicate merely the vegetation characteristics in the

catchment. Geochemical factors suggest the redox and antigenic state of the core sediments,

they are however not direct fingerprints of the lake level variances. In order to better discern

hydrological states, direct imprints such as aquatic organisms need to be focused on.

A sharp transition from the late Pleistocene to the Holocene (i.e., Termination I) is illuminated

by palynological and geochemical imprints. Lower area of the Orog Nuur catchment was

dominated by Artemisia steppe in the late Pleistocene and was altered to Chenopodiaceae

95

desert steppe in the Holocene. The water body in the Holocene appears to be a distinct

alkaline environment which was subjected to frequently allogenic input and disturbance of the

late Pleistocene anoxic states.

The main humid pulse in MIS 3 might have been the main driving mechanism for the MIS 3

glacial advance, while the pronounced low temperature in the higher elevated catchment in

the eastern Khangai Mountains is presumably the forcing factor for the LGM glacial advance.

The early Holocene in the Gobi Desert experienced the other humid pulse, which was

concurrent with lacustrine conditions and the degradation of the permafrost features as well as

the glacial retreat.

Four major arid playa phases (~47 ka, ~37 ka, ~19 ka and ~13-~11 ka) were determined, of

which the latter two sand layers (i.e., the LGM, YD event, respectively) may provide an

opportunity to be labelled as chronological benchmarks for other lacustrine sequences in the

Gobi Desert of Mongolia. Nonetheless, further investigations still need to be carried out to

test its reliability in a larger spatial scale.

In the central to southern Mongolia, the late Quaternary moisture and thermal history may be

modulated, if not controlled, by coupled atmospheric components embracing both westerlies

and the propagation of the East Asian Summer Monsoon into the Asian interior, albeit more

details with respect to the quantified contribution from these systems remain an open

question.

96

Fig. 5-S1: Diagram of analyzed taxa of pollen, fungal spores, aquatic algae and non-pollen


aggregated into groups consist of panel (A) arboreal plants, humidity indicators, aquatic

algae, and panel (B) aridity indicators and grazing indicators.

97

“It is now clear that lacustrine sediments and

sedimentary rocks represent one of the best archives for

paleoenvironmental information available within the

entire realm of terrestrial settings” (William M. Last and

John P. Smol, 2002)

6. Geochemical imprints of coupled paleoenvironmental and provenance change in the

lacustrine sequence of Orog Nuur, Gobi Desert of Mongolia

Kaifeng Yu 1*

, Frank Lehmkuhl 1,Bernhard Diekmann

2, Christian Zeeden

1, Veit Nottebaum

1, Georg Stauch

1


Germany

2 Alfred Wegener Institute, Helmholtz Center for Polar and Marine Research, Telegrafenberg

A43, 14473 Potsdam, Germany

In the arid environment, due to the scarcity of a continuous terrestrial archive, lacustrine

sequences are often employed as a paleoenvironmental repository. However, numerous spatial

and temporal heterogeneities exist concerning previously studied sites in arid central Asia.

Furthermore, surveys using a XRF core scanning technique on lacustrine sequences retrieved

in hyperarid desert settings are largely rare. Hence, two parallel sediment cores (ONW I;

ONW II) were retrieved from Orog Nuur, in the Gobi Desert of Mongolia. Continuous, high-

resolution elemental abundances at a 1 cm scanning step size were examined in core ONW II

using XRF core scanning. Based on multivariate statistical evaluation, the bulk-geochemistry

of the core sediments are governed by (i) grain-size composition, (ii) authigenic productivity

(Ca, Cl, CaCO3) in an alkaline environment, (iii) allochthonous organic material (TOC and

C/Natomic), and (iv) terrigenous input via fluvial inflows, as well as quasi-constant aeolian

input through the late Quaternary (Al, Si, K, Ti, and Fe). To constrain the data quality,

elements with high error margins relative to measured peak areas and those elements/proxies

below the significance level during the multivariate statistics are excluded for

environmental/provenance implications. Disparate source lithotypes, as well as authigenic

productivity of the lake system existed before and after Termination I. The Holocene was

dominated by a distinct high productivity alkaline environment with more felsic and alkaline

input relative to the late Pleistocene. This might be attributed to an increased hydrodynamic

98

strength of riverine inflow and/or intensified erosion and weathering of felsic source rocks in

the upper catchment of the Orog Nuur. Therefore, in order to gain a better understanding of

the bulk-geochemistry of lake sediments, the coupled provenance and environmental

signatures, as well as land surface processes in the catchment need to be systematically

discerned. Thus, the XRF core scanning data obtained in this study would have practical and

complimentary merit for other lacustrine studies focused on the desert realm across the globe.

6.1. Introduction

Lacustrine sequences are critical archives for the reconstruction of climate and provenance

signatures. In particular, in arid central Asia, due to a scarcity of continuous loess-paleosol or

oceanic archives, lacustrine records were often employed as Quaternary paleoenvironmental

repositories (Boyle, 2001; Mischke et al., 2005). Southern Mongolia is located on the margin

of the westerlies-dominated arid central Asia and the East Asian Summer Monsoon domain of

eastern Asia (Chen et al., 2008). Acquisition of a continuous and chronologically reliable past

environmental record in this region is of substantial importance in refining our understanding

of comprehensive paleoenvironmental patterns and possible driving mechanisms. However,

numerous spatial and temporal heterogeneities exist concerning the lacustrine sequences here.

Compared with more intensively studied lake records in northern China and northern

Mongolia/Lake Baikal catchment of Russia, due to the remoteness of the region, only two

continuous lacustrine records were previously reported in the Gobi Desert of Mongolia,

namely Bayan Tohomin Nuur (nuur = lake; Felauer, 2011) and Ulaan Nuur (Lee et al., 2011).

Nonetheless, these two cores cover only the last ~15 ka. Furthermore, a commonly high

reservoir effect (contributing to 14

C dating uncertainty) of the non-varved lacustrine

sediments, as well as a lack of datable organic material in this arid context have largely

hampered a robust age-depth reconstruction (Felauer, 2011; Lee et al., 2011). A longer record

along with a well constrained chronological model (for instance using Bayesian age-depth

modelling) is therefore indispensable in providing a complementary repository to unravel the

aforementioned issues.

99

Fig. 6-1. Present-day distribution of geologic terranes in the Orog Nuur catchment. Panel A:

simplified land cover transect of Mongolia: mountain taiga, forest steppe, steppe, desert

steppe, and Gobi Desert in the south. Panel B: schematic geological setting of the catchment,

which is marked by dashed line. Northern region of catchment is dominated with Devonian-

Triassic felsic quartz syenite, granite, plagiogranite, and alkaline sediments, accompanied by

patchy intermediate mafic fractions. The southern region of Orog Nuur exhibits more of

mafic/ultramafic basalt and andesite-basalt rocks relative to the northern catchment. Alkaline

syenite occurs sporadically in the lower reach of Orog Nuur catchments. The depositional

basin is bounded by eastern Khangai in the north and the Gobi Altai Mountains in the south.

Among a suite of multidisciplinary studies on argillaceous sediments, geochemistry appears

to be a promising tool for deciphering the interplay between environmental change, source

lithotypes and sediment bulk-composition (Boyle, 2001). When considering late Quaternary

100

lacustrine sediments, the bulk-geochemistry may be controlled by source terranes, authigenic

or allogenic input, which may be modulated by past environmental conditions, while

pedogenesis and diagenesis exert merely limited overprints (Yu et al., 2016). Knowledge of

the downcore variability of major, trace, and minor element abundance along with field

investigation and careful examination of previous geologic mapping will enable us to

decipher the climate-induced provenance change throughout the depositional processes. On

the other hand, surveys considering the bulk-geochemistry and corresponding environmental

interpretations in the marine realm have been systematically reviewed (Calvert and Pedersen,

2007), whilst elucidations of the elemental fingerprints incorporated in the lacustrine

sediments still await more investigation (Roser and Korsch, 1988; Norman and De Deckker,

1990; Sinha et al., 2006). In particular, in contrast to conventional XRF measurement using

pressed pellets, the application of high resolution, time saving, and continuous XRF core

scanning to lake sediments has only attracted limited attention in the less than ten years since

its earlier applications in the marine realm (Melles et al., 2012). The XRF scanning technique

has not yielded a complete consensus on how the provenance, depositional environment, and

corresponding earth surface processes affect a suite of elements in different lacustrine settings

(Richter et al., 2006; Shanahan et al., 2008). A recent review by Davies et al. (2015)

highlights that several studies have been conducted to interpret elements, ratios and associated

environmental interpretations in the lacustrine realm. However, those implications should not

be taken as universally applicable, as there is to date rare XRF core scanning data that has

been generated in the hyperarid (desert) setting where largely no varved laminae can be

observed in the lacustrine sequences.

This work aims to provide a complementary XRF scanning logged record to fill in the gaps in

the above mentioned spatial/temporal heterogeneity of data repositories in the hyperarid

desert setting. In this study, major and trace element contents accompanied by calcium

carbonate concentration and C/Natomic ratios of the Orog Nuur lacustrine sequence are

examined in an attempt to: (i) test the possible plots of element ratios that could be employed

to extract the broad pattern of diversified source regions throughout the depositional period;

and (ii) ultimately evaluate the set of multi-proxy fingerprints and parse out the interplay

between the coupled past environmental processes, land surface processes, and provenance

change in the arid depositional system of the Orog Nuur catchment.

101

6.2. Regional setting

The Orog Nuur (45°01’-45°04’N 100°33’-100°53’E; ~1,219 m a.s.l.) is located in a

hydrologically closed basin system in the Valley of Gobi Lakes (Fig. 6-1). It is a brackish lake

subject to desiccation and occasionally turns into playa system. The Tuyn Gol (gol = river) is

the main inflow originating from the southern slopes of the Khangai Mountains around 220

km north of the lake. The continental climate of Gobi Desert is characterized by a dry / cold

winter with limited snow. More than 65% of the rainfall occurs in the summer. Annual

precipitation reaches ~100 mm in the vicinity of the Orog Nuur in the Gobi Desert of

Mongolia, while the higher elevated Khangai Mountains receive up to ~400 mm/a. The mean

annual air temperature is -2 to 5 °C. The mean air temperatures during July and January in the

Gobi Desert are ~16 °C and ~-20 °C, respectively (Felauer, 2011). The Valley of Gobi Lakes

(1,400-1,800 m a.s.l.) is an elongated intramontane depression between the Siberian Craton

and the Tarim and Sino-Korean Cratons (Baljinnyam et al., 1993; Cunningham et al., 2005).

Paleozoic volcanic and plutonic rocks are exposed in the basin (Fig. 6-1). The southern part of

the Orog Nuur watershed is represented by the tectonic escarpment of the Gobi Altai, as well

as gently inclined alluvial fans (Baljinnyam et al., 1993). As illustrated in Fig. 6-1, the

northern catchment (Tuyn Gol watershed) is dominated with Devonian-Triassic felsic quartz

syenite, granite, plagiogranite, and alkaline sediments, accompanied by minor proportions of

(intermediate) mafic rocks. The southern watershed exhibits more mafic to ultramafic basalt

and andesite-basalt rocks relative to the northern catchment. The Orog Nuur might have

occupied a much larger area in the Eemian and Marine Isotope Stage 3 (Lehmkuhl and Lang,

2001).

6.3. Material and methods

6.3.1. Core retrieval and Bayesian age-depth modelling

During the playa phase in 2007 and 2008, two parallel sediment cores (ONW I: 6.00 m,

45°03’48”N, 100°34’39”E; ONW II: 13.35 m, 45°04’28”N, 100°35’08”E, 1,219 m a.s.l.)

were recovered from the western part of Orog Nuur (Fig. 6-1B). The core was opened,

described and documented in the lab. Samples were separated and transported to (i) RWTH

for grain size and TOC analyses, and (ii) AWI Bremerhaven for X-ray fluorescence scanning

and X-ray powder diffraction analyses. The study here is conducted primarily on the longer

core ONW II.

102

Fig. 6-2. Lithostratigraphy and Bayesian age-depth modelling for core ONW II. R package

Bchron (Parnell et al., 2008) was employed. Gray shaded area represents maximum and

minimum age-depth estimates based on 100,000 smooth spline fits through the calibrated age

distributions (95% highest density region). The Bayesian age-depth model was further cross-

validated by several independent approaches: (i) two ages from the parallel core ONW I

(marked as black rectangles), (ii) sand layer and luminescence-determined age model of

Ulaan Nuur (Lee et al., 2011), (iii) downcore variance of CaCO3 concentration.

In total ten radiocarbon ages were determined from bulk materials (Tab. 6-1; Fig. 6-2) using

accelerator mass spectrometry (AMS) at the Erlangen-Nürnberg University and subsequently

calibrated to calendar years using CALIB rev. 7.1 (Stuiver et al., 1998) with the latest IntCal

13.14c calibration curve (Reimer et al., 2013). All radiocarbon ages are presented as

calibrated years (cal a BP) with a 2σ error bar (Tab. 6-1). The upper 24 cm of the core ONW

II were not retrieved. The R package “Bchron” (Parnell et al., 2008) was employed here to set

up a Bayesian age-depth model for core ONW II (Tab. 6-1; Fig. 6-2) including its calibration

using the IntCal calibration curve mentioned above. Reservoir effects at different anchor

points were assessed by (i) identifying Termination I in bulk-geochemical (sharp increase of

Ca peak areas, and CaCO3 %; Lauterbach et al. 2011) and palynological sequences (sharp

increase of pollen grains/concentration; Murad, 2011), and (ii) comparing the Younger Dryas-

related sand layer in Orog Nuur and Ulaan Nuur (Fig. 6-1A; Lee et al., 2011). All of the

radiocarbon ages were first subtracted by the reservoir effect and then computed to calibrated

103

calendar years and used in the establishment of the Bayesian age-depth model (Tab. 6-1; Fig.

6-2). The exact implementation procedures can be reproduced using appended R script. In

addition, two radiocarbon ages were determined for core ONW I (386 cm, 588 cm) to

compare and validate the timing of stratigraphic units of the counterpart depths in core ONW

II (Fig. 6-2).

Table 6-1 The AMS 14




ages are subtracted by a reservoir effect of ~3,400±2,100 a BP before computing to

calibrated ages and generating the Bayesian age-depth model as given in Fig. 6-2.

Core

Nr.

Dept

h

(cm)

Lab Nr. Dating

material

14C ages

(a BP)

δ13C

(%)

2σ (cal a

BP)

(95.4 %)*

Estimated

real age

according to

lithostratigr

aphy

(cal a BP)*

Calibrated

Age based on

Bayesian age-

depth

modelling (cal

a BP)

Median Age

based on

Bayesian age-

depth

modelling (cal

a BP)

ONW

II

94 Erl-15622 Bulk

material

12,581±

103

-25.4 14,325-

15,192

- 5,054-16,854 10,697

ONW

II

199 Erl-15623 Bulk

material

13,242±

85

-24.2 15,691-

16,138

11,000-

13,000

5,988-17,523 11,499

ONW

II

271 Erl-15624 Bulk

material

17,733±

149

-25.4 21,042-

21,819

13,000-

18,000

10,905-22,893 17,703

ONW

II

392 Erl-15625 Bulk

material

17,713±

180

-24.5 20,945-

21,857

18,000-

19,000

10,827-22,924 17,046

ONW

II

571 Erl-15626 Bulk

material

19,738±

179

-23.8 23,344-

24,150

- 13,528-25,249 19,604

ONW

II

778 Erl-15627 Bulk

material

21,501±

217

-24.2 25,386-

26,122

- 15,408-27,018 21,685

ONW

II

1,19

2

Erl-15628 Bulk

material

46,103±

2,281

-25.4 44,908-

49,914

- 37,748-49,716 45,303

ONW

II

1,27

2

Erl-15629 Bulk

material

40,429±

1,340

-24.9 42,221-

46,226

- 34,226-47,355 40,873

ONW

I

386 Erl-12107 Bulk

material

16,020±

105

-28.1 19,049-

19,560

18,000-

19,000

- -

ONW

I

588 Erl-12108 Bulk

material

17,642±

112

-23.4 21,020-

21,653

- - -

* The reservoir effect is determined by taking average of the difference between the boldface ages in these two columns.


The grain size distributions (GSDs) of 258 samples from ONW II and 114 samples from

ONW I at a 5 cm interval were determined using a Beckman Coulter LS13320 Laser

Diffraction Particle Size Analyzer. All measurements were performed uniformly without

hydrochloric acid treatment (Felauer, 2011; Schulte et al., 2016). 700 µl of 35% H2O2 was

added to remove the organic components. Samples were subsequently dried at 70 °C for two

days. Sodium pyrophosphate was added to avoid coagulation. The final GSD represents the

average of two subsamples. Each of them was measured twice to ensure the reliability. In any

104

case of conspicuous offsets exist between subsamples, the procedure was repeated for the

sample (Schulte et al., 2016).

6.3.3. TOC, CaCO3 and C/Natomic analyses

Fig. 6-3. Downcore variance of abundance of the major and trace elements expressed by

integrated peak areas (total counts per second), accompanied by the C/Natomic ratios, CaCO3

content, Rb/Sr ratio (lower values= higher catchment weathering; Davies et al., 2015), Al/Si

ratio (lower values=higher coarse-grained sands; Davies et al., 2015), and the XRD

determined minerals in sparser intervals. Silicon and K share similar patterns as Al, whilst

Mn and Co share similar patterns as Fe. Further details are presented in the supplementary

material of this paper. The grey bars depict the sand layers inferred from the granulometric

and lithostratigraphic characteristics of the core, and the dashed line mark the transition of

Termination I. The consistency between the grey bars and the elements spikes implies that the

sand fractions (Factor 1 as determined by robust statistics; Fig. 6-4) have played a pivotal

role in the alteration of bulk-geochemistry of the lacustrine sediments.

In total 258 samples (ONW II) were dried at 36 °C for 24 hours. The prepared samples in zinc

capsules were analyzed in a CHNS Analyzer EA3000. The Scheibler gas method (DIN

19684) was used to calculate the CaCO3 content. Total organic carbon (TOC) was computed

following Scheffer and Schachtschabel (2002). The C/Natomic ratios of core ONW II were

calculated for the following discussion (Meyers and Lallier-Vergès, 1999).

105

6.3.4. X-ray fluorescence scanning and X-ray powder diffraction analysis

The X-ray fluorescence (XRF) analysis was conducted using an Avaatech XRF Core Scanner

configured with a Canberra X-Pips 1500-1.5 detector (20 s dwell time, slit 10 mm, 10, 30

keV, 300 and 700 µA) (Richter et al., 2006) at the AWI Bremerhaven. For core ONW II, in

total 1,335 results at a 1 cm scanning step size were retrieved. Subtraction of a background

curve, application of statistical corrections for physical processes in the detector, and

deconvolution of overlapping peaks were performed using a scanner equipped WinAxil

package (NIOZ and Avaatech, 2007). Element abundance is given in integrated peak areas

(total counts per second or cps). To help discern their environmental and/or provenance

implications, the whole set of elements were further divided into four groups based on robust

multivariate statistics as outlined in next chapter (Fig. 6-3).

To further interpret the bulk-geochemical composition of samples from different

sedimentological settings, a standard X-ray diffraction (XRD) technique using a Philips

X’Pert MPD diffractometer with a divergence slit using array Cu Kα radiation (40 kV, 50mA)

was employed to detect the bulk mineralogical composition of 14 samples at the depth of 50,

200, 300, 350, 400, 500, 600, 700, 800, 900, 1000, 1100, 1200, 1300 cm, respectively (Fig. 6-

3). Minerals were identified from XRD patterns of samples at 3.34 Å (quartz), 3.24 Å (K-

feldspar), 3.19 Å (plagioclase), 3.024 Å (calcite), 6.15 Å (mica), 14/7.02/4.72 Å (kaolinite-

chlorite/chlorite), 3.57 Å (kaolinite) 8.47 Å (hornblende), and minor spikes at 10/5 Å (illite).

The measurement was performed from 2-35° 2θ with scan speed of 0.02° 2θ/2 sec. The

integrated peak areas (total counts) of the above mineral groups were calculated and

transformed into mineral percentages using the freeware MacDiff (Petschick et al., 1996). A

suite of original measurements of Stein et al. (1994) and Vogt (1997) that were previously

carried out at AWI were compared to ensure the comparability between the peak areas and

percent calculations.

6.3.5. Robust multivariate statistics

To provide a quantitative manner in which to segregate the whole set of multi-proxies into

detailed assemblages, a robust multivariate statistical analysis is computed using the R

package “Robustfa” (https://cran.r-project.org/web/packages/robustfa/robustfa.pdf). A

detailed computing process and the manner of data interpretation were given in Yu et al.

(2016). The robust evaluation parameters for each factor and sediment sample are presented

in Tab. 6-2. Those elements with high positive or negative (> 0.75) loadings are shown in

106

boldface and will be included for further discussion regarding their integrated environmental

processes, provenance, and land surface processes.

Table 6-2 Standard deviation, variance/cumulative proportion, multi-proxies loading for each

factor, and communalities for each proxy generated by robust multivariate statistics. High

positive or negative (> 0.75) loadings are marked in boldface. Blank values represent those

loadings which are under significance level in robust multivariate statistics. Communalities

for each element are computed by taking sum of the seared loadings, it can be referred to the

length of the element arrows in loadings plot (Fig. 6-4A), a value greater than 0.90 suggests a

well representative of this proxy in the statistical model. Factors 1 (26.4%) and 4 (25.9%)

explain more than have of the variance, they are therefore illustrated in the biplots of Fig. 6-4

for further implications. More details concerning the statistics refer to Yu et al. (2016).

Multi-proxies Factor 1: Sand

fraction

Factor 2: Authigenic

productivity

Factor 3: Organic

material

Factor 4:

Terrigenous input

Communalit

y estimates

Al -0.249 -0.129 - 0.930 0.944

Si - -0.117 - 0.911 0.844

S - -0.448 - - 0.201

Cl - 0.756 - -0.118 0.585

K -0.340 - - 0.935 0.990

Ca -0.494 0.780 - - 0.852

Ti -0.459 -0.105 - 0.826 0.904

Cr -0.243 0.197 - 0.194 0.135

Fe -0.554 - - 0.773 0.904

Rb - -0.367 - 0.675 0.590

Sr -0.108 0.697 - -0.153 0.521

Zr - -0.109 - 0.724 0.536

N (CNS) -0.804 -0.147 0.280 0.141 0.766

CaCO3 -0.548 0.779 - -0.244 0.967

TOC -0.382 - 0.954 - 1.056

C/Natomic - - 0.818 - 0.669

Mean grain-size 0.914 -0.242 - -0.234 0.949

Clay fraction -0.780 0.470 - - 0.829

Silt fraction -0.926 0.122 - 0.269 0.945

Sand fraction 0.934 -0.258 - -0.212 0.984

Standard

deviation

5.281 3.097 1.685 5.172 -

Variance

proportions/%

26.4 15.5 8.4 25.9 -

Cumulative

proportions/%

26.4 41.9 50.3 76.2 -

6.4. Results

6.4.1. Lithostratigraphic units and Bayesian chronology

According to the Bayesian age-depth model, the basal age of the core ONW II was

determined to be between 38 and 52 ka (Fig. 6-2). Termination I manifests as a remarkable

transition in geochemical patterns in core ONW II (Fig. 6-3), namely an abrupt increase of

CaCO3, indicating the onset of the Holocene unit in the age model (Fig. 6-2) (Lauterbach et

107

al., 2011). To further corroborate the depth of Termination I, a luminescence-constrained

Younger Dryas sand layer in Ulaan Nuur sediments (Fig. 6-1A) is used. In this adjacent core,

the sand layer was determined to be some 13-11 ka (Fig. 6-2). Thus, ~205 cm is presumed to

be ~11,000 a BP, which is ~4,900 years younger than that in the age model. The age of

sediment at ~248 cm is assigned to ~13,000 a BP, which is ~3,000 yrs younger than in the

Bayesian age model. Likewise, based on the estimation of the real ages of the

lithostratigraphy, the raw radiocarbon ages were subtracted by a reservoir effect of c.

3,400±2,100 yrs (Tab. 6-2) and then computed to calibrated calendar years and employed in

the establishment of the Bayesian age-depth model (Fig. 6-2). In light of Felauer (2011) and

Lee et al. (2011), a reservoir effect of greater than 3,000 yrs in the Orog Nuur cores is highly

feasible. Furthermore, the lowermost two ages are near the limit of radiocarbon dating.

However, the application of the newly established calculation curve IntCall13, as well as the

employment of the Bayesian modeling still assures a relatively reliable basal age that is

younger than ~50 ka (Tab. 6-1; Fig. 6-2).

On the basis of lithostratigraphic, granulometric, and geochemical characteristics, the entire

sequence is delimited into five units:

Unit I (1,335-390 cm; ~50-~19 cal ka BP)

Unit I is dominated by greyish clayey-silty sediments (on average 60-80%; Fig. 6-2) with

intercalations of several brownish sand layers that are marked as grey bars in Fig. 6-3. CaCO3

remains generally stable with low values (~10%) within Unit I, while the C/Natomic ratio

displays a relatively varied pattern (Fig. 6-3) with several laminations with values greater than

20.

Unit II (390-350 cm, ~19-~18 cal ka BP)

Unit II is a salient brownish coarse sand layer which is also captured by the parallel core

ONW I (Fig. 6-2). The sand fraction ranges from 91.5% to 96.1%. CaCO3 abundance exhibits

extremely low values while C/Natomic ratios display high values with oscillations (Fig. 6-2).

Unit III (350-248 cm; ~18-~13 cal ka BP)

Unit III is a dark greyish clayey-silty layer similar to Unit I. The sand fraction ranges from

10% to 35% with an abrupt increase in the basal part of Unit III to ~35% and a subsequent

decrease to 10% at 283 cm. The CaCO3 and C/Natomic ratio depict similar patterns to Unit I.

108

Unit IV (248-205 cm; ~13-~11 ca ka BP)

Unit IV is dominated by the sand proportion (50-90%). The sand layer is also present in the

parallel core ONW I and the adjacent Ulaan Nuur (Fig. 6-1; Lee et al., 2011). CaCO3 content

remains similar to Unit III, while the C/Natomic ratio shows more fluctuations similar to Unit II.

Throughout Unit IV, the S content exhibits low values (Fig. 6-3).

Unit V (205-25 cm; <~11 cal ka BP)

Unit V is characterized by fine sediments with the highest clay proportion (30-50%) and the

lowest sand fraction (1-10%). A sharp increase in CaCO3 (calcite; Fig. 6-3) and concomitantly

diminished S contents are present in Unit V. C/Natomic ratios decrease from ~20 to <10 upward

and increase incrementally approaching the top of the core.

6.4.2. Elements/minerals downcore variance and multi-proxy assemblages revealed by

multivariate statistics

The major and trace element concentrations, as well as mineral compositions throughout the

core ONW II were plotted against the depths and stratigraphic units (Fig. 6-3). In light of the

multivariate statistics (Fig. 6-4; Tab. 6-2), the whole set of elemental compositions and other

proxies are evaluated using a robust factor analysis. They are divided into four factors: (1)

Factor 1 (variance proportion: 26.4%) is positively correlated with mean grain-size and sand

percentages of the sediments and negatively correlated with the N content and silt/clay

fraction of the sediments. (2) Factor 2 (15.5%) exhibits significantly high loadings of Cl, Ca

and CaCO3 content. In general, Ca and Sr concentrations display a unique pattern that

resembles CaCO3 and hornblende, with relatively low abundances throughout the late

Pleistocene and sharp increases after Termination I. Cl exhibits a long-term trend of upcore

increase along with superimposed fluctuations. In general, the factor is an indicator for the

authigenic productivity of the lake system (see following discussion). (3) Factor 3 (8.4%)

delineates higher loadings for TOC and C/Natomic and an insignificant correlation with

elemental compositions and grain-size composition. (4) Factor 4 (25.9%) has significantly

high loadings of Al, Si, K, Ti, and Fe. This set of elements exhibits higher abundance in Units

I, III and V (relatively lower in Unit I), while the contents in Units II and IV are markedly

lower. They share broadly similar patterns to mica and chlorite. This factor is explained as

terrigenous input brought by two pathways: (i) riverine inflows via Tuyn Gol, and (ii) quasi-

constant aeolian inputs. In addition, note the relatively high variability of the element

109

concentrations of, in particular, Unit I (Fig. 6-3), which may result from the frequently

occurrence of sand layers as illustrated by the grey bars in Fig. 6-3. In this scenario, insight

into the bulk-geochemistry of each unit is more readily obtained in the dispersion plots,

wherein Units I and III share broadly similar median values and upper / lower quartiles (Fig.

6-3; supplementary figure). Factors 1 (26.4%), 4 (25.9%), and 3 (15.5%) explain ~70% of the

multi-proxy variations; they are the primary governing factors for the sediment bulk-

geochemistry.

Fig. 6-4. Loadings and scores plot of the multi-proxy data set. Panel A: loading plot of the

robust Factor 1 (grain-size) vs. 4 (terrigenous input). Panel B: score plot of the robust Factor

1 vs. 4. In multivariate statistics, the maximum direction of variance is regarded as axes.

Rotation is performed to get varimax rotated axis (x- and y-axis in Panel A) during the

analysis attempting to archive a higher loading for each elements (Tab. 6-2). Cosine of the

angle between the proxies and varimax rotated axes are factor loadings, length of the arrows

are corresponding to the proxy communalities. In panel B, all units of samples are defined

based on their lithostratigraphic units.

In terms of the minerals, mica, chlorite, quartz, plagioclase, K-feldspar, hornblende, and

calcite are determined at coarser intervals throughout the core ONW II (Fig. 6-3). Amongst

those minerals, (1) quartz and plagioclase display only minor variation through the core; (2)

mica and chlorite exhibit a highly coupled pattern that has higher values in the lake phases

(Unit I, III, and V); (3) calcite and hornblende have a similar pattern with higher values

specifically in the Holocene (Unit V) relative to the Pleistocene units; (4) K-feldspar denotes

110

a broadly reversed pattern compared with that of mica and chlorite, and appears to have a

higher content corresponding to the sand layers (grey bars in Fig. 6-3). In general, the set of

XRD data shed light on the general mineralogical composition of the studied core, and

thereby aid in further interpretation of the XRF scanning data.

6.5. Discussion

6.5.1. Evaluating the quality of the X-ray scanning data

As mentioned above, the retrieved XRF core scanning data are expressed in semi-quantified

integrated peak area/total counts per second (cps) without conversion to oxide wt % or molar

abundance.

Amongst the entire suite of elements, Ni, Cu, Zn, Ga, Br, Au, Pb, and Bi have a relatively

high error margin compared to their measured peak areas. They are therefore removed from

the multivariate statistics (for detailed data and error margin see the supplementary material of

this paper). In the statistical report (Tab. 6-2), the proxies that are under significance level are

ruled out for the following discussions as well. The remaining set of elements show measured

peak areas that are well above the theoretical detection limit of the Avaatech XRF scanning

infrastructure (Richter et al., 2006; NIOZ and Avaatech, 2007). Certain low count rates (in

particular for light elements such as Si) may be ascribed to intervals with very coarse-grained

sediments such as those in Unit II and Unit IV in this core (Richter et al., 2006; Shanahan et

al., 2008; Weltje and Tjallingii, 2008; Ohlendorf et al., 2015). However, as demonstrated in

the recently published monograph concerning the XRF scanning technique and its

applications on marine, fluvial, and lacustrine sediments (Bertrand et al., 2015), in general,

the precision of data obtained by XRF core scanning is not significantly altered by grain-size

variations. They correct the XRF peak areas for water content and grain-size, which only

slightly improves the correlation between XRF peak areas and Inductively Coupled Plasma

Mass Spectrometry (ICP-MS) determined bulk-geochemistry concentrations. This

improvement is, however, relatively minor; the difference between the correlation coefficients

is never significant at p < 0.05 level (Bertrand et al., 2015). Furthermore, Schillereff et al.

(2015) demonstrate for the first time that the employment of different XRF scanning

instruments (mainly Avaatech and Itrax, etc.) yields similar elemental patterns, implying a

robust and sound quality of the retrieved data. Meanwhile, Dulski et al. (2015) analyzed

parallel subsamples using XRF scanning and ICP-MS, also revealing a systematic relationship

between the scanned data and element concentrations.

111

In terms of late Quaternary lacustrine sediments, although diagenesis may indeed exert certain

influence on the fate of bulk-composition (Johnsson, 1993), it is, however, a less crucial

aspect compared with the alteration of paleoenvironment and source lithotypes in the entire

catchment system. Pettijohn (1973) concluded that due to the limited burial time, calcite

cements along with the negligible quartz overgrowths are virtually the only diagenetic

overprints. Thus, source terranes, paleoenvironment-induced weathering and transportation

adaptations are more likely the governing factors for the late Quaternary lake sediments. On

this account, the following discussion is generally demarcated and outlined in twofold: (i) the

paleoenvironmental (weathering, land surface processes) implications, and (ii) provenance

signals.

6.5.2. Governing factors for sediment bulk-geochemistry

In the marine realm, researches have been systematically undertaken to examining the

sediment bulk-geochemistry and its environmental interpretations (Calvert and Pedersen,

2007; Diekmann et al., 2008; Martinez-Ruiz et al., 2015). It has been suggested that

assemblages of the elements exhibit specific indications of the redox condition (Mn and Co),

diagenetic processes (Fe), authigenic productivity (Al and Ba) and the allogenic

aeolian/fluvial inputs (Zr and Rb) during sedimentary processes. Here, in the lacustrine realm,

a holistically refined view of the governing factors of sediment bulk-geochemistry, taken in

consideration of other multidisciplinary proxies, is employed using robust multivariate

statistics as a prerequisite for the following discussions.

6.5.2.1. Factor 1 (26.4%): Grain-size variation of the lacustrine sediments

This factor has the highest proportion, 26.4%, as determined by the factor analysis (Tab. 6-2).

In arid central Asia, aeolian sand plays a primary role in the terrestrial context (Kazancı et al.,

2016) and the northwest Pacific pelagic sediments (Pye and Zhou, 1989). The abundance of

the sand percentage (63-2,000 µm fractions) is also well presented as spikes for Al, Fe, and

Al/Si ratios (Fig. 6-3). In order to examine the sand-induced influence on the bulk-

geochemistry, all samples are plotted in the scores plot based on the multivariate statistics

(Fig. 6-4B). Samples with a great contribution from (aeolian/fluvial) sand fractions

particularly approach the positive end of Factor 1, while those samples of clay or silty-clay

sediments are scattered at intermediate to negative positions on the Factor 1 axis.

Interestingly, all coarse-grained samples from the Unit II (LGM) strata are relatively

clustered, while those from the Unit IV (YD) are more scattered. This may be because the

112

Unit IV strata correspond to a playa phase that is more frequently influenced by the riverine

input (mixture of both fluvial sands and finer suspended river sediments; Fig. 6-2), instead of

predominantly coarse-grained sands as in Unit II. In summary, the coarse sands have a large

influence on the sediment bulk-geochemistry via (i) fluvial inflows, and (ii) direct aeolian

inputs. However, it must also be remembered that the sediment bulk-geochemistry is not

solely a reflection of the sediment grain-size composition, indicated by the fact that that no

elements in Factor 1 are well above the significance level (Tab. 6-2).

6.5.2.2. Factor 2 (15.5%): Authigenic productivity of the saline/alkaline lacustrine

environment

Factor 2 has significantly high loadings of Cl, Ca, and CaCO3 (Tab. 6-2). Ca is generally

included in carbonates which are also common minerals in lacustrine environment (Tucker

and Wright, 1990). Calcite, low-Mg calcite, and rarely occurring aragonite are generally

authigenic precipitates in lake sediments while dolomite (along with quartz and feldspar)

could be regarded as detrital input (Sinha et al., 2006; Wünnemann et al., 2010). In core ONW

II, the abundance of calcite, Ca and CaCO3 share analogous patterns, implying that in this

study Ca is present mainly as calcite, and Unit V (Holocene) depicts notably higher authigenic

productivity relative to the late Pleistocene epoch (Fig. 6-3). In contrast, Al and Si display

lower values in the Holocene, acting as an analogue for plagioclase and K-feldspar (Fig. 6-3),

suggesting a non-productivity signature. Therefore, compared with the marine pelagic realm,

Al is not an indicator for the authigenic productivity. On the other hand, Sr has relatively high

loadings, but its content variation is still under significance level (Tab. 6-2). Sr behaves quite

similarly to Ca due to their similar hydro-geochemical characteristics (Calvert and Pedersen,

2007; Kylander et al., 2011). Likewise, Davies et al. (2015) also suggest that the presence of

Ca and Sr is commonly associated with authigenic carbonate minerals in catchments located

in arid and carbonate environments. In the lacustrine realm, halite is a ubiquitous species that

may partly be generated by re-precipitation from pore solutions (Sinha et al., 2006). Similar to

halite, the halogen element bromine may also be readily incorporated in halide salts. The

upcore increase of these halogen elements (Fig. 6-3) implies a higher salinity/alkalinity

approaching the upper part of core ONW II. Particularly in the Holocene, this high salinity

and/or alkalinity are corroborated by the significantly elevated CaCO3 abundance (Fig. 6-3).

In addition, Sulfur is in part negatively correlated with the Factor 2, albeit under the

significance level (Tab. 6-2). Sulfur might be present as sulfate (gypsum and anhydrite),

sulfide, and pyrite (Millot, 1970). As illustrated in Fig. 6-3, S exhibits higher content in Units

113

I-III, low values in Unit IV and is slightly elevated in the uppermost unit. In the late

Pleistocene units, the less disturbed lake states may have led to a relatively higher content of

sulfide and pyrite precipitation at the sediment-water interface. The sulfide minerals were then

almost entirely oxygenated when exposed in the sand/playa layer and subsequently produced

sulfate precipitation in the alkaline environment of Unit V. Alternatively, the sporadically

deposited gypsum and mirabilite in such a cold, dry lake system may also account for these S

spikes determined in Fig. 6-3. However, above all, the insignificance level of S loading for

Factor 2 (Tab. 6-2) suggests that sulfide or sulfate mineral genesis is of minor importance for

the bulk-geochemistry. This is also corroborated by clear evidence that above mentioned

minerals are not detected by the XRD analysis in this study (Fig. 6-3).

6.5.2.3. Factor 3 (8.4%): Organic material content

Factor 3 is represented by significantly high loadings of TOC and C/Natomic (Tab. 6-2). The

TOC content incorporated in the core sediments is relatively low (Fig. 6-2). The coupled

variation of TOC and C/Natomic values implies that the lacustrine system has merely minor

biogenic activity, and that the low content of the organic matter incorporated in the core was

brought into the depocenter via allogenic fluvial processes, given that C/Natomic ratios greater

than 20 indicate allochthonous input into the lacustrine system (Meyers and Lallier-Vergès,

1999).

6.5.2.4. Factor 4 (25.9%): Terrigenous components via riverine input and quasi-constant

aeolian input

Factor 4 has high loadings for Al, Si, K, Ti and Fe (Tab. 6-2). Clay or silty-clay is the

predominant sediment in the lacustrine environment (Millot, 1970). The fine detrital matrix

formed mainly from the weathering decomposition of the primary aluminosilicates, which are

the main host for Al (Calvert and Pedersen, 2007). On the other hand, Al may also be

incorporated in feldspar and micas (Millot, 1970). In core ONW II, Al and Si display broadly

identical patterns (Fig. 6-3), indicating a close link due to similar host minerals in fine

sediments, either in a feldspathic structure or as a set of clay minerals in water bodies (Millot,

1970). In addition, in an arid context such as the Gobi Desert, in situ evaporated alkaline

water was reported to be rich in Si and depleted in Al (Millot, 1970), which may account for

the lower Al abundance in Unit V when calcite or carbonates are the predominant minerals in

the water body (Fig. 6-3). In addition, several works suggest that the organic content in a

lacustrine environment may exert strong influence on the Al abundance (Davies et al., 2015).

114

However, in the arid context of the Orog Nuur, the TOC values are broadly no greater than 2-

3% (Fig. 6-2), thus making the above Factor 2 less important.

Due to the complex characteristics of its provenance ranging from fine clays to coarse sands

(Yu et al., 2016), Si presents an unclear indication of the grain-size composition (unpresented

in Factor 1; Tab. 6-2). Indeed, aeolian sands exhibit higher amounts of SiO2 (Yu et al., 2016).

However, in this context in Orog Nuur, Units II and IV may not be ascribed to purely aeolian

sand. Grain-size alone cannot be used precisely to conform that the coarse-grained layers in

this study are purely aeolian sands (Confroy et al., 2013). In a recent study on the Gobi Desert

archives, SiO2 content was found to be 71% (aeolian sands), 55% (lacustrine clay), 47%

(fluvial sands), 69% (sandy loess), and 36% (alluvial gravels), respectively (Yu et al., 2016).

This relatively lower Si content in the coarser-grained fluvial sand layer is also suggested by

Dulski et al. (2015) based on scanning of the Piànico paleolake sequence. Likewise, Hunt et

al. (2015) also demonstrate that increasing grain-size composition will not necessarily result

in an elevated Si value. Thus, the fluvial sands may indeed display lower Si abundance

relative to lacustrine clays in the core. In addition, as given in Yu et al. (2016), the sand

fraction of the fluvial sands may be as high as that of the aeolian sand, albeit more spikes in

finer fractions may be observed in the granulometic curves of fluvial sands. Thus, the

complex provenance of lacustrine sediments from both coarse sands and alluminosilicate

clays leads us to assume that the relatively coarse-grained layers (Units II and IV) may be

largely attributed to fluvial transportation, rather than purely aeolian sands. On the other hand,

in this aeolian sand and silt source region, the fluvial system normally experiences multi-cycle

deposition of the coarse material. The SiO2 values in this aeolian material show a significant

decrease through long-term weathering processes (Bridge and Demicco, 2008), rendering a

considerably lower SiO2 abundance of the fluvial sands detected by the core scanning in this

study. Multi-cycle deposition would account for the relatively lower SiO2 values in those

coarse-grained layers relative to the lacustrine clays in other units. Furthermore, the various

Al/Si ratios through the core indicate that the Si pattern is also not governed by the

terrigenous aluminosilicate constituents (Fig. 6-5; Richer et al., 2006), demonstrating a

twofold, mixed provenance for Si: (i) quartz in aeolian sands, and (ii) fine-grained terrigenous

aluminosilicates.

115

Fig. 6-5. Bivariate plots of major and trace elements in peaks areas (y-axis) against Al/Si (x-

axis) according to their stratigraphic units. Al/Si ratios indicate generally the grain-size

compositions of the bulk sediments, which display lower values/spikes according to the

coarse-grained sand layers (Fig. 6-3). The biplot of Ca vs. Al/Si seems like a useful tool to

properly discriminate the instinct (i) source rocks changes, and (ii) authigenic productivity in

comparison with the late Pleistocene units. This distinct Holocene samples characteristics are

more clearly elaborated in the scores plot of Fig. 6-4B.

In this study, K displays a divergent variance pattern compared with Ca and reveals

convergent characteristics with Si, Al and Fe (Figs. 6-3 and 6-4). The behavior may result

from mechanisms suggested by Millot (1970), namely, in young aqueous sediments, Ca is

preferentially fixed into carbonates while K enters silicates. Furthermore, as illustrated in Fig.

6-3, the most likely host for K in the Orog Nuur sediments is K-feldspar with partial

contributions from micas or chlorite, which may all be associated with allogenic fluvial

inputs, as well as the aeolian processes (Costa et al., 2014).

In core ONW II, hornblende and micas are the potential hosts for Iron (Fig. 6-3). In general,

hornblende is the main constituent of certain mafic source terranes in the catchments (Fig. 6-

1B) and has weak resistance to weathering (Millot, 1970). Consequently, decomposed and

retained hornblende together with micas all concentrate in fine clay sediments which are

liable to be transported into the Orog Nuur by hydraulic flow and/or mass movement (Nesbitt

116

and Young, 1996) from southern alluvial fans (Fig. 6-1). In general, Fe has two modes of

occurrence, namely the trivalent state and bivalent state when it accompanies Mg or K

(Millot, 1970). As illustrated in Fig. 6-3, Fe acts in a manner analogous to Al and Si. Thus,

rather than primary hornblende, Fe is mainly found in the micas and chlorite in the clayey

sediments, which seems to be the same case with Ti, Rb and Co (Fig. 6-3). Due to the short

burial time and resulting weak diagenetic influence, Fe may not be an indicator of diagenetic

processes in the lacustrine realm.

Furthermore, several other elements share similar patterns as that of the above elements, they

illustrate insignificant loadings for Factor 4. (1) Zirconium: there is ample evidence that Zr is

relatively stable and is almost exclusively mechanically transported and fractionated

(Pettijohn et al., 1973). Mukherji (1970) determined that in an acid aqueous media, Zr may

also exist in polymeric species, e.g., zirconyl halides in the case of higher Cl anion abundance

(Fig. 6-3). However, in terms of alkaline environments with higher concentrations of calcite

or carbonate (higher abundance of [HCO3−]+2[CO3

2−]; Chilingar et al., 1979; Dapples, 1979;),

zirconyl halides seem not to be the host for Zr. Here, the downcore Zr variance reveals no

association between sand fractions and Zr abundance. However, it displays certain links with

the quartz, plagioclase and K-feldspar abundance throughout the core, implying that Zr may

be correlated to the precursor lithotype-governed allogenic fluvial clastic input, but is not

necessarily a faithful recorder of aeolian input (Sinha et al., 2006). (2) Manganese: in spite of

the two sand layers (Units II and IV), the lacustrine phase throughout the core depicts quasi-

constant Mn abundance. In general, insoluble oxi-hydroxides (Mn (III) and Mn (IV) ion)

occur in oxygenated environments while soluble Mn (II) occurs in anaerobic environments

(Calvert and Pedersen, 2007). Thus, in terms of higher lake levels with suboxic to anoxic

water bottoms, more soluble Mn will be readily incorporated into the carbonate lattice and

thereby share variance patterns similar to Ca or CaCO3, which seems largely untenable in this

study (Fig. 6-3). The Orog Nuur therefore represents a frequently riverine inflow-disturbed

shallow hydrological environment, rather than a stagnant water body. (3) Cobalt: the element

Co behaves similarly to Mn and is generally enriched in clayey deposits (Costa et al., 2014).

Both Mn and Co are therefore depleted in sand layers (Fig. 6-3). A slightly lower value in

Unit V can be observed with respect to Co, albeit the broad pattern remains analogous to Mn.

This lowering may be ascribed to the sensitive response of the elements to the slightly

decreasing mica abundance (Fig. 6-3).

117

In summary, Factor 4, represented by Al, Si, K, Ti, and Fe, indicates the allochthonous input

into the lacustrine system. This terrigenous input implication is consistent with several other

works, including Kylander et al. (2011) and Davies et al. (2015).

6.5.3. Biplots of elements and scores plot as indicators of a distinct provenance signal

and high authigenic productivity of the lake system in the Holocene

There is broad consensus that rare earth elements (REE), which are rarely influenced by

diagenesis and metamorphic processes, are common indicators for provenance change

(Pettijohn et al., 1973). Here, in terms of the dominating mudstone, ratios between major and

trace elements were computed in an attempt to examine the possible lithotype signatures (Fig.

6-5). As postulated by Pettijohn et al. (1973), the Al/Si ratio is the commonly applied value in

discriminating the provenance of sandstone and mudstones. In general, higher Al/Si denotes

more clayey or detrital aluminosilicates while the quartz-rich sandstones have low Al/Si ratios

due to the virtual absence of clay-bearing silicates (Davies et al., 2015).

Distinct assemblages related to the lithostratigraphic units are discriminated in the biplots

against Al/Si (Fig. 6-5). Three main groups have been recognized: (i) Fe (K with similar

pattern) abundances are broadly altered by silicate contents in authigenic mud and allogenic

fluvial clastics brought by the Tuyn Gol. Due to the lack of a clay fraction, the Last Glacial

Maximum (LGM) sediments (Unit II) are distinctly distributed with low Fe and Al/Si values,

whereas the mud dominant sediments from Units I and III occupy the opposite end of the

biplot. Furthermore, apart from the lacustrine sediments in this study, the quasi-linear

regression (r=0.66-0.97; R2=0.43-0.97) pattern between Al/Si and Fe was also delineated in

the sandstones and argillites, wherein both depositional conditions and provenance

fingerprints might have been incorporated (Roser and Korsch, 1988); (ii) Mn (Zr and Co with

similar patterns), similarly, is also governed by clay and silt fractions. However, through

sand-dominated sediments may confound the pattern of Mn (Fig. 6-5), data points of the

LGM sediments (Unit II) are more scattered compared with those of Fe. The entire set of data

points is therefore less linearly correlated (r=0.18-0.70; R2=0.03-0.57) relative to the first

group of elements. In addition, as stated above, the whole suite of samples in the biplot of Zr

vs. Al/Si present a slightly more horizontal and scattered regression (r=0.36-0.61; R2=0.13-

0.38), implying that Zr is hosted in both clay enriched and depleted sediments, which might

have complicated the biplot pattern; (iii) Ca appears to be the distinct element that allows us

to discriminate between late Pleistocene and Holocene signatures. Before Termination I, the

sediments exhibit a virtually linear regression (r=0.06-0.97; R2=0.00-0.95) in a manner

118

analogous to Fe, suggesting a quasi-consistent alteration kinetics governed by the clay and silt

fraction input (Fig. 6-5). Nevertheless, although they share similar granulometric

compositions, the Holocene sediments in Unit V exhibit distinctly higher Ca values relative to

the granulometrically similar Units I and III (r=-0.42; R2=0.17) (Fig. 6-5). On the other hand,

the LGM and Younger Dryas sand layers (Units II and IV) represent distinctly different

patterns, namely the Younger Dryas sediments are more scattered relative to LGM sediments

regarding the Al/Si axis. Unlike the dominantly playa condition during the LGM, the Younger

Dryas may have been subject to frequent riverine inflows, resulting in a broader covering of

mud-, silt- to sandstone suites (Fig. 6-2). As a consequence, compared with those in Unit II,

the set of Unit IV data points is more scattered in Fig. 6-5.

6.5.3.1. The disparate provenance signatures and authigenic productivity of the

Holocene sediments

The set of late Pleistocene sediments exhibits coherent regression patterns in the biplots (Fig.

6-5), indicating insignificantly alteration of source lithotypes through Units I-IV. This quasi-

constant provenance delivery/authigenic productivity through Unit I to IV is also documented

in the broadly unbiased quartz, plagioclase, hornblende and minor biased Rb/Sr ratios

(catchment weathering) throughout the late Pleistocene (Fig. 6-3). Apart from Ca, all of the

above three groups of elements display a tendency of quasi-linear regression which is altered

by the amount of clay and silty-clay proportions, albeit Zr is slightly biased and scattered due

to contributions from both finer and coarse sand fractions. The constant delivery of riverine

inflows during the lake phase may have transported the felsic and weathered intermediate or

mafic plutonic rocks into the Orog Nuur (Fig. 6-1B), although this phase was interrupted by

two desiccated playa periods (the local LGM and Younger Dryas) with a sporadic or cessation

of riverine inflows. Alternatively, in this hyperarid environment, airborne material is a

permanent, vital input for the catchment depocenter. The minor variance of quartz and

plagioclase content implies that the aeolian activity that conveyed coarse-grained sediments

into the Orog Nuur catchment basin was quasi-constant through the late Quaternary.

Likewise, as demonstrated by robust multivariate statistics, in the context of the Eijina Basin,

there is no sharp variation of aeolian sands input through the late Quaternary (Yu et al., 2016),

albeit the LGM might have been subject to a slightly higher aeolian deflation in the arid

central Asia (Nilson and Lehmkuhl, 2001). This amount of airborne material was thus

incorporated into the sediment trap at a quasi-constant rate, in spite of the lacustrine setting,

119

playa, or other land surfaces. Thus, the sharp increase of Ca since the very beginning of the

Holocene seems not to have been caused solely by the quansi-constant aeolian activity.

From the onset of the Unit V, both hornblende and CaCO3 demonstrate a coeval abrupt

increase. This phenomenon is also documented in calcite and Ca (peak areas) (Fig. 6-3). The

distinct behavior of Ca in the Holocene sediments is also properly depicted in the biplot

against Al/Si (Fig. 6-5), as well as the scores plot revealed by robust multivariate statistics

(Fig. 6-4B). All of the evidence corroborates the theory that the past environment is not the

sole governing factor exerting influence on the sediment bulk-composition, and that

prominently different source lithotypes and/or higher authigenic productivity may have

existed before and after Termination I. This different provenance may be attributed to twofold

pathways given that the catchment basin is intimately linked to the higher catchment via the

Tuyn Gol: (i) in the stadial (Younger Dryas event) of the late glacial period, the glacial-

pluvial debris accumulated in the higher Khangai Mountains may not have been subject to

abrasion and transport into the Orog Nuur. Subsequently, immediately after Termination I,

due to climate amelioration-induced deglaciation and the hydrodynamically strengthened

riverine inflows, abrupt increases of clastic minerals were available for transport into the Orog

Nuur (Fig. 6-1B), resulting in a sharp increase of the Ca, calcite, and hornblende abundance in

the Holocene, as well as the distinct biplot pattern of Ca vs. Al/Si (Figs. 6-3 and 6-5).

Furthermore, this increased weathering and fluvial input is corroborated by the negative

values of Factor 4 (high terrigenous input) of the Holocene samples in the scores plot (Fig. 6-

4B) and the decreased Rb/Sr ratios of the Holocene samples (Fig. 6-3; Davies et al., 2015).

(ii) Due to the temperature amelioration and better moisture conditions, markedly intensified

chemical weathering (decrease of Rb/Sr) and denudation (increased mica, chlorite and

hornblende contents) occurred in the upper catchment of Orog Nuur, where the source

terranes were broadly occupied by felsic/alkaline and patchy mafic plutonic rocks (Fig. 6-1B

and 6-3). In addition, as suggested by Millot (1970), when hypersaline brines reach saturation,

alkaline earth carbonates will be the first precipitates followed by calcite and aragonite. Thus,

an immense amount of alkaline carbonates and subsequent authigenic calcite have been

generated from the felsic and alkaline precursors, causing a distinct alkaline environment

(higher authigenic productivity) in the Holocene. Thus, only the biplot of Ca vs. Al/Si has

rendered a sound and distinct picture of conditions of the Holocene relative to late Pleistocene

sediments.

120

Fig. 6-6. Schematic model of the integrated past environmental change, land surface

processes and provenance signals (carbonate geochemistry) of the Orog Nuur catchment over

the last ~50 ka. The eastern parts of Khangai Mountains (higher reach of the Orog Nuur

catchment) with altitudes between 3,500 m and 3,700 m asl. have no modern glaciers. More

details regarding glacial and vegetation developments in the catchment are given in

Lehmkuhl and Lang (2001), Rother et al. (2014), and Lehmkuhl et al. (2016). Distinct

geochemical signals for each phase also refer to the table in supplementary material of this

paper.

In summary, influences from the integrated past environment (weathering and land surface

processes) and provenance signals cannot be separately discerned in the arid context. They

more likely represent inherently intercorrelated mechanisms regarding the bulk-geochemistry

of the lacustrine sediments (Fig. 6-6).

121

6.5.4. Summary of coupled paleoenvironmental, weathering, catchment land surface

processes, and provenance signals over the past ~50 ka BP in the Orog Nuur catchment

The overall knowledge gained above has allowed gleaning of the hydro-geochemical

implications of the Orog Nuur water body, as well as the reconstruction of

paleoenvironmental conditions and provenance signals in the drainage catchment (Fig. 6-6).

6.5.4.1. Unit I (~50-~19 cal ka BP): shallow to lentic water body, moderate catchment

weathering

Ample proxy and modelling evidence illustrates that it was generally humid in arid

northwestern China over MIS 3 (Herzschuh, 2006; Yu et al., 2007). It has been suggested that

this may have been caused either by the strengthening of the Asian moisture-laden monsoon

or by a strengthened moisture supply by the westerlies (Yang and Scuderi, 2010). This

enhanced moisture supply is documented in the Orog Nuur sediment core by high amounts of

clay and silt, and is also indicated by a high abundance of the minerals mica and chlorite that

are fluvially transported to the lake. Factor 4 elements, plagioclase, and K-feldspar retain

generally high levels in Unit I, indicating allogenic fluvial input linked to adjacent source

lithotypes in the catchment (Fig. 6-6). On the other hand, however, several intercalated sand

layers with corresponding spikes in the C/Natomic ratio point to sporadic desiccation of the

water column and abrupt fluvial-induced allogenic terrestrial input. The depths of these sand

layers (playa phases) exhibit broad consistency with the spikes in the elemental variances and

the Al/Si ratio (Fig. 6-3). Apart from the sand layers, most phases in Unit I reveal a slightly

reducing environment as inferred from the Mn abundance (Fig. 6-3). A moderate weathering

rate is demonstrated by the largely higher values of the Rb/Sr ratios throughout Unit I. In

summary, Unit I developed in a humid environment with regular riverine inflows of the Tuyn

Gol. Throughout the time period of Unit I, the water body was generally shallow and

sporadically converted into playa conditions (Fig. 6-6). This generally humid phase in MIS 3

is corroborated by low dust supply in the northwestern Pacific pelagic sequences (Leinen,

1989; Nilson and Lehmkuhl, 2001).

6.5.4.2. Unit II (~19-~18 cal ka BP): playa phase and minor riverine inflow

This unit, related to the global LGM, is represented by a massive sand layer that marks an

abrupt deterioration of the hydro-environment in the catchment (Fig. 6-6). The Orog Nuur

was completely desiccated and changed to a playa environment, which served as an allogenic

122

dust sink with ephemeral alluvial/fluvial deposition. In a modern context, aeolian silts and

mobile dunes were generally accumulated in the (winter monsoon) downwind direction,

namely the southeastern section of the Orog Nuur (Lehmkuhl and Lang, 2001). The Factor 4

elements, mud clay, and fluvial clastic bearing minerals mica, chlorite, plagioclase, and

hornblende all exhibit drastically declining contents due to the virtual absence of argillaceous

sediments (Fig. 6-3). Conversely, allogenic K-feldspar displays an abrupt increase related to

the sharply increased contribution of feldspar-bearing sand. The C/Natomic ratios with broadly

greater values than 20 also suggest a more frequent input of allogenic matter. The slightly

reduced redox condition was broken down, as shown by extremely low concentrations of Mn

(Calvert and Pedersen, 2007; Wirth et al., 2013). The in situ weathered materials from the

higher catchment in the Khangai Mountains were preserved locally due to reduced fluvial

transport capacity, and thus were seldom transported into the Orog Nuur.

6.5.4.3. Unit III (~18-~13 cal ka BP): shallow saline lake, moderate weathering in higher

catchment

This period corresponds with the commencement of the last deglaciation in the southern

Mongolian Plateau. Climate amelioration after the LGM led to the melting of glaciers (Rother

et al., 2014) in the higher Khangai Mountains and resulted in a significantly raised amount of

riverine inflows into the Orog Nuur (Fig. 6-6; C/Natomic exhibits slightly higher values relative

to Unit I). The water body was instantly recovered to a state that is comparable with

hydrological conditions prior to the LGM. The relatively stagnant environment is also

demonstrated by high Mn abundance and low C/Natomic ratios (Fig. 6-3). The Factor 4

elements Si, Al, and Fe exhibit similar abundances relative to those in Unit I, suggesting

analogous hydro-geochemical and moderately higher catchment weathering conditions (Figs.

6-4B and 6-6). A slightly increased Cl abundance implies raised salinity of the water body in

contrast to Unit I (Sinha et al., 2006). In contrast, Factor 2 proxies Ca, calcite, and CaCO3 are

low compared to the older late Pleistocene units, indicating a broadly unchanged authigenic

condition within the lake system. Furthermore, similar Rb/Sr ratios relative to those in Unit I

also reflect a largely unchanged weathering intensity in the catchment.

6.5.4.4. Unit IV (~13-~11 cal ka BP): playa phase with slightly more riverine inflows

relative to LGM

This unit again, as in the LGM, is represented by a sand layer (Fig. 6-4B). It corresponds to a

climate deterioration episode associated with the late glacial Younger Dryas event. Although

123

the Orog Nuur system did not completely turn into a playa system as during the LGM (Unit

II; Fig. 6-4B), it experienced repeated desiccations giving rise to the deposition of coarse

sands brought in by fluvial and aeolian processes (Fig. 6-6). The sharp decrease in clay and

silt-clay is displayed by exceedingly low abundances of Fe, Mn and Co, whereas marked

fluctuations in C/Natomic ratios hint at the frequent occurrence of allogenic fluvial inputs into

the playa system (Fig. 6-3). In addition, the (i) sand layer, (ii) reduced run-off input, and (iii)

varved sediments altered to unvarved strata related to the Younger Dryas are also documented

in the nearby lacustrine sequence of Ulaan Nuur in the Valley of Gobi Lakes (Fig. 6-1A; Lee

et al., 2011), in the Lake Tana in northern Ethiopia (Lamb et al., 2007), and in the Lake

Mondsee in Europe (Lauterbach et al., 2011), respectively. These features presumably

characterize a benchmark lithofacies in the lacustrine sequences in the Gobi Desert of

Mongolia or even on a larger northern hemispheric scale. In particular, this YD event-related

Ca depletion with sharp boundaries at c. 11,500 and 12,500 cal a BP is also clearly recorded

in the varved lacustrine core from Lake Mondsee (Luterbach et al., 2011), albeit the

corresponding archive is not a sand layer as in this study in the desert realm.

6.5.4.5. Unit V (<~11 cal ka BP): lentic hypersaline lake with abruptly increased

terrigenous inputs and authigenic productivity

An abrupt recovery of lacustrine conditions marks the termination of the late glacial stage as a

chronological benchmark in the Orog Nuur catchment. This rapid increase of Ca abundance

linked to the ameliorated Holocene climate is also recorded in Lake Mondsee in the

northeastern Alps (Lauterbach et al., 2011). The granulometric composition is akin to that of

Units I and III with slightly higher clay and silt concentrations (Fig. 6-2; Al/Si ratio in Fig. 6-

3). The most remarkable feature of the Holocene phase is the amount of hornblende and

calcite corresponding to high Ca and CaCO3 concentrations that are significantly greater

relative to late Pleistocene units (Fig. 6-5). This feature suggests a considerably different

hydro-geochemical habitat associated with paleoclimatic amelioration. The characteristic

presence of carbonate indicates a predominant alkaline environment (higher Cl content) with

abundant moisture supply and raised authigenic productivity (Wünnemann et al., 2010). This

postulation is reinforced by the distinct assemblage of Ca in the biplot against Al/Si (Figs. 6-

4B and 6-5; Norman and De Deckker, 1990). Another reason for its occurrence, however, is

that the calcite may also result from its recrystallization from detrital precursors (neogenic

calcite; Pettijohn et al., 1973). In this case, more eroded clastic materials from the northern

felsic and alkaline source terranes were brought into the Orog Nuur (Figs. 6-1B and 6-6).

124

However, in terms of late Quaternary terrestrial (lacustrine) sediments, this diagenesis

(recrystallization) may only have exerted limited influence on the bulk-geochemistry relative

to the climate-induced calcite precipitation as well as allochthones input (Pettijohn, 1973; Yu

et al., 2016). Furthermore, the significantly increased input of weathering materials is also

demonstrated by the sharp increase of the Rb/Sr ratio (Fig. 6-3). In general, the higher

Holocene lake level is in agreement with that from Lehmkuhl and Lang (2001). In the Baikal

catchment further to the north (Fig. 6-1A), a wealth of palynological signals was assembled

and assessed by Tarasov et al. (2007), who revealed a broadly warm and humid early

Holocene as well as a deteriorated climate since the mid-Holocene. Nonetheless, in this study,

due to the loss of the uppermost 24 cm of sediments and a relatively high uncertainty in the

Holocene age in the Bayesian age model (Fig. 6-2), a feasible interpretation for the late

Holocene signals is difficult to construct. Accordingly, the lentic hypersaline habit proposed

here might be restrained to the early- to mid-Holocene period.

In summary, the distinct Holocene water body is attributed to (i) a more intensively weathered

felsic/alkaline material transported from the higher catchment into the Orog Nuur depocenter,

and (ii) sharply increased authigenic productivity of the alkaline environment thereafter.

6.6. Conclusions

We present a continuous, high-resolution elemental record attained in the lacustrine sequence

of Orog Nuur, southern Mongolia. An integrated paleoenvironmental and provenance history

in the Gobi Desert that trace back to the MIS 3 are outlined for the first time.

To constrain the data quality, elements with high error margins relative to measured peak

areas, and those elements/proxies under a significance level during the robust multivariate

statistics are ruled out for further implications. As determined by robust multivariate statistics,

the bulk-geochemistry of the Orog Nuur core sediments are governed by Factor 1 (grain-size

composition), Factor 2 (authigenic productivity in the saline/evaporitic environment: Ca, Cl,

CaCO3, Factor 3 (allochthonous organic material input: TOC, C/Natomic, and Factor 4 (fluvial

terrigenous input, as well as quasi-constant aeolian input through the late Quaternary period:

Al, Si, K, Ti, Fe, chlorite, and hornblende). As inferred from the biplot of Ca vs. Al/Si,

disparate source lithotypes along with authigenic productivity existed before and after

Termination I. In contrast to the late Pleistocene, the Holocene was dominated by a distinctly

high productivity alkaline environment due to the increased hydrodynamic strength of riverine

inflow and intensified source rock weathering/erosion in the upper catchments of Orog Nuur.

125

In order to gain a better understanding of the bulk-geochemistry of lake sediments, the

coupled provenance and environmental signatures, such as weathering and land surface

processes in the catchment, need to be systematically discerned. Thus, considering the bulk-

geochemistry of lacustrine sediments, paleoclimate would not be the sole governing factor;

the coupled alteration of precursor rocks and accompanying decomposition / denudation

intensity, as well as the transportation process may exert pivotal influences.

In the arid realm of Gobi Desert of Mongolia, two coarse sand layers corresponding to the

LGM and Younger Dryas are present. In the Gobi Desert, these two layers provide a potential

opportunity to act as chronological benchmarks. However, more investigation needs to be

carried out to test the reliability of these beds on larger spatial scales and their synchrony in

time.

Table 6-S1 Synoptic description of the coupled paleoenvironmental and provenance

signatures during the past ~50 ka BP as inferred from the lacustrine sequence of Orog Nuur.

A schematic model comprising these factors is proposed in Fig. 6-6 in the main text.

Lithostrati

graphic

units

Correspondi

ng ages (cal

ka BP)

Water column and

climatic interpretations

Broad source lithotypes and

land surface processes

Distinct bulk-geochemistry

I ~50-~19 shallow to lentic water

body

Weathering of felsic and few

mafic plutonic rocks

High S, Al, and Si; low Cl, Ca,

calcite and hornblende

II ~19-~18 arid playa phase Intensified aeolian sands

deposited in fluvial coarse-

grained sediments

High K-Feldspar, C/N ratio;

low terrestrial elements, mica,

chlorite, and calcite

III ~18-~13 shallow saline lake Felsic and few mafic plutonic

rocks similar like phase I

High S and Co; low Ca, calcite

and hornblende

IV ~13-~11 arid playa phase Intensified aeolian sands

input and seldom fluvial

activity

High C/N ratio; low terrestrial

elements and Ca/CaCO3

V <~11 lentic & hypersaline water

body

alkaline, felsic, and few mafic

plutonic rocks, together with

intensive erosion of bedrocks

High Ca, Cl, calcite, mica,

chlorite and hornblende; low S,

quartz, plagioclase, and K-

feldspar

126

Fig. 6-S1. Dispersion plots of the major and trace element concentrations (peak areas)

according to their stratigraphic units. Numbers of observations for each unit are marked on

the upmost axis. The median values are linked with dash lines, while the upper and lower

quartiles are marked as error bars.

127

7. Synthesis: Paleoclimatic implications from late Quaternary terrestrial archives in the

Gobi Desert

This dissertation presents results from the multidisciplinary study on terrestrial archives

embracing aeolian, fluvial, alluvial and lacustrine sediments in the arid realm, and their

corresponding thermal and moisture implications. To close the loop of this dissertation, main

thermal and moisture history of the Gobi Desert of Mongolia is illustrated in Fig. 7-1. The

core part is summarized as follows, reflecting the significant findings in this research.

Fig. 7-1: Synoptic compilation of hydrologic, vegetation and glacial history of the Orog Nuur

catchment over the last ~50 ka. Playa & dune phase is recovered from the granulometric and

lithostratigraphic characteristics of the core ONW II. Likewise, this late Glacial high lake

levels are also recorded in the Ejina Basin as lacustrine facies.

7.1. Bulk-geochemistry of different terrestrial archives in the arid realm and their

corresponding implications for the paleoenvironmental, pedogenesis and diagenesis

processes

A practical strategy is presented to discriminate sediment archive properties and thereafter

sequence the controlling factors of sedimentary processes. The considered records of northern

Ejina Basin can be discriminated into three broad lithological units: basal aeolian strata,

lacustrine strata and the uppermost poorly-sorted mixture of coarse sand and gravels.

128

Elemental characteristics in conjunction with the granulometric composition in different

sediment archives in the northern Ejina Basin are obtained. Aeolian sands have high contents

of Zr, Hf, Ba and Si. Especially Hf can be regarded as good indicator to discriminate the

coarse sand proportion. Sandy loess has high contents of Ca and Sr which both exhibit broad

correlations with medium to coarse silt proportions. Lacustrine clay has high contents of

felsic, ferromagnesian and mica source elements. Fluvial sands have distinctly high contents

of Mg, Cl and Na which are enriched in evaporate minerals. Alluvial gravels have especially

high contents of Cr which may originate from nearby Cr-rich bedrock.

Multivariate statistical methods applied to geochemical compositions is a useful tool to

discriminate archive types in our case. Considering the suite of late Quaternary sediments in

the arid realm, temporal variation of the bulk-geochemistry can be well explained by four

semi-quantified sediment components: (weak-) weathering intensity, silicate-bearing mineral

abundance, saline / alkaline influence of sediments, and the quasi-constant aeolian input into

the arid endorheic basin system. In contrast, pedogenesis and diagenesis exert only a limited

influence.

7.2. Sedimentological, palynological and ostracod studies of the lacustrine sediments and

their references to the paleohydrological and vegetation history

A multidisciplinary investigation is conducted on a 13.35 m lacustrine core retrieved from

Orog Nuur, in the Gobi Desert of Mongolia. Methods embracing sedimentological,

geochemical, palynological and ostracod studies combined with sets of dated terrace ridges

and moraines were conducted to illustrate the vegetation development, hydrologic and

paleoglaciation history during the last ~ 50 ka. For the first time, in the Gobi Desert, semi-

quantified moisture and thermal sequences that track back to the MIS 3 are presented.

In Gobi Desert of Mongolia, the MIS 3 experienced sufficient moisture supply in particular

between ~36 ka and ~24 ka; MIS 2 was subjected to cool temperature and moisture-deficient,

which was interrupted by two exceedingly cold and dry playa phases related to the LGM and

YD event; The early Holocene exhibited ameliorated climatic condition characterized by

higher lake levels. In addition, a sharp transition of Termination I is illuminated by

palynological and geochemical imprints. Lower area of the Orog Nuur catchment was

dominated by Artemisia steppe community in the late Pleistocene and altered to

Chenopodiaceae desert steppe in the Holocene. Water body in the Holocene appears to be a

129

distinct alkaline environment which was subjected to frequently allogenic input and

disturbance of the late Pleistocene anoxic states.

The main humid pulse in MIS 3 might have been the main driving mechanism for the MIS 3

glacial advance, while the pronounced low temperature in the higher elevated catchment in

the eastern Khangai Mountains is presumably the forcing factor for the LGM glacial advance.

The early Holocene in the Gobi Desert experienced the other humid pulse, which was

concurrent with the lacustrine condition and the degradation of the permafrost features as well

as the glacial retreat. On the other hand, four arid playa phases (~47 ka, ~37 ka, ~19 ka and

~13-~11 ka) were determined, of which the latter two sand layers (i.e., the LGM, YD event,

respectively) recorded in the Orog Nuur may provide an opportunity to be labelled as

chronological benchmarks for other lacustrine sequences in the Gobi Desert of Mongolia.

Nonetheless, further investigations still need to be carried out to test its reliability in a larger

spatial scale.

In the central to southern Mongolia, the late Quaternary moisture and thermal history may be

modulated, if not controlled, by coupled atmospheric components embracing both Westerlies

and the propagation of the East Asian Summer Monsoon into the Asian interior, albeit more

details with respect to the quantified contribution from these systems remain an open

question.

7.3. Geochemical imprints of the lacustrine sediments and integrated

paleoenvironmental and provenance implications

Based on a continuous, high-resolution elemental records attained in the lacustrine sequence

of Orog Nuur, Gobi Desert of Mongolia, the coupled paleoenvironmental and provenance

history are outlined. Due to the predominant sediments of mud- to siltstone, Al and Si display

broadly identical pattern in the lake sediments. Ca behaviors may be ascribed to the

authigenic calcite abundance. As equally applicable to pelagic sediments, Mn and Co may act

as indicator for the redox condition in lacustrine realm. On the other hand, owing to the short

burial time and weak diagenetic influence, Fe is not an indicator to the diagenetic processes.

Likewise, Zr may be associated to fluvial clastic in relation to the lithotypes but not

necessarily linked to the aeolian sands concentration as in the pelagic realm. Furthermore, S

in lake sediments may denote the redox condition and K is more likely linked to the K-

feldspar which is associated with the allogenic fluvial inputs.

130

The two sand laminations corresponding to the LGM and YD are dominated by coarse sands.

In the Gobi Desert of southern Mongolia, these two laminations provide a potential

opportunity to be regarded as chronological benchmarks. However, more investigation needs

to be conducted to test its reliability in larger spatial scales.

As inferred from the biplot between Ca and Al/Si, disparate source lithotypes may exist

before and after the Termination I. The Holocene appears to be a distinct alkaline

environment relative to the late Pleistocene. This may be ascribed to either strengthened

hydrodynamic strength of the riverine inflows and / or intensified source rocks erosion in the

upper catchments of Orog Nuur. In order to gain a better understanding of the bulk-

geochemistry of lake sediments, the coupled provenance and environmental signatures need

to be properly discerned. On the other hand, in arid realm as Gobi Desert of Mongolia, sands

may exert a considerable influence on the bulk-geochemistry of lacustrine sediments.

In summary, considering the bulk-geochemistry of lacustrine sediments, paleoclimate would

not be the solely governing factor, the coupled alteration of precursor rocks and

decomposition intensity may exert pivotal influences.

7.4. Outlook

For the upcoming decades, several pivotal but challenging tasks need to be carried out to

better decipher the paleoclimate signals and controlling atmospheric circulations in the arid

central Asia, on perspectives of both terrestrial and pelagic archives. A fundamental step is to

refine the understandings concerning the proxies and their corresponding paleoenvironmental

implications. The tasks are assembled as follows:

(1) As the purposes of the paleoclimatologist are primarily high resolution and longer time

spanning of the repository, it is therefore indispensable to take part in IODP / ECORD

excursion in e.g., NW Pacific, retrieving continuous, high resolution and well constrained

chronological framework pelagic sequences;

(2) Get innovative methods, e.g., neodymium or other isotopes signals incorporated in the

pelagic sediments, to discriminate Westerlies signals from EASM, and build thereby

robust Westerlies Intensity indices. Eventually, compare with our published arid central

Asia data in this dissertation. Thereafter, it is possible to define and decipher the moisture

provenance in the Gobi Desert of Mongolia;

131

(3) Investigate long time scale lacustrine core retrieved in southern Mongolia, from where

great scarcity of paleoenvironmental repository exists. Alternatively, terrestrial drilling

project (e.g., lacustrine, loess-paleosol sequences) could be carried out underneath the

pathway of the mid-latitude Westerlies. Compared to the pelagic sequences,

environmental interpretations concerning those elements such as trace and REE elements

abundance as well as non-conventional isotopes signals in lacustrine realm are less studied

and understood. A time spanning of greater than the late glacial period is in urgent need in

order to gain a comprehensive understanding of the kinetics of central to eastern Asian

paleoclimate patterns;

(4) Gain robust and quantitatively defined indices for thermal and moistures signals.

Promising proxies are geobiology (pollen / ostracod / diatom), geochemistry (elements,

minerals and non-conventional isotopes) and geophysics (paleomagnetic), of which the

paleomagnetic is of special importance because of its application of refinement of the

surface sediments’ age model and thereafter the estimation of reservoir effect from the

radiocarbon determined age model;

(5) In particular, the geologic repository may not illustrate complete consistency with the

numerical simulation results (e.g., Jiang et al., 2012). It is therefore indispensable to

provide high resolution, (semi-) quantified proxy sequence as parameters for the modelers.

In turn, the high resolution simulation results should be taken into account in trying to

better understand the atmospheric patterns and underlying mechanisms of the

paleoclimate changes.

The fulfillment of above points will largely fill the knowledge gaps in the arid central Asia

and therefore benefit the paleoclimatological communities.

132

References

Aitken, M.J., 1998. An Introduction to Optical Dating. Oxford University Press, Oxford, pp.

43-44.

Amorosi, A., Centineo, M.C., Dinelli, E., Lucchini, F. Tateo, F., 2002. Geochemical and

mineralogical variations as indicators of provenance changes in Late Quaternary

deposits of SE Po Plain. Sedimentary Geology 151, 273-292.

An, C.B., Chen, F.H., Barton, L., 2008. Holocene environmental changes in Mongolia: A

review. Global and Planetary Changes 63, 283-289.

An, C.B., Zhao, J.J., Tao, S.C., Lv, Y.B., Dong, W.M., Li, H., Jin, M., Wang, Z.L., 2011.

Dust variation recorded by lacustrine sediments from arid Central Asia since ~15 cal ka

BP and its implication for atmospheric circulation. Quaternary Research 75, 566-573.

An, Z.S., Li, L., Burr, G.S., Song, Y.G., Yan, L.B., Chang, H., Sun, Y.B., Cai, Y.J., Shi, Z.G.,

Xu, H., Zhao, H.L., Zhou, W.J., 2014. Introduction. In: An, Z.S., (Eds.). Late Cenozoic

Climate Change in Asia. Developments in Palaeoenvironmental Research 16, Springer,

Dordrecht, pp. 1-22.

An, Z.S., Porter, S.C., Kutzbach, J.E., Wu, X.H., Wang, S.M., Liu, X.D., Li, X.Q., Zhou,

W.J., 2000. Asynchronous Holocene optimum of the East Asian Monsoon. Quaternary

Science Reviews 19, 743-762.

Anadon, P., Utrilla, R., Julià, R., 1994. Palaeoenvironmental reconstruction of a Pleistocene

lacustrine sequence from faunal assemblages and ostracode shell geochemistry, Baza

Basin, SE Spain. Palaeogeography, Palaeoclimatology, Palaeoecology 111, 191-205.

Arz, H.W., Lamy, F., Pätzold, J., Müller, P.J., Prins, M., 2003. Mediterranean moisture source

for an Early-Holocene humid period in the Northern Red Sea. Science 300, 118-121.

Baljinnaym, I., Bayasgalan, A., Brorisov, B.A., Cisternas, A., Dem'yanovich, M.G.,

Ganbaatar, L., Kochetkov, V.M., Kurushin, R.A., Molnar, P., Philip, H., Vashchilov,

Y.Y., 1993. Ruptures of major earthquakes and active deformation in Mongolia and its

surroundings. Geology Society of America Memoirs 181, 26-52.

Barry, R.G., Chorley, R.J., 2009. Atmosphere, Weather and Climate. Routledge, Abingdon,

pp. 127-161.

Bayasgalan, A., Jackson, J., Ritz, J.-F., Carretier, S., 1999. “Forebergs”, flower structures, and

the development of large intra continental strike-slip faults: the Gurvan Bogd fault

system in Mongolia. Journal of Structural Geology 21, 1285-1302.

Becken, M., Hölz, S., Fiedler-Volmer, R., Hartmann, K., Wünnemann, B., Burkhardt, H.,

2007. Electrical resistivity image of the Jingsutu Graben at the NE margin of the Ejina

Basin (NW China) and implications for the basin development. Geophysical Research

Letters 34, doi: 10.1029/2007GL029412.

Benn, D.I., Owen, L.A., 1998. The role of the Indian Summer Monsoon and the mid-latitude

weterlies in Himalayan glaciation: review and speculative discussion. Journal of the

Geological Society, London 155, 353-363.

133

Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 million years.

Quaternary Science Reviews 10, 297-317.

Bertrand, S., Hughen, K., Giosan, L., 2015. Limited Influence of Sediment Grain Size on

Elemental XRF Core Scanner Measurements. In: Croudace, I.W., Guy Rothwell, R.,

(eds) Micro-XRF Studies of Sediment Cores. Developments in Paleoenvironmental

Research 17, pp 473-490.

Bhatia, M.R., 1983. Plate tectonics and geochemical composition of sandstones. The Journal

of Geology 91, 611-627.

Blott, S.J., Pye, K., 2012. Particle size scales and classification of sediment types based on

particle size distributions: Review and recommended procedures. Sedimentology 59,

2071-2096.

Bond, G., Broecker, W., Johnsen, S., McManus, J., Labeie, L., Jouzel, J., Bonani, G., 1993.

Correlations between climate records from North Atlantic sediments and Greenland ice.

Nature 365, 143-147.

Bøtter-Jensen, L., Thomsen, K.J., Jain, M., 2010. Review of optically stimulated

luminescence (OSL) instrumental developments for retrospective dosimetry. Radiation

Measurements 45, 253-257.

Boyle, J.F., 2001. Inorganic geochemical methods in palaeolimnology. In: Last, W.M., Smol,

J.P., (Eds.). Tracking Environmental Change Using Lake Sediments Volume 2: Physical

and Geochemical Methods. Developments in Paleoenvironmental Research 2, pp. 83-

130.

Boyle, E.A., Keigwin, L., 1987. North Atlantic thermohaline circulation during the past

20,000 years linked to high-latitude surface temperature. Nature 330, 35-40.

Bridge, J.S., Demicco, R.V., 2008. Earth surface processes, landforms and sediment deposits.

Cambridge: Cambridge University Press, p 61.

Broccoli, A.J., Dahl, K.A., Stouffer, R.J., 2006. Response of the ITCZ to Northern

Hemisphere cooling. Geophysical Research Letters 33, L01702, doi:

10.1029/2005/2005GL024546.

Broecker, W.S., 1994. Massive iceberg discharges as triggers for global climate change.

Nature 372, 421-424.

Brooks, S.J., 2006, Fossil midges (Diptera: Chironomidae) as palaeoclimatic indicators for the

Eurasian region. Quaternary Science Reviews 25, 1894-1910.

Buggle, B., Glaser, B., Zöller, L., Hambach, U., Marković, S., Glaser, I., Gerasimenko, N.,

2008. Geochemical characterization and origin of Southeastern and Eastern European

loess (Serbia, Romania, Ukraine). Quaternary Science Reviews 27, 1058-1075.

Bush, A.B.G., 2004. Modelling of Late Quaternary climate over Asia: a synthesis. Boreas 33,

155-163.

Calvert, S.E., Pedersen, T.F., 2007. Elemental proxies for palaeoclimatic and

palaeoceanographic variability in marine sediments: Interpretation and application. In:

134

Hillaire-Marcel, C., De Vernal., A., (Eds.). Proxies in late Cenozoic paleoceanography.

Developments in Marine Geology 1, 567-625.

Chen, C.-T. A., Lan, H.-C., Lou, J.-Y., Chen, Y.-C., 2003. The dry Holocene megathermal in

Inner Mongolia. Palaeogeography, Palaeoclimatology, Palaeoecology 193, 181-200.

Chen, F.H., Dong, G.H., Zhang, D.J., Liu, X.Y., Jia, X., An, C.B., Ma, M.M., Xie, Y.W.,

Barton, L., Ren, X.Y., Zhao, Z.J., Wu, X.H., Jones, M.K., 2015. Agriculture facilitated

permanent human occupation of the Tibetan Plateau after 3600 B.P. Science 347, 248-

250.

Chen, F.H., Fan, Y.X., Chun, X., Madsen, D.B., Oviatt, C.G., Zhao, H., Yang, L.P., Sun, Y.,

2008a. Preliminary research on Megalake Jilantai-Hetao in the arid areas of China

during the Late Quaternary. Chinese Science Bulletin 53, 1725-1739.

Chen, F.H., Yu, Z.C., Yang, M.L., Ito, E., Wang, S.M., Madsen, D.B., Huang, X.Z., Zhao, Y.,

Sato, T., Birks, H.J.B., Bommer, I., Chen, J.H., An, C.B., Wünnemann, B., 2008b.

Holocene moisture evolution in arid central Asia and its out-of phase relationship with

Asian monsoon history. Quaternary Science Reviews 27, 351-364.

Chen, H.F., Song, S.R., Lee, T.Q., Löwemark, L., Chi, Z.Q., Wang, Y., Hong, E., 2010. A

multiproxy lake record from Inner Mongolia displays a late Holocene teleconnection

between Central Asian and North Atlantic climates. Quaternary International 227, 170-

182.

Chen, J., Li, G.J., 2011. Geochemical studies on the source region of Asian dust. Science

China Earth Sciences 54, 1279-1301.

Chen, J.H., 2009. Moisture variability over the last millennium recorded by chironomids from

lacustrine sediments in arid northwestern China. Dissertation to Lanzhou University.

Cheng, H., Spötl, C., Breitenbach, S.F.M., Sinha, A., Wassenburg, J.A., Jochum, K.P.,

Scholz, D., Li, X.L., Yi, L., Peng, Y.B., Lv, Y.B., Zhang, P.Z., Votintseva, A., Loginov,

V., Ning, Y.F., Kathayat, G., Edwards, R.L., 2016. Climate variations of Central Asia

on orbital to millennial timescales. Scientific Reports 6, 36975, doi: 10.1038/srep36975.

Chiang, J.C.H., Fung, I.Y., Wu, C.-H., Cai, Y.J., Edman, J.P., Liu, Y.W., Day J.A.,

Bhattacharya, T., Mondal, Y., Labrousse, C.A., 2015. Role of seasonal transitions and

westerly jets in East Asian paleoclimate. Quaternary Science Reviews 108, 111-129.

Chilingar, G.V., Zenger, D.H., Bissell, H.J., Wolf, K.H., 1979. Dolomites and dolomitization.

In: Larsen, G., Chilingar, G.V., (Eds.). Diagenesis in Sediments and Sedimentary

Rocks. Developments in Sedimentology 25A, pp. 428-536.

Clark, P.U., Shakun, J.D., Baker, P.A., Bartlein, P.J., Brewer, S., Brook, E., Carlson, A.E.,

Cheng, H., Kaufman, D.S., Liu, Z.Y., Marchitto, T.M., Mix, A.C., Morrill, C., Otto-

Bliesner, B.L., Pahnke, K., Russell, J.M., Whitlock, C., Adkins, J.F., Blois, J.L., Clark,

J., Colman, S.M., Curry, W.B., Flower, B.P., He, F., Johnson, T.C., Lynch-Stieglitz, J.,

Markgraf, V., McManus, J., Mitrovica, J.X., Moreno, P.I., Williams, J.W., 2012. Global

climate evolution during the last deglaciation. Proceedings of the National Academy of

Sciences 109, doi: 10.1073/pnas.1116619109.

http://www.pnas.org/search?author1=Peter+U.+Clark&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Jeremy+D.+Shakun&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Paul+A.+Baker&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Patrick+J.+Bartlein&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Simon+Brewer&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Ed+Brook&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Anders+E.+Carlson&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Hai+Cheng&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Darrell+S.+Kaufman&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Zhengyu+Liu&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Thomas+M.+Marchitto&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Alan+C.+Mix&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Carrie+Morrill&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Bette+L.+Otto-Bliesner&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Bette+L.+Otto-Bliesner&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Katharina+Pahnke&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=James+M.+Russell&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Cathy+Whitlock&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Jess+F.+Adkins&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Jessica+L.+Blois&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Jorie+Clark&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Steven+M.+Colman&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=William+B.+Curry&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Ben+P.+Flower&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Feng+He&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Thomas+C.+Johnson&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Jean+Lynch-Stieglitz&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Vera+Markgraf&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Jerry+McManus&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Jerry+X.+Mitrovica&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=Patricio+I.+Moreno&sortspec=date&submit=Submit

http://www.pnas.org/search?author1=John+W.+Williams&sortspec=date&submit=Submit

135

Colman, S.M., Yu, S.Y., An, Z.S., Shen, J., Henderson, A.C.G., 2007. Late Cenozoic climate

changes in China’s western interior: a review of research on Lake Qinghai and

comparison with other records. Quaternary Science Reviews 26, 2281-2300.

Conroy, J.L., Restrepo, A., Overpeck, J.T., Steinitz-Kannan, M., Cole, J.E., Bush, M.B.,

Colinvaux, P.A., 2009. Unprecedented recent warming of surface temperatures in the

eastern tropical Pacific Ocean. Nature Geoscience 2, 46-50.

Costa, K.M., Russel, J.M., Vogel, H., Bijaksana, S., 2014. Hydrological connectivity and

mixing of Lake Towuti, Indonesia in response to paleoclimatic changes over the last

60,000 years. Palaeogeography Palaeoclimatology Palaeoecology 417, 467-475.

Cunningham,W.D., 2005. Active intracontinental transpressional mountain building in the

Mongolian Altai: Defining a new class of orogeny. Earth and Planetary Science Letters

240, 436-444.

Cunningham, W.D., 2013. Mountain building processes in intracontinental oblique

deformation belts: Lessons from the Gobi Corridor, Central Asia. Journal of Structural

Geology 46, 255-282.

Cunningham, W.D., Windley B.F., Dorjnamjaa, D., Badamgarov, J., Saandar, M., 1996. Late

Cenozoic transpression in southwestern Mongolia and the Gobi Altai-Tien Shan

connection. Earth and Planetary Science Letters 140, 67-81.

Cunningham, W.D., Windley, B.F., Owen, L.A., Barry, T., Dorjnamjaa, D., Badamagarav, J.,

1997. Geometry and style of partitioned deformation within a late Cenozoic

transpressional zone in the eastern Gobi Altai Mountains, Mongolia. Tectonophysics

277, 285-306.

Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrup, N.S., Hammer,

C.U., Hvldberg, C.S., Steffensen, J.P., Sveinbjörnsdottir, A.E., Jouzel, J., Bond, G.,

1993. Evidence for general instability of past climate from a 250-k ice-core record.

Nature 364, 218-220.

Dansgaard, W., White, J.W.C., Johnsen, S.J., 1989. The abrupt termination of the Younger

Dryas climate event. Nature 339, 532-534.

Dapples, E.C., 1979. Silica as an agent in diagenesis. In: Larsen, G., Chilingar, G.V., (Eds.).

Diagenesis in Sediments and Sedimentary Rocks. Developments in Sedimentology 25A,

pp. 99-142.

Davies, S.J., Lamb, H.F., Roberts, S.J., 2015. Micro-XRF core scanning in palaeolimnology:

recent developments. In: Croudace, I.W., Guy Rothwell, R., (Eds.). Micro-XRF Studies

of Sediment Cores. Developments in Paleoenvironmental Research 17, pp. 189-226.

Demske, D., Heumann, G., Granoszewski, W., Nita, M., Mamakowa, K., Tarasov, P.E.,

Oberhänsli, H., 2005. Late glacial and Holocene vegetation and regional climate

variability evidenced in high-resolution pollen records from Lake Baikal. Global and

Planetary Change 46, 255-279.

Derbyshire, E., Meng, X.M., Kemp, R.A., 1998. Provenance, transport and characteristics of

modern aeolian dust in western Gansu Province, China, and interpretation of the

Quaternary loess record. Journal of Arid Environments 39, 497-516.

136

Dietze, E., Maussion, F., Ahlborn, M., Diekmann, B., Hartmann, K., Henkel, K., Kasper, T.,

Lockot, G., Opitz, S., Haberzettl, T., 2014. Sediment transport processes across the

Tibetan Plateau inferred from robust grain-size end members in lake sediments. Climate

of the Past 10, 91-106.

Ding, Z.L., Liu, T.S., Rutter, N.W., Yu, Z.W., Guo, Z.T., 1995. Ice-volume forcing of East

Asian Winter Monsoon variations in the past 800,000 years. Quaternary Research 44,

149-159.

Divis, A.F., Anne McKenzie, J., 1975. Experimental authigenesis of phyllosilicates from

feldspathic sands. Sedimentology 22, 147-155.

Doberschütz, S., Frenzel, P., Haberzettl, T., Kasper, T., Wang, J.B., Zhu, L.P., Daut, G.,

Schwalb, A., Mäusbacher, R., 2013. Monsoonal forcing of Holocene

paleoenvironmental change on the central Tibetan Plateau using a sediment record from

Lake Nam Co (Xizang, China). Journal of Paleolimnology 51, 253-266.

Dorofeyuk, N.I., Tarasov, P.E., 1998. Vegetation and lake levels in Northern Mongolia in last

12,500 years as indicated by data of pollen and diatom analyses. Stratigraphy and

Geological Correlation 6, 70-83.

Duce, R.A., Unni, C.K., Ray, B.J., Prospero, J.M., Merrill, J.T., 1980. Long-range

atmospheric transport of soil dust from Asia to the tropical North Pacific: Temporal

variability. Science 209, 1522-1524.

Dulski, P., Brauer, A., Mangili, C., 2015. Combined µ-XRF and microfacies techniques for

lake sediment analyses. In: Croudace, I.W., Guy Rothwell, R., (eds) Micro-XRF Studies

of Sediment Cores. Developments in Paleoenvironmental Research 17, pp 325-349.

Egozcue, J.J., Pawlowsky-Clahn, V., Mateu-Figueras, G., Barceló-Vidal, C., 2003. Isometric

logratio transformations for compositional data analysis. Mathematical Geosciences 35,

279-230.

Eichler, A., Olivier, S., Henderson, K., Laube, A., Beer, J., Papina, T., Gäggeler, H.W.,

Schwikowski, M., 2009. Temperature response in the Altai region lags solar forcing.

Geophysical Research Letters 36, L01808, doi: 10.1029/2008GL035930.

Erdtman, G., 1960. The acetolysis method. Svensk Botanisk Tidskrift 54, 564-564.

Fedotov, A.P., Chebykin, E.P., Yu, S.M., Vorobyova, S.S., Yu, O.E., Golobokova, L.P.,

Pogodaeva, T.V., Zheleznyakova, T.O., Grachev, M.A., Tomurhuu, D., Oyunchimeg,

Ts., Narantsetseg, Ts., Tomurtogoo, O., Dolgikh, P.T., Arsenyuk, M.I., De Batist, M.,

2004. Changes in the volume and salinity of Lake Khubsugul (Mongolia) in response to

global climate changes in the upper Pleistocene and the Holocene. Palaeogeography,

Palaeoclimatology, Palaeoecology 209, 245-257.

Fedotov, A.P., Phedorin, M.A., De Batist, M., Ziborova, G.A., Yu, K.A., Yu., S.M.,

Matasova, G.G., Khabuev, A.V., Kugakolov, S.A., Rodyakin, S.V., Krapivina, S.M.,

Pouls, T., 2008. A 450-ka long record of glaciation in Northern Mongolia based on

studies at Lake Khubsugul: high-resolution reflection seismic data and grain-size

variations in cored sediments. Journal of Paleolimnology 39, 335-348.

137

Felauer, T., 2011. Jungquartäre Landschafts und klimagschichte der Südmongolei.

Dissertation zu Rheinishch-Westfälischen Technischen Hochshule Aachen.

Felauer, T., Schlütz, F., Murad, W., Mischke, S., Lehmkuhl, F., 2012. Late Quaternary

climate and landscape evolution in arid Central Asia: A multiproxy study of lake

archive Bayan Tohomin Nuur, Gobi desert, southern Mongolia. Journal of Asian Earth

Science 48, 125-135.

Feng, Z.D., Chen, F.H., Tang, L.Y., Kang, J.C., 1998. East Asian monsoon climates and Gobi

dynamics in marine isotope stages 4 and 3. Catena 33, 29-46.

Feng, Z.D., Wang, W.G., Guo, L.L., Khosbayar, P., Narantsetseg, T., Jull, A.J.T., An, C.B.,

Li, X.Q., Zhang, H.C., Ma, Y.Z., 2005. Lacustrine and eolian records of Holocene

climate changes in the Mongolian Plateau: preliminary results. Quaternary

International 136, 25-32.

Feng, Z.D., Zhai, X.W., Ma, Y.Z., Huang, X.Z., Wang, W.G., Zhang, H.C., Khosbayar, P.,

Narantsetseg, T., Liu, K.-B., Rutter, N.W., 2007. Eolian environmental changes in the

Northern Mongolian Plateau during the past ~35,000. Palaeogeography,


Filzmoser, P., Hron, K., Reimann, C., Garrett, R., 2009. Robust factor analysis for

compositional data. Computers & Geosciences 35, 1854-1861.

Fitzsimmons, K.E., Magee, J.W., Amos, K.J., 2009. Characterization of aeolian sediments

from the Strzelecki and Tirari Deserts, Australia: Implications for reconstructing

palaeoenvironmental conditions. Sedimentary Geology 218, 61-73.

Fleitmann, D., Burns, S.J., Mudelsee, M., Nef, U., Kramers, J., Mangini, A., Matter, A., 2003.

Holocene forcing of the Indian Monsoon recorded in a stalagmite from southern Oman.

Science 300, 1737-1739.

Folk, R.L., 1966. A review of grain-size parameters. Sedimentology 6, 73-93.

Forman, S.L., Pierson, J., Gómez, J., Brigham-Grette, J., Nowaczyk, N.R., Melles, M., 2007.

Luminescence geochronology for sediments from Lake Elgygytgyn, northeast Siberia,

Russia: constraining the timing of paleoenvironmental events for the past 200 ka.

Journal of Paleolimnology 37, 77-88.

Fowell, S.J., Hansen, B.C.S., Peck, J.A., Khosbayer, P., Ganbold, E., 2003. Mid to Late

Holocene climate evolution of the Lake Telmen Basin, north central Mongolia, based on

palynological data. Quaternary Research 59, 353-363.

Fumiko Watanabe, N., Watanabe, T., Kakegawa, T., Minoura, K., Imai, A., Fagel, N.,

Horiuchi, K., Nakamura, T., Kawai, T., 2014. Biological nitrate utilization in south

Siberian lakes (Baikal and Hovsgol) during the Last Glacial period: the influence of

climate change on primary productivity. Quaternary Science Reviews 90, 69-79.

Fralick, P.W., Kronberg, B.I., 1997. Geochemical discrimination of clastic sedimentary rock

sources. Sedimentary Geology 113, 111-124.

138

Franklin, S.B., Gibson, D.J., Robertson, P.A., Pohlmann, J.T., Fralish, J.S., 1995. Parallel

analysis: a method for determining significant principal components. Journal of

Vegetation Science 6, 99-106.

Gallet, S., Jahn, B., Van Vliet-Lanoë, B., Dia, A., Rossello, E., 1998. Loess geochemistry and

its implications for particle origin and composition of the upper continental crust. Earth

Planetary Science Letters 156, 157-172.

Ganning, B., 1971. On the ecology of Hetericyoris salinus, H. incongruens and Cypridopsis

aculeate (Crustacea: Ostracoda) from Baltic brackish-water rock pools. Marine Biology

8, 271-279.

Garzanti, E., Andò, S., France-Lanord, C., Vezzoli, G., 2009. Grain-size dependence of

sediment composition and environmental bias in provenance studies. Earth Planetary

Science Letters 277, 422-432.

Gillespie, A.R., Burke, R.M., Komatsu, G., Bayasgalan, A., 2008. Late Pleistocene glaciers in

Darhad Basin, northern Mongolia. Quaternary Research 69, 169-187.

Grunert, J., Lehmkuhl, F., Walther, M., 2000. Paleoclimatic evolution of the Uvs Nuur basin

and adjacent areas (western Mongolia). Quaternary International 65/66, 171-192.

Grunert, J., Stolz, C., Hempelmann, N., Hilgers, A., Hülle, D., Lehmkuhl, F., Felauer, T.,

Dasch, D., 2009. The evolution of small lake basins in the Gobi Desert in Mongolia.

Quaternary Sciences 29, 678-686.

Guérin, G., Mercier, N., Adamiec, G., 2011. Dose-rate conversion factors: update. Ancient TL

29, 5-8.

Gunin, P.D., Vostokova, E.A., Dorofeyuk, N.I., Tarasov, P.E., Black, C.C., 1999. Vegetation

dynamics of Mongolia. Geobotany 26, 233-238.

Guo, Y.J., Li, B.S., Wen, X.H., Wang, F.N., Niu, D.F., Shia, Y.J., Guo, Y.H., Jiang, S.P., Hu,

G.G., 2013. Holocene climate variation determined from rubidium and strontium

contents and ratios of sediments collected from the Badain Jaran Desert, Inner

Mongolia, China. Chemie der Erde-Geochemistry, doi: 10.1016/j.chemer.2013.09.001.

Gupta, A.K., Anderson, D.M., Overpeck J.T., 2003. Abrupt changes in the Asian southwest

monsoon during the Holocene and their links to the North Atlantic Ocean. Nature 421,

354-357.

Guttman, L., 1954. Some necessary conditions for common factor analysis. Psychometrika

19, 149-161.

Guy, A., Schulmann, K., Munschy, M., Miehe, J.-M., Edel, J.-B., Lexa, O., Fairhead, D.,

2014. Geophysical constraints for terrane boundaries in southern Mongolia. Journal of

Geophysical Research Solid Earth 119, 7966-7991.

Hammer, U.T., 1986. Saline lake ecosystems of the world. Monographiae Biologicae 59.

Kluwer Acad. Publ., Dordrecht.

139

Hartmann, K., Wünnemann, B., 2009. Hydrological changes and Holocene climate variations

in NW China, inferred from lake sediments of Juyanze palaeolake by factor analyses.

Quaternary International 194, 28-44.

Hartmann, K., Wünnemann, B., Hölz, S., Hrätschell, A., Zhang, H.C., 2011. Neotectonic

constraints on Gaxun Nur inland basin in north-central China, derived from remote

sensing, geomorphology and geophysical analyses. In: Gloaguen, R., Ratschbacher, L.,

(Eds.). Growth and collapse of the Tibetan Plateau. Geological Society London Special

Publications 353, 221-233.

Haug, G.H., Hughen, K.A., Sigman, D.M., Peterson, L.C., Röhl, U., 2001. Southward

migration of the Intertropical Convergence Zone through the Holocene. Science 293,

1304-1308.

Hempelmann, N.H., 2010. Aeolian geomorphodynamics in endorheic basins of Mongolian

Gobi Desert. Dissertation at Johannes Gutenberg-Universität Mainz, pp.31-57.

Herzschuh, U., 2006. Palaeo-moisture evolution in monsoonal Central Asia during the last

50,000 years. Quaternary Science Reviews 25, 163-178.

Heusser, L.E., 1989. Northeast Asian climatic change over the last 140,000 years inferred

from pollen in marine cores taken off the Pacific coast of Japan. In: Leinen, M.,

Sarnthein, M., (Eds.). Paleoclimatology and Paleometeorology: Modern and Past

Patterns of Global Atmospheric Transport. Dordrecht/Bosten/London, Kluwer

Academic Publishers, pp. 665-692.

Hilbig, W., 1995. Vegetation of Mongolia. SPB Academic Publishing, Amsterdam.

Hiller, D., 1972. Untersuchungen zur Biologie und zur Ökologie limnischer Ostracoden aus

der Umgebung von Hamburg. Archiv für Hydrobiologie, Supplement 40, 400-497.

Hillier, S., 1993. Origin, diagenesis, and mineralogy of chlorite minerals in Devonian

lacustrine mudrocks, Orcadian Basin, Scotland. Clays and Clay Minerals 41, 240-259.

Hölz, S., Polag, D., Becken, M., Fiedler-Volmer, R., Zhang, H.C., Hartmann, K., Burkhardt,

H., 2007. Electromagnetic and geoelectric investigation of the Gurinai Structure, Inner

Mongolia, NW China. Tectonophysics 445, 26-48.

Honda, M., Shimizu, H., 1998. Geochemical, mineralogical and sedimentological studies on

the Taklimakan Desert sands. Sedimentology 45, 1125-1143.

Honda, M., Yabuki, S., Shimizu, H., 2004. Geochemical and isotopic studies of aeolian

sediments in China. Sedimentology 51, 211-230.

Hong, B., Gasse, F., Uchida, M., Hong, Y.T., Leng, X.T., Shibata, Y., An, N., Zhu, Y.X.,

Wang, Y., 2014. Increasing summer rainfall in arid eastern-Central Asia over the past

8500 years. Scientific Report, doi:10.1038/srep05279.

Horiuchi, K., Minoura, K., Hoshino, K., Oda, T., Nakamura, T., Kawai, T., 2000.

Palaeoenvironmental history of Lake Baikal during the last 23,000 years.

Palaeogeography, Palaeoclimatology, Palaeoecology 157, 95-108.

140

Hövermann, J., 1988. The Sahara, Kalahari and Namib-Deserts: a geomorphological

comparison. In: Dardis, G.F., Moon, B.P., (Eds.). Geomorphological Studies in

Southern Africa. Balkema, Rotterdam, pp. 71-83.

Hovsgol Drilling Project Members, 2009. Sedimentary record from Lake Hovsgol, NW

Mongolia: Results from the HDP-04 and HDP-06 drill cores. Quaternary International

205, 21-37.

Hu, Z.X., Niu, Y.L., Liu, Y., Zhang, G.R., Sun, W.L., Ma, Y.X., 2014. Petrogenesis of

Ophiolite-type chromite deposits in China and some new perspectives. Geological

Journal of China Universities 20, 9-27.

Huang, W., Chen, J.H., Zhang, X.J., Feng, S., Chen, F.H., 2015. Definition of the core zone of

the “Westerlies-dominated climatic regime” and its controlling factors during the

instrumental period. Science China Earth Sciences 45, 379-388.

Hunt, J.E., Croudace, I.W., MacLachland, S.E., 2015. Use of calibrated ITRAX XRF data in

determining turbidite geochemistry and provenance in Agadir Basin, Northwest African

passive margin. In: Croudace, I.W., Guy Rothwell, R., (eds) Micro-XRF Studies of

Sediment Cores. Developments in Paleoenvironmental Research 17, pp 127-146.

Hülle, D., 2011. Lumineszenzdatierung von Sedimenten zur Rekonstruktion der jungquartären

Landschaftsentwicklung in der Mongolei. Dissertation zu Universität zu Köln.

IJmker, J.M., Stauch, G., Diekmann, B., Wünnemann, B., Lehmkuhl, F., 2012a. Dry periods

on the NE Tibetan Plateau during the late Quaternary. Palaeogeography,

Palaeoclimatology, Palaeoecology 346-347, 108-119.

IJmker, J.M., Stauch, G., Hartmann, K., Diekmann, B., Dietze, E., Opitz, S., Wünnemann, B.,

Lehmkuhl, F., 2012b. Environmental conditions in the Donggi Cona lake catchment,

NE Tibetan Plateau, based on factor analysis of geochemical data. Journal of Asian

Earth Sciences 44, 176-188.

Ilyashuk, B.P., Ilyashuk, E.A., 2007. Chironomid record of Late Quaternary climatic and

environmental changes from two sites in Central Asia (Tuva Republic, Russia) – local,

regional or global causes? Quaternary Science Reviews 26, 705-731.

ISO 14688-1, 2013. Geotechnical investigation and testing – Identification and classification

of soil – Part 1: Identification and description, ISO 14688-1: 2002 / Amd 1: 2013.

International Organization for Standardization, Geneva.

Ivanova, E.V., 2009. The Global Thermohaline Paleocirculation. Heidelberg, London, New

York, Springer Dordrecht.

Jahn, B.-M., Gallet, S., Han, J.M., 2001. Geochemical of the Xining, Xifeng and Jixian

sections, Loess Plateau of China: aeolian dust provenance and paleosol evolution during

the last 140 ka. Chemical Geology 178, 71-94.

Jankovská, V., Komárek, J., 2000. Indicative value of Pediastrum and other coccal green

algae in palaeoecology. Folia Geobotanica 35, 59-82.

141

Jiang, D.B., Lang, X.M., Tian, Z.P., Wang, T., 2012. Considerable model-data mismatch in

temperature over china during the Mid-Holocene: results of PMIP simulations. Journal

of Climate 25, 4135-4153.

Jiang, W.Y., Guo, Z.T., Sun, X.J., Wu, H.B., Chu, G.Q., Yuan, B.Y., Hatté, C., Guiot, J.,

2006. Reconstruction of climate and vegetation changes of Lake Bayanchagan (Inner

Mongolia): Holocene variability of the East Asian Monsoon. Quaternary Research 65,

411-420.

Jin, Z.D., An, Z.S., Yu, J.M., Li, F.C., Zhang, F., 2015. Lake Qinghai sediment geochemistry

linked to hydroclimate variability since the last glacial. Quaternary Science Reviews

122, 67-73.

Johnson, C.L., 2004. Polyphase evolution of the East Gobi basin: sedimentary and structural

records of Mesozoic-Cenozoic intraplate deformation in Mongolia. Basin Research 16,

79-99.

Johnsen, S.J., Dahl-Jensen, D., Gundestrup, N., Steffensen, J.P., Clausen, H.B., Miller, H.,

Masson-Delmotte, V., Sveinbjörnsdottir, A.E., White, J., 2001. Oxygen isotope and

palaeotemperature records from six Greenland ice-core stations: Camp Century, Dye-3,

GRIP, GISP2, Renland and NorthGRIP. Journal of Quaternary Sciences 16, 299-307.

Johnsson, M.J.,1993. The system controlling the composition of clastic sediments. Geological

Society of America Special Paper 284, 1-20.

Juschus, O., Pavlov, M., Schwamborn, G., Preusser, F., Fedorov, G., Melles, M., 2011. Late

Quaternary lake-level changes of Lake El’gygytgyn, NE Siberia. Quaternary Research

76, 441-451.

Kaiser, H.F., 1958. The varimax criterion for analytic rotation in factor analysis.

Psychometrika 23, 187–200.

Kaiser, H.F., 1960. The application of electronic computers to factor analysis. Educational

and Psychological Measurement 20, 141-151.

Karabanov, E.B., Prokopenko, A.A., Williams, D.F., Khursevich, G.K., 2000. A new record

of Holocene climate change from the bottom sediments of Lake Baikal.

Palaeogeography, Palaeoclimatology, Palaeoecology 156, 211-224.

Karabanov, E.B., Williams, D., Kuzmin, M., Sideleva, V., Khursevich, G., Prokopenko, A.,

Solotchina, E., Tkachenko, L., Fedenya, S., Kerber, E., Gvozdkov, A., Khlustov, O.,

Bezrukova, E., Letunova, P., Krapivina, S., 2004. Ecological collapse of Lake Baikal

and Lake Hovsgol ecosystems during the Last Glacial and consequences for aquatic

species diversity. Palaeogeography, Palaeoclimatology, Palaeoecology 209, 227-243.

Kazancı, N., Gulbabazadeh, T., Leroy, S.A.G., Ataselim, Z., Gürbüz, A., 2016. Aeolian

control on the deposition of high altitude lacustrine basins in the Middle East: The case

of Lake Neor, NW Iran. Quaternary International, doi:

dx.doi.org/10.1016/j.quaint.2015.11.040.

Kerr, R.A., 2003. Tropical Pacific a key to deglaciation. Science 299, 183-184.

142

Klinge, M., Böhner, J., Lehmkuhl, F., 2003. Climate pattern, snow- and timberlines in the

Altai Mountains, central Asia. Erdkunde 57, 296-308.

Klinge, M., Lehmkuhl, F., 2013. Geomorphologic map of Tsetseg Nuur Basin, Mongolian

Altai-lake development, fluvial sedimentation and aeolian transport in a semi-arid

environment. Journal of Maps 9, 361-366.

Komatsu, G., Brantingham, P.J., Olsen, J.W., Baker, V.R., 2001. Paleoshoreline

geomorphology of Böön Tsagaan Nuur, Tsagann Nuur and Orog Nuur: the Valley of

Lakes, Mongolia. Geomrphology 39, 83-98.

Komatsu, T., Tsukamoto, S., 2015. Late Glacial lake-level changes in the Lake Karakul basin

(a closed glacierized-basin), eastern Pamirs, Tajikistan. Quaternary Research 83, 137-

149.

Konert, M., Vandenberghe, J., 1997. Comparison of laser grain size analysis with pipette and

sieve analysis: a solution for the underestimation of the clay fraction. Sedimentology 44,

523-535.

Kostrova, S.S., Meyer, H., Chapligin, B., Tarasov, P.E., Bezrikova, E.V., 2014. The last

glacial maximum and late glacial environmental and climate dynamics in the Baikal

region inferred from an oxygen isotope record of lacustrine diatom silica. Quaternary


Kovács, J., 2008. Grain-size analysis of the Neogene red clay formation in the Pannonian

Basin. International Journal of Earth Sciences 97, 171-178.

Kramer, A., Herzschuh, U., Mischke, S., Zhang, C.J., 2010. Late Quaternary environment

history of the south-eastern Tibetan plateau inferred from the lake Naleng non-pollen

palynomorph record. Vegetation History and Archaeobotany 19, 453-468.

Kylander, M.E., Ampel, L., Wohlfarth, B., Veres, D., 2011. High-resolution X-ray

fluorescence core scanning analysis of Les Echets (France) sedimentary sequence: new

insights from chemical proxies. Journal of Quaternary Science 26, 109-117.

Lai, Z.P., Mischke, S., Madson, D., 2014. Paleoenvironmental implications of new OSL dates

on the formation of the “Shell Bar” in the Qaidam Basin, northeastern Qinghai-Tibetan

Plateau. Journal of Paleolimnology 51, 197-210.

Lamb, H., Bates, C., Coombes, P., Marshall, M., Umer, M., Davies, S., Dejen, E., 2007. Late

Pleistocene desiccation of Lake Tana, source of the Blue Nile. Quaternary Science

Reviews 26, 287–299.

Last, W.M., 1990. Lacustrine dolomite-an overview of modern, Holocene, and Pleistocene

occurrences. Earth-Science Reviews 27, 211-263.

Lauterbach, S., Brauer, A., Andersen, N., Danielopol, D., Dulski, P., Hüls, M., Milecka, K.,

Namiotko, T., Obremska, M., Grafenstein, U., Participants, D., 2011. Environmental

responses to Lateglacial climatic fluctuations recorded in the sediments of pre-alpine

Lake Mondsee (northeastern Alps). Journal of Quaternary Science 26, 253–267.

143

Lee, M.K., Lee, Y.I., Lim, H.S., Lee, J.I., Choi, J.H., Yoon, H.I., 2011. Comparison of

radiocarbon and OSL dating methods for a Late Quaternary sediment core from Lake

Ulaan, Mongolia. Journal of Paleolimnology 45, 127-135.

Lee, M.K., Lee, Y.I., Lim, H.S., Lee, J.I., Yoon, H.I., 2013. Late Pleistocene-Holocene

records from Lake Ulaan, southern Mongolia: implications for East Asian

palaeomonsoonal climate changes. Journal of Quaternary Sciences 28, 370-378.

Lehmkuhl, F., 1997. Flächenhafte Erfassung der Landschaftsdegradation im Becken von

Zoige (Osttibet) mit Hilfe von Landsat-TM-Daten. Göttinger Geographische

Abhandlungen 100, 179-194.

Lehmkuhl, F., Böhner, J., Stauch, G., 2003. Geomorphologische Formungs- und

Prozessregionen in Zentralasien. Petermanns Geographische Mitteilungen 147, 6-13.

Lehmkuhl, F., Haselein, F., 2000. Quaternary paleoenvironmental change on the Tibetan

Plateau and adjacent areas (western China and western Mongolia). Quaternary

International 65/66, 121-145.

Lehmkuhl, F., Klinge, M., Rother, H., Hülle, D., 2015. Distribution and timing of Holocene

and late Pleistocene glacier fluctuations in western Mongolia. Annals of Glaciology 71,

doi: dx.doi.org/10.3189/2016AoG71A030.

Lehmkuhl, F., Klinge, M., Stauch, G., 2011. The extent and timing of Late Pleistocene

glaciations in the Altai and neighboring mountain systems. In: Ehlers, J., Gibbard, P.L.,

Hughes, P.D., (Eds.). Developments in Quaternary Sciences 15, Amsterdam, The

Netherlands, 967-979.

Lehmkuhl, F., Lang, A., 2001. Geomorphological investigations and luminescence dating in

the southern part of the Khangay and the Valley of the Gobi Lakes (Central Mongolia).

Journal of Quaternary Sciences 16, 69-87.

Lehmkuhl, F., Schulte, P., Zhao, H., Hülle, D., Protze, J., Stauch, G., 2014. Timing and

spatial distribution of loess and loess-like sediments in the mountain areas of the

northeastern Tibetan Plateau. Catena 117, 23-33.

Leinen, M., 1989. The late Quaternary record of atmospheric transport to the northwest

Pacific from Asia. In: Leinen, M., Sarnthein, M., (Eds.). Paleoclimatology and

Paleometeorology: Modern and Past Patterns of Global Atmospheric Transport.

Dordrecht/Bosten/London, Kluwer Academic Publishers, pp. 693-732.

Li, X.Q., Zhao, K.L., Dodson, J., Zhou, X.Y., 2011. Moisture dynamics in central Asia for the

last 15 k: new evidence from Yili Valley, Xinjiang, NW China. Quaternary Science

Reviews 30, 3457-3466.

Liang, L.J., Jiang, H.C., 2015. Geochemical composition of the last deglacial lacustrine

sediments in East Tibet and implications for provenance, weathering, and earthquake

events. Quaternary International, doi:10.1016/j.quaint.2015.07.037.

Lisiecki, L.E., Raymo, M.E., 2005. A Pliocene-Pleistocene stack of 57 globally distributed

benthic δ18O records. Paleoceanography 20, PA1003, doi: 10.1029/2004PA001071.

144

Lister, G.S., Kelts, K., Chen, K.Z., Yu, J.Q., Niessen, F., 1991. Lake Qinghai, China: closed-

basin lake levels and the oxygen isotope record for ostracoda since the latest

Pleistocene. Palaeogeography, Palaeoclimatology, Palaeoecology 84, 141-162.

Liu, C.Q., Masuda, A., Okada, A., Yabuki, S., Zhang, J., Fan, Z.L., 1993. A geochemical

study of loess and desert sand in northern China: Implications for continental crust

weathering and composition. Chemical Geology 106, 359-374.

Liu, X.J., Lai, Z.P., Fan, Q.S., Long, H., Sun, Y.J., 2010. Timing for high lake levels of

Qinghai Lake in the Qinghai-Tibetan Plateau since the Last Interglaciation based on

quartz OSL dating. Quaternary Geochronology 5, 218-222.

Liu, Z.Y., Otto-Bliesner, B.L., He, F., Brady, E.C., Tomas, R., Clark, P.U., Carlson, A.E.,

Lynch-Stieglitz, J., Curry, W., Brook, E., Erickson, D., Jacob, R., Kutzbach, J., Cheng,

J., 2009. Transient simulation of Last Deglaciation with a now mechanism for Bølling-

Allerød warming. Science 325, 310-314.

Loibl, D., 2015. Late Holocene glacier change in the eastern Nyainqêntanglha Range, SE

Tibet. Dissertation to RWTH Aachen University, Aachen, p. 5.

Long, H., Shen, J., 2015. Underestimated 14C-based chronology of late Pleistocene high lake-

level events over the Tibetan Plateau and adjacent areas: evidence from the Qaidam

Basin and Tengger Desert. Science China Earth Sciences 58, 183-194.

Lu, H.Y., Sun, D.H., 2000. Pathways of dust input to the Chinese Loess Plateau during the

last glacial and interglacial periods. Catena 40, 251-261.

Lu, H.Y., Vandenberghe, J., An, Z.S., 2001. Aeolian origin and palaeoclimatic implications of

the 'red clay' (north China) as evidenced by grain-size distribution. Journal of

Quaternary Science 16, 89-97.

Lu, H.Y., Yi, S.W., Xu, Z.W., Zhou, Y.L., Zeng, L., Zhu, F.Y., Feng, H., Dong, L.N., Zhuo,

H.X., Yu, K.F., Mason, J., Wang, X.Y., Chen, Y.Y., Lu, Q., Wu, B., Dong, Z.B., Qu,

J.J, Wang, X.M., Guo, Z.T., 2013. Chinese deserts and sand fields in Last Glacial

Maximum and Holocene Optimum. Chinese Science Bulletin 58, 2775-2783.

Lu, H.Y., Zhao, C.F., Mason, J., Yi, S.W., Zhao, H., Zhou, Y.L., Ji, J.F., Swinehart, J., Wang,

C.M., 2010. Holocene climate changes revealed by aeolian deposits from the Qinghai

Lake area (northeastern Qinghai-Tibetan Plateau) and possible forcing mechanisms. The

Holocene 21, 297-304.

Lupker, M., France-Lanord, C., Lavé, J., Bouchez, J., Galy, V., Métivier, F., Gaillardet, J.

Lartiges, B., Mugnier, J., 2011. A Rouse-based method to integrate the chemical

composition of river sediments: Application to the Ganga basin. Journal of Geophysical

Research 116, doi: 10.1029/2010JF001947.

Ma, Y.Z., Liu, K.-B., Feng, Z.D., Meng, H.W., Sang, Y.L., Wang, W., Zhang, H.C., 2013.

Vegetation changes and associated climate variations during the past ~38,000 years

reconstructed from the Shaamar eolian-paleosol section, northern Mongolia. Quaternary


145

Ma, Y.Z., Liu, K.-B., Feng, Z.D., Sang, Y.L., Wang, W., Sun, A.Z., 2008. A survey of

modern pollen and vegetation along a south-north transect in Mongolia. Journal of

Biogeography 35, 1512-1532.

Mackay, A.W., Bezrukova, E.V., Leng, M.J., Meaney, M., Nunes, A., Piotrowska, N., Self,

A., Shchetnikov, A., Shilland, E., Tarasov, P., Wang, L, White, D., 2012. Aquatic

ecosystem responses to Holocene climate change and biome development in boreal,

central Asia. Quaternary Science Review 41, 119-131.

Madsen, D.B., Ma, H.Z., Rhode, D., Jeffrey Brantingham, P., Forman, S.L., 2008. Age

constraints on the late Quaternary evolution of Qinghai Lake, Tibetan Plateau.

Quaternary Research 69, 316-325.

Maher, B.A., Mutch, T.J., Cunningham, D., 2009. Magnetic and geochemical characteristics

of Gobi Desert surface sediments: Implications for provenance of the Chinese Loess

Plateau. Geology 37, 279-282.

Martinez-Ruiz, F., Kastner, M., Gallego-Torres, D., Rodrigo-Gámiz, M., Nieto-Moreno, V.,

Ortega-Huertas, M., 2015. Paleoclimate and paleoceanography over the past 20,000 yr

in the Mediterranean Sea Basins as indicated by sediment elemental proxies.


Mason, J.A., Greene, R.S.B., Joeckel, R.M., 2011. Laser diffraction analysis of the

disintegration of aeolian sedimentary aggregates in water. Catena 87, 107-118.

McLennan, S.M., 1993. Weathering and global denudation. The Journal of Geology 101, 295-

303.

McLennan, S.M., 2001. Relationships between the trace element composition of sedimentary

rocks and upper continental crust. Geochemistry, Geophysics, Geosystems 2, doi:

10.1029/2000GC000109.

Meisch, F., 2000. Freshwater ostracoda of western and central Europe. Spektrum, Gustav

Fischer, p. 522.

Melles, M., Brigham-Grette, J., Minyuk, P.S., Nowaczyk, N.R., Wennrich, V., DeConto,

R.M., Anderson, P.M., Andreev, A.A., Coletti, A., Cook, T.L., Haltia-Hovi, E.,

Kukkonen, M., Lozhkin, A.V., Rosén, P., Tarasov, P., Vogel, H., Wagner, B., 2012. 2.8

million years of Arctic climate change from Lake El’gygytgyn, NE Russia. Science 337,

315-320.

Meyers, P.A., Lallier-Vergès, E., 1999. Lacustrine sedimentary organic matter records of Late

Quaternary paleoclimates. Journal of Paleolimnology 21, 345-372.

Meyers, P.A., Teranes, J.L., 2001. Sediment organic matter. In: Last, W.M., Smol, J.P.,

(Eds.). Tracking Environmental Changes Using Lake Sediments. Vol. 2: Physical and

Geochemical Methods, pp. 239-270.

Miehe, G., Schlütz, F., Miehe, S., Opgenoorth, L., Cermak, J., Samiya, R., Jäger, E.J.,

Wesche, K., 2007. Mountain forest islands and Holocene environmental changes in

Central Asia: A case study from the southern Gobi Altay, Mongolia. Palaeogeography,


146

Millot, G., 1970. Geology of Clays, Weathering, Sedimentology, Geochemistry. Springer-

Verlag Wien Gmbh, Vienna, pp. 357-381.

Mischke, S., Demske, D., Shudack, M.E., 2003. Hydrologic and climatic implications of a

multidisciplinary study of the mid to late Holocene Lake Eastern Juyanze. Chinese

Science Bulletin 48, 1411-1417.

Mischke, S., Herzschuh, U., Zhang, C.J., Bloemendal, J., Riedel, F., 2005. A Late Quaternary

lake record from the Qilian Mountains (NW China): lake level and salinity changes

inferred from sediment properties and ostracod assemblages. Global and Planetary

Changes 46, 337-359.

Mischke, S., Weynell, M., Zhang, C.J., Wiechert, W., 2013. Spatial variability of 14C

reservoir effects in Tibetan Plateau lakes. Quaternary International 313-314, 147-155.

Mischke, S., Zhang, C.J., 2011. Ostracod distribution in Ulungur Lake (Xinjiang, China) and

a reassessed Holocene record. Ecological Research 26: 133-145.

Mölg, T., Maussion, F., Scherer, D., 2013. Mid-latitude westerlies as a driver of glacier

variability in monsoonal High Asia. Nature Climate Change 4, 68-73.

Moore, P.D., Webb, J.A., Collinson, M.E., 1991. Pollen analysis, 2nd. Ed. Oxford: Blacwell

Scientific Publication.

Moros, M., Emeis, K., Risebrobakken, B., 2004. Sea surface temperature and ice rafting in the

Holocene North Atlantic: Climate influences on northern Europe and Greenland.

Quaternary Science Reviews 23: 2113-2160.

Mukherji, A.K., 1970. Analytical chemistry of zirconium and hafnium. Pergamon Press,

Oxford, pp. 1-11.

Murad, W., 2011. Late Quaternary vegetation history and climate change in the Gobi Desert,

south Mongolia. Dissertation at Georg-August-Universität Göttingen, pp.67-82.

Murakami, T., Katsuta, N., Yamamoto, K., Takamatsu, N., Takano, M., Oda, T., Matsumoto,

G.I., Horiuchi, K., Kawai, T., 2010. A 27-k record of environmental change in central

Asia inferred from the sediment record of Lake Hovsgol, northwest Mongolia. Journal

of Paleolimnology 43, 369-383.

Murakami, T., Takamatsu, T., Katsuta, N., Takano, M., Yamamoto, K., Takahashi, Y.,

Nakamura, T., Kawai, T., 2012. Centennial- to millennial-scale climate shifts in

continental interior Asia repeated between warm-dry and cool-wet conditions during the

last three interglacial states: evidence from uranium and biogenic silica in the sediment

of Lake Baikal, southeast Siberia. Quaternary Science Reviews 52, 49-59.

Murray, A.S., Wintle, A.G., 2003. The single aliquot regenerative dose protocol: potential for

improvements in reliability. Radiation Measurements 37, 377-381.

Murray, R.W., Buchholtz Ten Brink, M.R., Gerlach, D.C., Price Russ III, G., Jones, D.L.,

1992. Rare earth, major, and trace element composition of Monterey and DSDP chert

and associated host sediment: Assessing the influence of chemical fractionation during

diagenesis. Geochemica et Cosmochimica Acta 56, 2657-2671.

147

Nakagawa, T., Kitagawa, H., Yasuda, Y., Tarasov, P.E., Nishida, K., Gotanda, K., Sawai, Y.,

Yangtze River Civilization Program Members, 2003. Asynchronous climate changes in

the north Atlantic and Japan during the Last Termination. Science 299, 688-691.

Nakai, S., Halliday, A.N., Rea, D.K., 1993. Provenance of dust in the Pacific Ocean. Earth

and Planetary Science Letters 119, 143-157.

Nagashima, K., Tada, R., Tani, A., Toyoda, S., Sun, Y.B., Isozaki, Y., 2007. Contribution of

aeolian dust in Japan Sea sediments estimated from ESR signal intensity and

crystallinity of quartz. Geochemistry, Geophysics, Geosystems 8, doi:

10.1029/2006GC001364.

Naumann, S., 1999. Spät- und postglaziale Landschaftsentwicklung im Bajan Nuur

Seebecken (Nordwestmongolei). Die Erde 130, 117-130.

Nesbitt, H.W., Young, G.M., 1982. Early Proterozoic climates and plate motions inferred

from major element chemistry of lutites. Nature 299, 715-717.

Nesbitt, H.W., Young, G.M., 1996. Petrogenesis of sediments in the absence of chemical

weathering: effects of abrasion and sorting on bulk composition and mineralogy.

Sedimentology 43, 341-358.

Nickel, E.H., 1954. The distribution of major and minor elements among some co-existing

ferromagnesian silicates. American Mineralogist 6, 486-493.

Nilson, E., Lehmkuhl, F., 2001. Interpreting temporal patterns in the late Quaternary dust flux

from Asia to the North Pacific. Quaternary International 76/77, 67-76.

NIOZ, Avaatech, 2007. XRF core scanner user manual version 2.0. Royal Netherlands

Institute of Sea Research and Avaatech XRF Core Scanner Technology.

Norman, M.D., De Deckker, P., 1990. Trace metals in lacustrine and marine sediments: A

case study from the Gulf of Carpentaria, northern Australia. Chemical Geology 82, 299-

318.

Nottebaum, V., Lehmkuhl, F., Stauch, G., Hartmann, K., Wünnemann, B., Schimpf, S., Lu,

H.Y., 2014. Regional grain size variations in aeolian sediments along the transition

between Tibetan highlands and north-western Chinese deserts – the influence of

geomorphology settings on aeolian transport pathways. Earth Surface Processes

Landforms, doi: 10.1002/sep.3590.

Nottebaum, V., Lehmkuhl, F., Stauch, G., Lu, H.Y., Yi, S.W., 2015a. Late Quaternary aeolian

sand deposition sustained by fluvial reworking and sediment supply in the Hexi

Corridor — An example from northern Chinese drylands. Geomorphology 250, 113-

127.

Nottebaum, V., Stauch, G., Hartmann, K., Zhang, J.R., Lehmkuhl, F., 2015b. Unmixed loess

grain size populations along the northern Qilian Shan (China): Relations between

geomorphologic, sedimentologic and climatic controls. Quaternary International 372,

151-166.

148

Ohlendorf, C., Wennrich, V., Enters, D., 2015. Experiences with XRF-Scanning of Long

Sediment Records. In: Croudace, I.W., Guy, Rothwell, R., (eds) Micro-XRF Studies of

Sediment Cores. Developments in Paleoenvironmental Research 17, pp 351-372.

Orkhonselenge, A., Krivonogov, S.K., Mino, K., Kashiwaya, K., Safonova, I.Y., Yamamoto,

M., Kashima, K., Nakamura, T., Kim, J.Y., 2013. Holocene sedimentary records from

Lake Borsog, eastern shore of Lake Khuvsgul, Mongolia, and their paleoenvironmental

implications. Quaternary International 290-291, 95-109.

Owen, L.A., Cunningham, W.D., Richards, B.W.M., Rhodes, E., Windley, B.F., Dorjnamajaa,

D., Badamgarav, J., 1999. Timing of formation of forebergs in the northeastern Gobi

Altai, Mongolia: implications for estimating mountain uplift rates and earthquake

recurrence intervals. Journal of the Geological Society, London 156, 457-464.

Owen, L.A., Richards, B., Rhodes, E.J., Cunningham, W.D., Windley, B.F., Badamgarav, J.,

Dorjnamjaa, D., 1998. Relic permafrost structures in the Gobi of Mongolia: age and

significance. Journal of Quaternary Sciences 13, 539-547.

Owen, L.A., Windley, B.F., Cunningham, W.D., Badamgarav, J., Dorjnamjaa, D., 1997.

Quaternary alluvial fans in the Gobi of southern Mongolia: evidence for neotectonics

and climate change. Journal of Quaternary Sciences 12, 239-252.

Owen, M.R., 1987. Hafnium content of detrital zircons, a new tool for provenance study.

Journal of Sedimentary Petrology 57, 824-830.

Özer, M., Orhan, M., Isik, N. S., 2010. Effect of particle optical properties on size distribution

of soils obtained by laser diffraction. Environmental & Engineering Geoscience 16,

163-173.

Pachur, H.-J., Wünnemann, B., 1995. Lake evolution in the Tengger Desert, northwestern

China, during the last 40,000 years. Quaternary Research 44, 171-180.

Parnell, A.C., Haslett, J., Allen, J.R.M., Buck, C.E., Huntley, B., 2008. A flexible approach to

assessing synchronicity of past events using Bayesian reconstructions of sedimentation

history. Quaternary Science Reviews 27, 1872–1885.

Peck, J.A., Khosbayar, P., Fowell, S.J., Pearce, R.B., Ariunbileg, S., Hansen, B.C.S.,

Soninkhishig, N., 2002. Mid to Late Holocene climate change in north central Mongolia

as recorded in the sediments of Lake Telmen. Palaeogeography, Palaeoclimatology,

Palaeoecology 183, 135-153.

Pederson, N., Hessl, A.E., Baatarbileg, N., Anchukaitis, K.J., Di Cosmo, N., 2014. Pluvials,

droughts, the Mongol Empire, and modern Mongolia. Proceedings of the National

Academy of Sciences, doi:10.1073/pnas.1318677111.

Peltier, W.R., Fairbanks, R.G., 2006. Global glacial ice volume and Last Glacial Maximum

duration from an extended Barnados sea level record. Quaternary Science Reviews 25,

3322-3337.

Petschick, R., Kuhn, G., Gingele, F.X., 1996. Clay mineral distribution in surface sediments

of the South Atlantic: Sources, transport, and relation to oceanography. Marrine

Geology 130, 203-229.

149

Pettijohn, F.J., Potter, P.E., Siever, R., 1973. Sand and Sandstone. Springer-Verlag, New

York, pp. 24-63.

Porter, S.C., An, Z.S., 1995. Correlation between climate events in the north Atlantic and

China during the last glaciation. Nature 375, 305-308.

Prescott, J.R., Hutton, J.R., 1994. Cosmic ray contribution to dose rate for luminescence and

ESR dating: Large depth and long-term variations. Radiation Measurements 23, 479-

500.

Prins, M.A., Vriend, M., Nugteren, G., Vandenberghe, J., Lu, H.Y., Zheng, H.B., Weltje, G.J.,

2007. Late Quaternary aeolian dust input variability on the Chinese Loess Plateau:

inferences from unmixing of loess grain-size records. Quaternary Science Reviews 26,

230-242.

Prokopenko, A.A., Karabanov, E.B., Williams, D.F., Kuzmin, M.I., Khursevich, G.K.,

Gvozdkov, A.A., 2001. The detailed record of climatic events during the past 75,000 s

BP from the Lake Baikal drill core BDP-93-2. Quaternary International 80/81, 59-68.

Prokopenko, A.A., Karabanov, E.B., Williams, D.F., Khursevich, G.K., 2002. The stability

and the abrupt ending of the last Interglaciation in southeastern Siberia. Quaternary

Research 58, 56-59.

Prokopenko, A.A., Khursevich, G.K., Bezrukova, E.V., Kuzmin, M.I., Boes, X., Williams,

D.F., Fedenya, S.A., Kulagina, N.V., Letunova, P.P., Abzaeva, A.A., 2007.

Paleoenvironmental proxy records from Lake Hovsgol, Mongolia, and a synthesis of

Holocene climate change in the Lake Baikal watershed. Quaternary Research 68, 2-17.

Pye, K., Zhou, L.P., 1989. Late Pleistocene and Holocene Eolian Dust Deposition in North

China and the Northwest Pacific Ocean. Palaeogeography, Palaeoclimatology,


Rahmstorf, S., 2002. Ocean circulation and climate during the past 120,000 years. Nature

419, 207-214.

Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G., Bronk Ramsey, C.,

Grootes, P.M., Guilderson, T. P., Haflidason, H., Hajdas, I., HattŽ, C., Heaton, T. J.,

Hoffmann, D. L., Hogg, A. G., Hughen, K. A., Kaiser, K. F., Kromer, B., Manning, S.

W., Niu, M., Reimer, R. W., Richards, D. A., Scott, E. M., Southon, J. R., Staff, R. A.,

Turney, C. S. M., van der Plicht, J., 2013. IntCal13 and Marine13 Radiocarbon Age

Calibration Curves 0-50,000 Years cal BP. Radiocarbon 55, 1869-1887.

Rhodes, T.E., Gasse, F., Lin, R.F., Fontes, J.-C., Wei, K.Q., Bertrand, P., Gibert, E., Mélières,

F., Tucholka, P., Wang, Z.X., Cheng, Z.Y., 1996. A Late-Holocene lacustrine record

from Lake Manas, Zunggar (northern Xinjiang, western China). Palaeogeography,


Richter, R.O., van der Gaast, S., Koster, R., Vaars, A., Gieles, R., de Stigter, H.C., de Haas,

H., van Weering, T.C.E., 2006. The Avaatech XRF Core Scanner: technical description

and applications to NE Atlantic sediments. In: Rothwell, R.G., (ed) New techniques in

sediment core analysis. Geological Society London Special Publications 267, 39-50.

150

Ritz, J.-F., Vassallo, R., Braucher, R., Brown, E.T., Carretier, S., Bourlès, D.L., 2006. Using

in situ-produced 10Be to quantify active tectonics in the Gurvan Bogd mountain range

(Gobi-Altay, Mongolia). Geological Society of America Special Paper 415, 87-109.

Rodwell, M.J., Rowell, D.P., Folland, C.K., 1999. Oceanic forcing of the wintertime North

Atlantic Oscillation and European climate. Nature 398, 320-323.

Roser, B.P., Korsch, R.J., 1988. Provenance signatures of sandstone-mudstone suites

determined using discriminant function analysis of major-element data. Chemical

Geology 67, 119-139.

Rother, H., Lehmkuhl, F., Fink, D., Nottebaum, V., 2014. Surface exposure dating reveals

MIS-3 glacial maximum in the Khangai Mountains of Mongolia. Quaternary Research

82, 297-308.

Roy, P.D., Caballero, M., Lozano, S., Morton, O., Lozano, R., Jonathan, M.P., Sánchez, J.L.,

Macías, M.C., 2012. Provenance of sediments deposited at plaeolake San Felipe,

western Sonora Desert: Implications to regimes of summer and winter precipitation

during last 50 cal kyr BP. Journal of Arid Environments 81, 47-58.

Rudaya, N., Tarasov, P., Dorofeyuk, N., Solovieva, N., Kalugin, I., Andreev, A., Daryin, A.,

Diekmann, B., Riedel, F., Tserendash, N., Wagner, M., 2009. Holocene environments

and climates in the Mongolian Altai reconstructed from the Hoton-Nur pollen and

diatom records: a step towards better understanding climate dynamics in Central Asia.


Rudersdorf, A., Hartmann, K., Yu, K.F., Stauch, G., Reicherter, K., 2015. Seismites as

indicators for Holocene seismicity in the northeastern Ejina Basin, Inner Mongolia,

China. In: Landgraf, S., Stein, S., Hintersbergen, E., (Eds.). Seismicity, Fault Rupture

and Earthquake Hazards in Slowly Deforming Regions. Geological Society London

Special Publications 432, doi: 10.1144/SP432.6.

Rudnick, R.L., Gao, S., 2003. Composition of the continental crust. In: Holland, H.D.,

Turekian, K.K., (Eds.). Treatise on Geochemistry, vol. 3. Elservier-Pergamon, Oxford-

London, pp. 53-54.

Sato, T., Tsujimura, M., Yamanaka, T., Iwasaki, H., Sugimoto, A., Sugita, M., Kimura, F.,

Davaa, G., Oyunbaatar, D., 2007. Water sources in semiarid northwest Asia as revealed

by field observations and isotope transport model. Journal of Geophysical Research,

doi: 10.1029/2006JD008321.

Scheffer, F., Schachtschabel, P., 2002. Lehrbuch der Bodenkunde. 15. Auflage. Spektrum

Akademischer Verlag, Derdelberg, Berlin.

Schettler, G., Romer, R.L., Qiang, M.R., Plessen, B., Dulski, P., 2009. Size-dependent

geochemical signatures of Holocene loess deposits from the Hexi Corridor (China).

Journal of Asian Earth Sciences 35, 103-136.

Schillereff, D.N., Chiverrell, R.C., Croudace, I.W., Boyle, J.F., 2015. An Inter-comparison of

µXRF Scanning Analytical Methods for Lake Sediments. In: Croudace, I.W., Guy

Rothwell, R. (eds) Micro-XRF Studies of Sediment Cores. Developments in

Paleoenvironmental Research 17, pp 583-600.

151

Schimpf, S., Nottebaum, V., Diekmann, B., Hartmann, K., Lehmkuhl, F., Wünnemann, B.,

Zhang, C.J., 2014. From source to sink in the sediment cascade of the Hei-River Basin:

Implications for late Quaternary landscape dynamics in the Gobi Desert, NW China.

Geophysical Research Abstracts 16, EGU2014-9905.

Schlütz, F., Lehmkuhl, F., 2007. Climatic change in the Russian Altai, southern Siberia based

on palynological and geomorphological results with implications for climatic

teleconnections and human history since the middle Holocene. Vegetation History and

Archaeobotany 16, 101-118.

Schulte, P., Lehmkuhl, F., Steininger, F., Loibl, D., Lockot, G., Protze, J., Fischer, P., Stauch,

G., 2016. Influence of HCl pretreatment and organo-mineral complexes on laser

diffraction measurement of loess-paleosol-sequences. Catena 137, 392-405.

Schwanghart, W., Frechen, M., Kuhn, N.J., Schütt, B., 2009. Holocene environmental

changes in the Uggi Nuur basin, Mongolia. Palaeogeography, Palaeoclimatology,


Schwanghart, W., Schütt, B., Walther, M., 2008. Holocene climate evolution of the Ugii Nuur

Basin, Mongolia. Advances in Atmospheric Sciences 25, 986-998.

Selvaraj, K., Chen, C.-T. A., Babu C.P., Lou, J.Y., Liu, C.-L., Hsu, K.J., 2012. Glacially

derived material in an Inner Mongolian desert lake during the Marine Isotope Stage 2.

Journal of Quaternary Science 27, 725-733.

Shanahan, T.M., McKay, N., Overpeck, J.T., Peck, J.A., Scholz, C., Heil, C.W., King, J.,

2013. Spatial and temporal variability in sedimentological and geochemical properties

of sediments from an anoxic crater lake in West Africa: implications for

paleoenvironmental reconstructions. Palaeogeography, Palaeoclimatology,


Shanahan, T.M., Overpeck, J.T., Hubeny, J.B., King, J., Hu, F.S., Hughen, K., Miller, G.,

Black, J., 2008. Scanning micro-X-ray fluorescence elemental mapping: A new tool for

the study of laminated sediment records. Geochemistry, Geophysics, Geosystems 9,

Q02016, doi: 10.1029/2007GC001800.

Shichi, K., Takahara, H., Krivonogov, S.K., Bezrukova, E.V., Kashiwaya, K., Takehara, A.,

Nakamura, T., 2009. Late Pleistocene and Holocene vegetation and climate records

from Lake Kotokel, central Baikal region. Quaternary International 205, 98-110.

Sirocko, F., Sarnthein, M., Erlenkeuser, H., 1993. Century-scale events in monsoonal climate

over the past 24,000 years. Nature 364, 322-324.

Smith, G.I., Haines, D.V., 1964. Character and distribution of nonclastic minerals in the

Searles Lake evaporate deposit, California. United States Government Publishing Office

1181, pp. 1-58.

Spearman, C., 1904. The proof and measurement of association between two things. Journal

of Psychology 15, 72-101.

Spectro Xepos, 2007. Analysis of trace elements in geological materials, soils and sludges.

SPECTRO XRF Report, 1-5.

152

Staubwasser, M., Weiss, H., 2006. Holocene climate and cultural evolution in late prehistoric-

early historic West Asia. Quaternary Research 66, 372-387.

Stebich, M., Mingram, J., Han, J.T., Liu, J.Q., 2009. Late Pleistocene spread of (cool-)

temperate forests in Northeast China and climate changes synchronous with the North

Atlantic region. Global and Planetary Change 65, 56-70.

Stein, R., Grobe, H., Wahsner, M., 1994. Organic carbon, carbonate, and clay mineral

distributions in eastern central Arctic Ocean surface sediments. Marine Geology 119,

269-285.

Stevens, T., Palk, C., Carter, A., Lu, H.Y., Clift, P.D., 2010. Assessing the provenance of

loess and desert sediments in northern China using U-Pb dating and morphology of

detrital zircons. Geological Society of America Bulletin 122, 1331-1344.

Stolz, C., Hülle, D., Hilgers, A., Grunert, J., Lehmkuhl, F., Dasch, D., 2012. Reconstructing

fluvial, lacustrine and aeolian process dynamics in Western Mongolia. Zeitschrift für

Geomorphologie 56, 267-230.

Stuiver, M.P., Reimer, J., Bard, E., Burr, G.S., Hughen, K.A., Kromer, B., McCormac, G.,

Plicht, J.v.d., Spurk, M., 1998. INTCAL98 radiocarbon age calibration. Radiocarbon

40, 1041-1083.

Sun, D.H., 2004. Monsoon and westerly circulation changes recorded in the late Cenozoic

aeolian sequences of Northern China. Global and Planetary Change 41, 63-80.

Sun, D.H., Lu, H.Y., David, R., Sun, Y.B., Wu, S.G., 2000. Bimode grain size distribution of

Chinese Loess and its paleoclimate implication. Acta Sedimentologica Sinica 18, 327-

335.

Sun, Y.B., Tada, R., Chen, J., Liu, Q.S., Toyoda, S., Tani, A., Ji, J.F., Isozaki, Y., 2008.

Tracing the provenance of fine-grained dust deposited on the central Chinese Loess

Plateau. Geophysical Research Letters, doi: 10.1029/2007GL031672.

Tamura, K., Asano, M., Jamsran, U., 2012. Soil diversity in Mongolia. In

: Yamamura, N., Fujita, N., Maekawa, A., (Eds.). The Mongolian Ecosystem Network,

Environmental Issues under Climate and Social Changes. Springer Tokyo, pp. 99-103.

Tapponnier, P., Molnar, P., 1979. Active faulting and Cenozoic tectonics of the Tien Shan,

Mongolia, and Baykal region. Journal of Geophysical Research 84, 3425-3459.

Tarasov, P.E., Bezrukova, E., Karabanov, E., Nakagawa, T., Wagner, M., Kulagina, N.,

Letunova, P., Abzaeva, A., Granoszewski, W., Riedel, F., 2007. Vegetation and climate

dynamics during the Holocene and Eemian interglacials derived from Lake Baikal

pollen records. Palaeogeography, Palaeoclimatology, Palaeoecology 252, 440-457.

Tarasov, P.E., Peyron, O., Guiot, J., Brewer, S., Volkova, V.S., Bezusko, L.G., Dorofeyuk,

N.I., Kvavadze, E.V., Osipova, I.M., Panova, N.K., 1999. Last Glacial Maximum

climate of the former Soviet Union and Mongolia reconstructed from pollen and plant

macrofossil data. Climate Dynamics 15, 227-240.

Tarasov, P.E., Volkova, V.S., Webb III, T., Guiot, J., Andreev, A.A., Bezusko, L.G.,

Bezosko, T.V., Bykova., G.V., Dorofeyuk, N.I., Kvavadze, E.V., Osipova, I.M.,

153

Panova, N.K., Sevastyanov, D.V., 2000. Last Glacial Maximum biomes reconstructed

from pollen and plant macrofossil data from northern Eurasia. Journal of Biogeography

27, 609-620.

Tian, F., Herzschuh, U, Dallmeyer, A., Xu, Q.H., Mischke, S., Biskaborn, B.K., 2013.

Environmental variability in the monsoon-westerlies transition zone during the last 1200

years: lake sediment analyses from central Mongolia and supra-regional synthesis.


Tian, F., Herzschuh, U., Telford, R.J., Mischke, S., Van der Meeren, T., Krengel, M., 2014. A

modern pollen-climate calibration set from central-western Mongolia and its application

to a late glacial-Holocene record. Journal of Biogeography 41, 1909-1922.

Torres, V., Vandenberghe, J., Hooghiemstra, H., 2005. An environmental reconstruction of

the sediment infill of the Bogotá basin (Colombia) during the last 3 million years from

abiotic and biotic proxies. Palaeogeography, Palaeoclimatology, Palaeoecology 226,

127-148.

Tucker, M.E., Wright, V.P., Dickson, J.A.D., 1990. Carbonate sedimentology. Blackwell

Scientific Publications, Oxford, pp. 284-292.

Tudhope, A.W., Chilcott, C.P., McCulloch, M.T., Cook, W.R., Chappell, J., Ellam, R.M.,

Lea, D.W., Lough, J.M., Shimmield, G.B., 2001. Variability in the El Niño-Southern

Oscillation through a Glacial-Interglacial cycle. Science 291, 1511-1517.

Újvári, G., Stevens, T., Svensson, A., Klötzli, U.S., Manning, C., Németh, T., Kovács, J.,

Sweeney, M.R., Gocke, M., Wiesenberg, G.L.B., Markovic, S.B., Zech, M., 2015. Two

possible source regions for Central Greenland last glacial dust. Geophysical Research

Letters, doi: 10.1002/2015GL066153.

Újvári, G., Varga, A., Raucsik, B., Kovács, J., 2014. The Paks loess-paleosol sequence: A

record of chemical weathering and provenance for the last 800 ka in the mid-Carpathian

Basin. Quaternary International 319, 22-37.

Van der Meeren, T., Verschuren, D., Ito, E., Martens, K., 2010. Morphometric techniques

allow environmental reconstructions from low-diversity continental ostracode

assemblages. Journal of Paleolimnology 44, 903-911.

Van Geel, B., Buurman, J., Brinkkemper, O., Schlelvis, J., Aptroot, A., Van Reenen, G.,

Hakbijl, T., 2003. Environmental reconstruction of a roman Period settlement site in

Uitgeest (Netherlands) with special reference to coprophilous fungi. Journal of

Archaeological Science 30, 873-883.

Vandenberghe, J., 2013. Grain size of fine-grained windblown sediment: A powerful proxy

for process identification. Earth-Science Reviews 121, 18-30.

Vandenberghe, J., Renssen, H., van Huissteden, K., Nugteren, G., Konert, M., Lu, H.Y.,

Dodonov, A., Buylaert, J.P., 2006. Penetration of Atlantic westerly winds into Central

and East Asia. Quaternary Science Reviews 25, 2380-2389.

Vanky, K., 1994. European smut fungi. Gustav Fischer Verlag, Stuttgart, New York.

154

Vassallo, R., Jolivet, M., Ritz, J.-F., Braucher, R., Larroque, C., Sue, C., Todbileg, M.,

Javkhlanbold, D., 2007. Uplift age and rates of the Gurvan Bogd system (Gobi-Altay)

by apatite fission track analysis. Earth and Planetary Science Letters 259, 333-346.

Vassallo, R., Ritzm J.-F., Carretier, S., 2011. Control of geomorphic processes on 10Be

concentrations in individual clasts: Complexity of the exposure history in Gobi-Altay

range (Mongolia). Geomorphology 135, 35-47.

Victor, R., Fernando, C.H., 1981. Ilyocypris dentifera Sars, 1903 versus Ilyocypris angulate

Sars, 1903. A taxonomic study of two species of Ilyocypris bradyi and Norman, 1889

(Crustacea, Ostracoda) described from China. Canadian Journal of Zoology 59, 1103-

1110.

Vogt, C., 1997. Regional and temporal variations of mineral assemblages in Arctic Ocean

sediments as climatic indicator during glacial/interglacial changes. Reports of Polar

Research 251, 1-309.

Von Eynatten, H., Barceló-Vidal, C., Pawlowsky-Glahn, V., 2003. Composition and

discrimination of sandstones: a statistical evaluation of different analytical methods.

Journal of Sedimentary Research 73, 47-57.

Von Eynatten, H., Tolosana-Delgado, R., Karius, V., 2012. Sediment generation in modern

glacial settings: Grain-size and source-rock control on sediment composition.

Sedimentary Geology 280, 80-92.

Vriend, M., Prins, M.A., 2005. Calibration of modelled mixing patterns in loess grain-size

distributions: an example from the north-eastern margin of the Tibetan Plateau, China.

Sedimentology 52, 1361-1374.

Walker, R.T., Nissen, E., Molor, E., Bayasgalan, A., 2007. Reinterpretation of the active

faulting in central Mongolia. Geology 35, 759-762.

Walker, R.T., Waugh, B., Grone, A.J., 1978. Diagenesis in first-cycle desert alluvium of

Cenozoic age, southern United States and northwestern Mexico. Geological Society of

America Bulletin 89, 19-32.

Walther, M., 1998. Paläoklimatische Untersuchungen zur jungpleistozänen

Landschaftsentwicklung im Changai-Bergland und in der nördlichen Gobi (Mongolei).

Petermanns Geographische Mitteilungen 142, 207-217.

Wang, G.X., Cheng, G.D., 1999. Water resource development and its influence on the

environment in arid areas of China – the case of the Hei River Basin. Journal of Arid

Environments 43, 121-131.

Wang, N.A., Li, Z.L., Cheng, H.Y., Li, Y., Huang, Y.Z., 2011. High lake levels on Alashan

Plateau during the Late Quaternary. Chinese Science Bulletin 56, 1367-1377.

Wang, W., Feng, Z.D., 2013. Holocene moisture evolution across the Mongolian Plateau and

its surrounding areas: A synthesis of climatic records. Earth-Science Reviews 122, 38-

57.

Wang, X.M., Xia, D.S., Zhang, C.X., Lang, L.L., Hua, T., Zhao, S., 2012. Geochemical and

magnetic characteristics of fine-grained surface sediments in potential dust source areas:

155

Implications for tracing the provenance of aeolian deposits and associated

palaeoclimatic change in East Asia. Palaeogeography, Palaeoclimatology,

Palaeoecology 323-325, 123-132.

Wang, Y.B., Liu, X.Q., Herzschuh, U., 2010. Asynchronous evolution of the Indian and East

Asian Summer Monsoon indicated by Holocene moisture patterns in monsoonal central

Asia. Earth-Science Reviews 103, 135-153.

Webb, L.E:, Johnson, C.L., 2006. Tertiary strike-slip faulting in southeastern Mongolia and

implications for Asian tectonics. Earth and Planetary Science Letters 241, 323-335.

Weltje, G.J., Tjallingii, R., 2008. Calibration of XRF core scanner for quantitative

geochemical logging of sediment cores: Theory and application. Earth and Planetary

Science Letters 274, 423-438.

Whitlock, C., Bartlein, P.J., 1997. Vegetation and climate change in northwest America

during the past 125 k. Nature 388, 57-61.

Williams, D.F., Peck, J., Karabanov, E.B., Prokopenko, A.A., Kravchinsky, V., King, J.,

Kuzmin, M.I., 1997. Lake Baikal record of continental insolation during the past 5

million years. Science 278, 1114-1117.

Windley, B.F., Allen, M.B., 1993. Mongolian plateau: evidence for a late Cenozoic mantle

plume under central Asia. Geology 21, 295-298.

Wirth, S.B., Gilli, A., Niemann, H., Dahl, T.W., Ravasi, D., Sax, N., Hamann, Y., Peduzzi,

R., Peduzzi, S., Tonolla, M., Lehmann, M.F., Anselmetti, F.S., 2013. Combining

sedimentological, trace metal (Mn, Mo) and molecular evidence for reconstructing past

water-column redox conditions: The example of meromictic Lake Cadagno (Swiss

Alps). Geochimica et Cosmochimica Acta 120, 220-238.

Wu, B.Y., Wang, J., 2002. Winter Arctic Oscillation, Siberian High and East Asian Winter

Monsoon. Geopgysical Research Letters 29, 1897, doi: 10.1029/2002GL015373.

Wünnemann, B., 1999. Untersuchungen zur paläohydrographie der endseen in der Badain

Jaran und Tennger Wüste, Innere Mongolei, Nordwest China. Habilitation Thesis, Freie

Universität Berlin, Berlin, Germany.

Wünnemann, B., Demske, D., Tarasov, P., Kotlia, B.S., Reinhardt, C., Bloemendal, J.,

Diekmann, B., Hartmann, H., Krois, J., Riedel, F., Arya, N., 2010. Hydrological

evolution during the last 15 k in the Tso Kar lake basin (Ladakh, India), derived from

geomorphological, sedimentological and palynological records. Quaternary Science

Reviews 29, 1138-1155.

Wünnemann, B., Hartmann, K., Altmann, N., Hambach, U., Pachur, H.-J., Zhang, H.C., 2007.

Interglacial and glacial fingerprints from lake deposits in the Gobi Desert, NW China.

In: Sirocko, F., Claussen, M., Sánchez-Goñi, M.F., Litt, T., (Eds.). The Climate of Past

Interglacials. Development in Quaternary Sciences 7, 323-347.

Wünnemann, B., Pachur, H.-J., 1998. Climatic and environmental changes in the deserts of

Inner Mongolia, China, since the Late Pleistocene. In: Alsharan, A.S., Glennie, K.W.,

Whittle, G.L., (Eds.). Quaternary Deserts and Climatic Change, Balkema, Rotterdam,

pp. 381-394.

156

Xiao, G.Q., Zong, K.Q., Li, G.J., Hu, Z.C., Dupont-Nivet, G., Peng, S.Z., Zhang, K.X., 2012.

Spatial and glacial-interglacial variations in provenance of the Chinese Loess Plateau.

Geophysical Research Letters 39, doi:10.1029/2012GL053304.

Xiao, J.L., Inouchi, Y., Kumai, H., Yoshikawa, S., Kondo, Y., Liu, T.S., An, Z.S., 1997.

Eolian quartz flux to Lake Biwa, central Japan, over the past 145,000 years. Quaternary

Research 48, 48-57.

Xiao, J.L., Xu, Q.H., Nakamura, T., Yang, X.L., Liang, W.D., Inouchi, Y., 2004. Holocene

vegetation variation in the Daihai Lake region of north-central China: a direct indication

of the Asian monsoon climatic history. Quaternary Science Reviews 23, 1669-1679.

Xu, X.K., Yang, J.Q., Dong, G.C., Wang, L.Q., Miller, L., 2009. OSL dating of glacier extent

during the Last Glacial and the Kanas Lake basin formation in Kanas River valley, Altai

Mountains, China. Geomorphology 112, 306-317.

Xu, Z.W., Lu, H.Y., Zhao, C.F., Wang, X.Y., Su, Z.Z., Wang, Z.T., Liu, H.Y., Wang, L.X.,

Lu, Q., 2011. Composition, origin and weathering process of surface sediment in

Kumtagh Desert, Northwest China. Journal of Geographical Sciences 21, 1062-1076.

Yancheva, G., Nowaczyk, N.R., Mingram, J., Dulski, P., Schettler, G., Negendank, J.F.W.,

Liu, J.Q., Sigman, D.M., Peterson, L.C., Haug, G.H., 2007. Influence of the

intertropical convergence zone on the East Asian Monsoon. Nature 445, 74-77.

Yang, X.P., Scuderi, L.A., 2010. Hydrological and climatic changes in deserts of China since

the late Pleistocene. Quaternary Research 73, 1-9.

Yang, B., Shi, Y.F., Braeuning, A., Wang, J.X., 2004. Evidence for a warm-humid climate in

arid northwestern China during 40-30 ka BP. Quaternary Science Reviews 23, 2537-

2548.

Yang, S.L., Ding, F., Ding, Z.L., 2006. Pleistocene chemical weathering history of Asian arid

and semi-arid regions recorded in loess deposits of China and Tajikistan. Geochimica et

Cosmochimica Acta 70, 1695-1709.

Yang, S.Y., Jung, H.-S., Li, C.X., 2004. Two unique weathering regimes in the Changjiang

and Huanghe drainage basins: geochemical evidence from river sediments. Sedimentary

Geology 164, 19-34.

Yang, X.P., Rost, K.T., Lehmkuhl, F., Zhu, Z.D., Dodson, J., 2004. The evolution of dry

lands in northern China and in the Republic of Mongolia since the Last Glacial

Maximum. Quaternary International 118/119, 69-85.

Yang, X.P., Scuderi, L.A., 2010. Hydrological and climatic changes in deserts of China since

the late Pleistocene. Quaternary Research 73, 1-9.

Yang, X.P., Scuderi, L., Paillou, P., Liu, Z.T., Li, H.W., Ren, X.Z., 2011. Quaternary

environmental changes in the drylands of China – A critical review. Quaternary Science

Reviews 30, 3219-3233.

Yang, X.P., Zhu, B.Q., White, P.D., 2007. Provenance of aeolian sediment in the Taklamakan

Desert of western China, inferred from REE and major-elemental data. Quaternary


157

Young, G., Nesbitt, H., 1998. Processes controlling the distribution of Ti and Al in

weathering, siliciclastic sediments and sedimentary rocks. Journal of Sedimentary

Research 68, 448-455.

Yu, G., Cui, F., Shi, Y.F., Zheng, Y., 2007. Late marine isotope stage 3 paleoclimate for East

Asia: a data-modal comparison. Palaeogeography, Palaeoclimatology, Palaeoecology

250, 167-183.

Yu, K.F., Hartmann, K., Nottebaum, V., Stauch, G., Lu, H.Y., Zeeden, C., Yi, S.W.,

Wünnemann, B., Lehmkuhl, F., 2016. Discriminating sediment archives and

sedimentary processes in the arid endorheic Ejina Basin, NW China using a robust

geochemical approach. Journal of Asian Earth Sciences 119, 128-144.

Yu, K.F., Lehmkuhl, F., Falk, D., 2017. Quantifying land degradation in the Zoige Basin, NE

Tibetan Plateau using satellite remote sensing data. Journal of Mountain Science 14, 77-

93.

Yu, K.F., Lu, H.Y., Lehmkuhl, F., Nottebaum, V., 2013. A preliminary quantitative

paleoclimate reconstruction of the dune fields of northern China during the Last Glacial

Maximum and Holocene optimum. Quaternary Sciences 33, 293-302.

Yu, Z.T., Liu, X.Q., Wang, Y., Chi, Z.Q., Wang, X.J., Lan, H.Y., 2014. A 48.5-ka climate

record from Wulagai Lake in Inner Mongolia, Northeast China. Quaternary


Zhai, D.Y., Xiao, J.L., Zhou, L., Wen, R.L., Chang, Z.G., Pang, Q.Q., 2010. Similar

distribution pattern of different phenotypes of Limnocythere inopinata (Baird) in a

brackish-water lake in Inner Mongolia. Hydrobiologia 651, 185-197.

Zhai, X.W., 2008. The Holocene climate and environment changes in the Mongolian Plateau.

Dissertation to Lanzhou University, Lanzhou.

Zhang, C.J., Zhang, W.Y., Feng, Z.D., Mischke, S., Gao, X., Gao, D., Sun, F.F., 2012.

Holocene hydrological and climatic change on the northern Mongolian Plateau based on

multi-proxy records from Lake Gun Nuur. Palaeogeography, Palaeoclimatology,

Palaeoecology 323-325, 75-86.

Zhang, H.C., Ming, Q.Z., Lei, G.L., Zhang, W.X., Fan, H.F., Chang, F.Q., Wünnemann, B.,

Hartmann, K., 2006. Dilemma of dating on lacustrine deposits in a hyperarid inland

basin of NW China. Radiocarbon 48, 219-226.

Zhang, Y.Y., 2014a. Robustfa: An object oriented solution for robust factor analysis. R

Package Version 1.0-5. http://CRAN.R-project.org/package=robustfa.

Zhang, Y.Y., Pe-Piper, G., Piper, D.J.W., 2014b. Sediment geochemistry as a provenance

indicator: Unrevealing the cryptic signatures of polycyclic sources, climate changes,

tectonism and volcanism. Sedimentology 61, 383-410.

Zhao, J.D., Yin, X.F., Harbor, J.M., Lai, Z.P., Liu, S.Y., Li, Z.Q., 2013. Quaternary glacial

chronology of the Kanas River valley, Altai Mountains, China. Quaternary


http://cran.r-project.org/package=robustfa

158

Zhao, Y., Yu, Z.C., Chen, F.H., Li, J.J., 2008. Holocene vegetation and climate change from a

lake sediment record in the Tengger Sandy Desert, northwest China. Journal of Arid

Environments 72, 2054-2064.

Zhao, Y., Liu, H.Y., Li, F.R., Huang, X.Z., Sun, J.H., Zhao, W.W., Herzschuh, U., Tang, Y.,

2012. Application and limitations of the Artemisia/Chenopodiaceae pollen ratio in arid

and semi-arid China. The Holocene 22, 1385-1392.

Zhu, B.Q., Yu, J.J., Rioual, P., Gao, Y., Zhang, Y.C., Min, L.L., 2014. Geomorphoclimatic

characteristics and landform information in the Ejina Basin, Northwest China.

Environmental Earth Sciences, doi: 10.1007/s12665-014-3927-9.

159

Appendix

Tab. A1: Ratios from major elements to oxides and corresponding functions.

Oxides Name Ratios

Na2O Sodium Oxide 1.3480

MgO Periclase 1.6583

Al2O3 Corundum 1.8895

SiO2 Quartz 2.1393

P2O5 Phosphorus Pentoxide 2.2914

K2O Potassium Oxide 1.2046

CaO Calcium Oxide 1.3992

TiO2 Rutile 1.6685

MnO Manganosite 1.2912

Fe2O3 Hematite 1.4297

𝑏𝑛 = 𝑎𝑛 ∗ r𝑛/104 Eq. (A.1)

𝑏𝑛′ = 𝑏𝑛 ∗ 100/ ∑ 𝑏𝑛

10𝑛=1 Eq. (A.2)

In Eq. (A.1), a denotes original major elements abundance in ppm. b denotes oxide data. r is ratio in

above table. n ranges from 1 to 10, i.e., ten major elements. respectively. In Eq. (A.2), bn’ is the

normalized oxides in wt %, which is used in the multivariate statistics.

160

R script employed in the dissertation

# this script is part of the manuscript ‘Discriminating sediment archives and sedimentary

processes in the arid endorheic Ejina Basin, NW China using a robust geochemical approach'

by Yu et al. in Journal of Asian Earth Science, and manuscript ‘Geochemical imprints of

coupled paleoenvironmental and provenance change in the lacustrine sequence of Orog Nuur,

Gobi Desert of Mongolia’ by Yu et al.

# load the 'boot' package, which must be installed

library(boot)

library(Hmisc)

# read in element data from clipboard, as in S3.

data<-read.table("clipboard",dec=",",header=TRUE,sep='\t')

# row names of archive types and sequence number

names1<-

c("A1","L1","F1","A2","A3","A4","A5","A6","A7","A8","A9","A10","A11","A12","A13","

A14","A15","F2","L2","A16","A17","A18","A19","A20","A21","L3","L4","A22","A23","A2

4","A25","A26","A27","A28","A29","F18","L5","A30","A31","A32","G1","G2","L6","L7","

L8","L9","F3","F4","F5","F6","A33","L10","S1","F7","L11","S2","L12","L13","F8","F9","L

14","L15","F10","L16","L17","F11","F12","A34","F13","F14","F15","F16","F17","A35","A

36","A37","A38","L18","L19","A39","A40","A41","A42","G3","A43","A44","A45","A46","

A47","A48","A49","A50","A51")

# column names of elements (raw data set)

names2<-

c("Na","Mg","Al","Si","P","K","Ca","Mn","Fe","Ti","S","Cl","V","Cr","Ni","Cu","Zn","Ga",

"As","R

b","Sr","Y","Zr","Nb","Ba","Nd","Hf","W","Pb","Th")

rownames(data)=make.names(names1,unique=TRUE)

colnames(data)=make.names(names2)

data # shows the data matrix with specific row names

# read in grain size AND element data from clipboard, as in S3.

data_all<-read.table("clipboard",dec=",",header=FALSE,sep='\t',stringsAsFactors = FALSE)

data_all <- do.call(rbind, data_all)

data_all2 <- apply(data_all, 1,as.numeric)

data_all2 <- t(as.numeric(data_all))

#[R,P] = corr(data_all, data_all)

Cor<-rcorr(data_all2) # generate correlation matrix and corresponding P values

dlmwrite('R1.txt',Cor$r)

dlmwrite('P1.txt',P) # imported into Excel>>R<-read.table("clipboard")

161

P<-read.table("clipboard")

mysub<-function(x){sub(",",".",x)} # create an function to replace "," with "."

dataR<-(apply(R,2,mysub))

R1<-data.frame(apply(dataR,2,as.numeric)) #change the matrix as numeric

dataP<-(apply(P,2,mysub))

P1<-data.frame(apply(dataP,2,as.numeric))

R<-R1

P<-P1

data_matrix<-(R)

data_matrix2<-matrix(nrow=length(P),ncol=length(P))

for(C in 1:40) # 40 can be dim(data[40])

{

for(Ro in 1:40)

{

if(P[Ro,C]<0.05){data_matrix2[Ro,C]<-data_matrix[Ro,C]}

# set criterion: P<0.05, can be set different, criterion here, >.5, this need to be improved later

}}

data_heatmap<-

heatmap(data_matrix2,Rowv=NA,Colv=NA,col=topo.colors(256),scale="column",margins=c

(5,10))>

install.packages("lattice")

library(lattice)

levelplot(data_matrix2,col.regions=topo.colors(256))

circle<-function(center=c(0,0),npoints=100){r=1

+ tt=seq(0,2*pi,length=npoints)

+ xx=center[1]+r*cos(tt)

+ yy=center[1]+r*sin(tt)

+ return(data.frame(x=xx,y=yy))

+ }

corcir=circle(c(0,0),npoints=100)

loadings_16<-as.data.frame(vRFs)

arrows=data.frame(x1=c(0,0,0,0),y1=c(0,0,0,0),x2=loadings_16$vRF1,y2=loadings_16$vRF

2)

ggplot()

+geom_path(data=corcir,aes(x=x,y=y))

+geom_segment(data=arrows,aes(x=x1,y=y1,xend=x2,yend=y2))

162

+geom_text(data=loadings_16,aes(x=vRF1,y=vRF2,label=rownames(loadings_16)))+geom_h

line(yint

ercept=0)

+geom_vline(xintercept=0)

+labs(x="Robust Factor 1",y="Robust Factor 2")

+ggtitle("Loadings Plot") # loadings plot for robust data

>robustfa_30<-

factorScorePfa(sum,factors=4,covmat=covwt,cor=TRUE,rotation="varimax",scoresMethod=c

("regres

sion")

>rownames(robustfa_30$loadings)=make.names(names2,unique=TRUE)

>scores_30=as.data.frame(robustfa_30$scores)

>ggplot(data=scores_30,aes(x=Factor1,y=Factor2,label=rownames(scores_30)))+geom_hline

(yinterce

pt=0)+geom_vline(xintercept=0)+geom_text(colour="tomato",alpha=0.8,size=4)+ggtitle("Sc

ores

Plot") # scores plot

>arrows_30=data.frame(x1=c(0,0),y1=c(0,0),x2=loadings_30$RF1_30,y2=loadings_30$RF2

_30)

>ggplot()+geom_path(data=corcir,aes(x=x,y=y))+geom_segment(data=arrows_30,aes(x=x1,y

=y1,xen

d=x2,yend=y2))+geom_text(data=loadings_30,aes(x=RF1_30,y=RF2_30,label=rownames(lo

adings_3

0)))+geom_hline(yintercept=0)+geom_vline(xintercept=0)+labs(x="Robust Factor

1",y="Robust

Factor 2")+ggtitle("Loadings Plot") # loadings plot for raw data

>RF1<-robustfa_16$scores[,1]

>RF2<-robustfa_16$scores[,2]

>labs_robustfa<-rownames(robustfa_16$scores)

>RF_scores<-data.frame(cbind(RF1,RF2))

>rownames(RF_scores)<-labs_robustfa

>ggplot(RF_scores,aes(RF1,RF2,label=rownames(RF_scores)))+geom_text()

>scores_16=as.data.frame(robustfa_16$scores)

>ggplot(data=scores_16,aes(x=Factor1,y=Factor2,label=rownames(scores_16)))+geom_hline

(yinterce

pt=0)+geom_vline(xintercept=0)+geom_text(colour="tomato",alpha=0.8,size=4)+ggtitle("Sc

ores

Plot") # scores plot for robust data

163

Acknowledgements

In 2012, the author got the chance to visit RWTH and started in 2013 with PhD works here. It

is great honor to spend three years’ time in Aachen to pursue knowledge concerning past

environmental changes and mechanisms. The author extends his sincere thanks to the

financial support from German Academic Exchange Service (DAAD) and China Scholarship

Council (CSC) through the Master and PhD studies. On the other hand, words cannot express

gratitude for all the support from advisor Prof. Frank Lehmkuhl, institute, colleagues, friends

and family members. So many people have contributed to the accomplishment of the

dissertation, making it seems impossible to name all. The author specially thanks:

The RWTH group:

First, the author want to express his deep grateful to advisor, Prof. Frank Lehmkuhl,

who make it possible for this PhD work and provides freedom to follow the author’s

own ideas and the training opportunities in summer schools, workshops and

conferences; the author also thanks Mrs. Carola Lehmkuhl, for the delicious food

during every year’s Christmas and nice memories of watching matches in the Stadium-

Tivoli of Alemannia Aachen;

Thanks for the accompanying through the field, lab works, manuscripts

writing/revision and daily life from Dr. Veit Nottebaum, Dr. Georg Stauch, Dr. David

Loibl, Dr. Christian Zeeden, Igor Obreht, Philipp Schulte, and Dr. Andreas

Rudersdorf; the author acknowledge deep grateful to Prof. Wolfgang Römer, Dr.

Martin Knippertz, Prof. Christoph Hilgers, Prof. Klaus Reicherter, Prof. Eileen

Eckmeier, Marianne Dohms, Lydia Krauß, Janina Bösken, Mona Ziemes, Anja Knops,

Verena Niedek, Isabel Pipaud, Bernhard Pröschel, Jörg Zens, Michael Buchty-Lemke,

for their kind help during the PhD works;

The author sincerely appreciate Dr. Eva Huintjes, Bastian Paas, Isabell Maras, René

Delpy, Dimitri Falk, Vera Schleiden, Annika Hörstmann-Jungemann, Matthias

Mensing, René Kocgazi, Gernot Frommelt, Hans-Joachim Ehrig and Dr. Gunnar

Ketzler, for accompanying in the Mensa, party and beer drinking.

The AWI group:

Prof. Berhard Diekmann has kindly supervised the PhD work regarding the bulk-

geochemistry in the lacustrine realm and corresponding paleoenvironmental

implications; Stefan Schimpf offered geological mapping of the Gaxun Nuur Basin.

The Mainz group:

Prof. Jörg Grunert has kindly provided his rich experience concerning the

geomorphological characteristics in the Valley of Gobi Lakes, Mongolia and the Gobi

Desert, and reviewed manuscripts about the Orog Nuur drilling core.

The Nanjing group:

The author’s Master advisor Prof. Huayu Lu, Prof. Bernd Wünnemann, Chuanbing

Yang and Shuangwen Yi have kindly supported and organized the OSL dating in

Nanjing lab and manuscripts writing/revision.

The Berlin group:

Dr. Kai Hartmann provides state-of-art techniques concerning the multivariate

statistics; J. Bölscher has kindly mailed the HC8 sediment samples; Dr. Gregori

Lockot provides vital information regarding reservoir effects in the lacustrine realm.

164

The NIHK/Göttingen group:

Dr. Frank Schlütz and Dr. Waheed Murad have kindly conducted the palynological

analysis and helped with the manuscript writing and revision, rendering the possibility

and the outline of the dissertation.

The Reykjavík/Potsdam group:

Prof. Stefan Mischke has conducted ostracod analysis in the dissertation and offered

pivotal experiences about scientific editing and reviewing.

The Wisconsin group:

Prof. Joseph Mason provides significant comments about the oral talk during the EGU

2016 and has kindly reviewed the manuscript.

The Xi’an group:

Prof. Youbin Sun has kindly reviewed the manuscript and offered helpful comments

during the INQUA 2015.

The Xining/Wuhan group:

Prof. Zhongping Lai provides significant remarks concerning the first manuscript

during the 2014 workshop in Nanjing.

The Vienna group:

Prof. Peter Filzmoser has kindly helped with the R scripts programming and

multivariate statistics.

Conferences and workshops:

ICAR VIII 2014: Prof. Jef Vandenberghe, Prof. Xiaoping Yang, Dr. Jan-Berend

Stuut, Dr. Jianbao Liu, Dr. Runqiang Zeng, Shibo Tuo, Dr. Zhiwei Xu, Han Feng,

Mengchun Cui, Dr. Yali Zhou and Prof. Hanchao Jiang provided pivotal remarks

concerning the oral talk and accompanied the impressive excursion to the Tengger

Desert;

AK Wüstenrandforschung 2014: Andrea Junge, Dr. Johanna Lomax provided their

interesting experiences regarding OSL dating in arid realm and its combination with

archaeological studies in the Middle East;

INQUA 2015: Prof. Zhaodong Feng, Prof. Jule Xiao, Prof. Slobodan Marković, Dr.

Sumiko Tsukamoto, Dr. Thomas Stevens, Dr. Jingran Zhang, Dr. Xiaozhong Huang

and Dr. Xin Jia has provided kindly comments regarding the oral talk and manuscripts;

And the author would like to express special thanks to Prof. Zhaodong Feng, for his

kind offering the book that reviews his career’s work in Mongolia: “Vegetation

changes and associated climate variations in the Mongolian Plateau during the past

40,000 years”;

2015 TPE & TiP Science & Technology Training School: Prof. Fan Zhang, Prof.

Erwin Appel, Prof. Antje Schwalb, Prof. Shichang Kang, all the lecturers and

classmates have given a unforgettable experiences in Alpine ecosystem and the

Yulong Glacier; the author especially thank the financial support from the German

Research Foundation (DFG) and Chinese Academy of Science (CAS);

Junge Geomorphologen 2015: Julia Meister, Dr. Sabine Kraushaar, Dr. Jan Blöthe,

Raphael Steup and Simon Meyer-Heintze have kindly helped the oral talk and

provided useful suggestions concerning the first manuscript;

165

EGU 2016: It my first time to take part in such sort of wide range subject conference,

Prof. Dominik Fleitmann, Prof. Markus Fuchs, Prof. Hans von Suchodoletz, Prof.

Achim Brauer, Prof. Fahu Chen, Prof. Xiaohua Gou and Dr. Gaojun Li have greatly

helped concerning the oral talk and manuscript revision; Special thanks are given to

the accompanying friends: Igor Obreht, Lydia Krauß, Janina Bösken and Han Feng;

As well, the author express his sincere thanks to the financial support from the AK

Geomorphologie (Prof. Markus Fuchs, Der Deutsche Arbeitskreis für

Geomorphologie);

USSP 2016 & ECORD Training Course 2017 (MARUM): Thanks Prof. Frank

Lehmkuhl, Dr. Christian Zeeden, and Igor Obrecht for the supporting to attend the

outstanding summer school and training courses about the state-of-the-art of the

coupled paleobiogeochemistry and climate modelling. It really extends the author’s

knowledge to a brand new level regarding the Cenozoic scale climate change and

proxies interpretations.

For the manuscripts:

Acknowledgements for the manuscript: “Discriminating sediment archives and

sedimentary processes in the arid endorheic Ejina Basin, NW China using a

robust geochemical approach” (Chapter 4): The study was funded by German

Federal Ministry of Education and Research (BMBF-CAME: 03G0814A), National

Natural Science Foundation of China (41371203, 41321062) and China Scholarship

Council (201306190112). Dr. C. Zeeden is supported by German Research Foundation

(DFG: SFB806). We gratefully thank Prof. C. Hilgers, Prof. W. Römer, I. Obreht, A.

Rudersdorf, P. Schulte, G. Lockot, N. Herrmann, S. Schimpf, Dr. D. Loibl, Dr. S.

Kraushaar and Dr. E. Eckmeier for constructive discussions about the methodology

and manuscript, Prof. Z.-P. Lai, Prof. F.-H. Chen, and Prof. K. Reicherter contributed

to the discussion. Prof. P. Filzmoser helped with the R calculation. J. Bölscher offered

the data of HC8 section, colleagues from Nanjing Uni., FU Berlin, AWI Potsdam and

Lanzhou Uni. helped during the field investigation, M. Dohms and C.-B. Yang are

acknowledged for parts of Lab work. Anonymous reviewers are sincerely appreciated

for improving the manuscript.

Acknowledgements for the manuscript: “Lake sediments documented late

Quaternary humid pulses in the Gobi Desert of Mongolia: Vegetation, hydrologic

and geomorphic implications” (Chapter 5): This study was funded by the German

Research Foundation (LE 730/16-1: Late Pleistocene, Holocene and ongoing

geomorpho-dynamics in the Gobi Desert, South Mongolia) and China Scholarship

Council (201306190112). The fieldwork was supported by the Institute of Geography

of the Mongolian Academy of Science (Prof. D. Dorjgotov, A. Tschimegsaichan). Dr.

T. Felauer, Dr. A. Hilgers and Dr. D. Hülle helped the field work. Palynological

analysis and radiocarbon dating were financed by a scholarship issued to Dr. W.

Murad. I. Pipaud offered the base map for geologic mapping. Prof. H. Rother offered

information concerning the paleoglaciation in the Khangai Mountains. We sincerely

appreciate the constructive discussions with Prof. F.-H. Chen, Prof. D. Fleitmann,

Prof. H.J. Spero, Prof. G.R. Dickens, Dr. S.R. Meyers, Prof. H.-Y. Lu, Prof. J.A.

Mason, Prof. Z.-D. Feng, Prof. J.-L. Xiao, I. Obreht, P. Schulte, Dr. C. Zeeden, Dr. F.

Muschitiello and Dr. J.-B. Liu during the INQUA 2015, EGU 2016 and USSP 2016.

Acknowledgments for the manuscript: “Geochemical imprints of coupled

paleoenvironmental and provenance change in the lacustrine sequence of Orog

Nuur, Gobi Desert of Mongolia” (Chapter 6): This study was funded by the German

Research Foundation (LE 730/16-1) and China Scholarship Council (201306190112).

166

Fieldwork was supported by the Institute of Geography of the Mongolian Academy of

Sciences (D. Dorjgotov, A. Tschimegsaichan). Radiocarbon dating was financed by a

scholarship issued to W. Murad. I. Pipaud helped with the geologic mapping. F.

Schlütz and T. Felauer supported the field work. We sincerely appreciate the

constructive discussion with S. Mischke, J. Grunert, F.H. Chen, A.C.G. Henderson,

H.Y. Lu, Z.D. Feng, J.L. Xiao, D. Fleitmann, P. Schulte, and I. Obreht during the

INQUA 2015 and EGU 2016. T.J. Whitmore, C. Zhao, M. Brenner, M. Riedinger-

Whitmore, and two anonymous reviewers are sincerely appreciated for improving the

manuscript.

Also, special thanks to friends who have visited the author during these three years: Dr.

Wentao Yang, Ning Ma, Tao Chen, Achyut Tiwari, Dr. Xianyan Wang, Hanzhi Zhang, Simon

Wittich and Chi Zhang. Thanks for the accompanying of Miaoqun Yin, Zhengxin Pu and

Nian Zhang. In the end, the author wants to acknowledge debt to his parents Zhongxiang Yu,

Lanying Zhang, dear Chathrin and kinsfolk, without whose support, assistance and patience,

this endeavor would not have been possible.

167

Abstract

Considerable efforts have been devoted to disentangle the late Quaternary moisture and

thermal evolution of arid central Asia. However, an array of paramount aspects has inhibited

our complete understanding of the broad pattern of Quaternary moisture and thermal history

of arid central Asia and underlying mechanisms.

The Ejina Basin, with its suite of different sediment archives, is known as one of the main

sources for the loess accumulation on the Chinese Loess Plateau. In order to understand

mechanisms along this supra-regional sediment cascade (aeolian, fluvial and alluvial

sediments), it is crucial to decipher the archive characteristics and formation processes. Five

sediment archives from three lithologic units exhibit geochemical characteristics as follows:

(i) aeolian sands have high contents of Zirconium and Hafnium, whereas only Hafnium can be

regarded as a valuable indicator to discriminate the coarse sand proportion; (ii) sandy loess

has high Calcium and Strontium contents which both exhibit broad correlations with the

medium to coarse silt proportions; (iii) lacustrine clays have high contents of felsic,

ferromagnesian and mica source elements e.g., Potassium, Iron, Titanium, Vanadium, and

Nickel; (iv) fluvial sands have high contents of Magnesium, Chlorine and Sodium which may

be enriched in evaporite minerals; (v) alluvial gravels have high contents of Chromium which

may originate from nearby Cr-rich bedrock. Temporal variations can be illustrated by four

robust factors: weathering intensity, silicate-bearing mineral abundance, saline / alkaline

magnitude and quasi-constant aeolian input. In summary, the bulk-composition of the late

Quaternary sediments in this arid context is governed by the nature of the source terrain, weak

chemical weathering, authigenic minerals, aeolian sand input, whereas pedogenesis and

diagenesis exert only limited influences. Hence, here demonstrates a practical geochemical

strategy to discriminate sediment archives and thereafter enhance our ability to offer more

intriguing information about the sedimentary processes in the arid central Asia.

On the other hand, two parallel cores (ONW I, 6.00 m; ONW II, 13.35 m) were retrieved from

lake Orog Nuur, in the Gobi Desert of Mongolia. Ample evidences reveal a marked moisture

pulse during the Marine Isotope Stage 3 (~36-~24 ka) which might have induced the

maximum last glacial expansion in the high elevated Khangai Mountains. A sharp transition

of Termination I (~11 ka) is illuminated by geochemical, palynological, and ostracod data.

Lower area of the Orog Nuur catchment was dominated by Artemisia steppe community in

the late Pleistocene and altered gradually to Chenopodiaceae desert steppe in the Holocene.

The early Holocene is also characterized by relatively humid environment, albeit discordant

168

downcore variability of moisture and thermal signals can be derived between palynological

and bulk-geochemical signals. Water body in the Holocene appears to be a distinct alkaline

environment which was subjected to frequent allogenic input and disturbance of the late

Pleistocene anoxic states. These two humid pulses may be the trait of a larger scale of arid

central Asia that would be documented in a suite of archives. Considering kinetics, a coupled

atmospheric component comprising both East Asian Summer Monsoon and strengthened

Westerlies moisture supply seems to be the driving mechanism, and this coupled mode might

have been modulated broadly by boreal insolation variances. On the other hand, four major

harsh climatic phases were documented in the core at ~47 ka, ~37 ka, ~19 ka (gLGM) and

~13-~11 ka (Younger Dryas) as playa phases. Reduced conveying of the Westerlies moisture

along with the retreated East Asian Summer Monsoon might have contributed to these playa

phases in the Gobi Desert.

Continuous, high-resolution elemental abundances at a 1 cm interval were examined on core

ONW II. Due to the predominant clay or silty-clay fraction in the sediments, Aluminium and

Silicon display broadly identical pattern. Calcium behaviors may be ascribed to the authigenic

calcite variance. Manganese and Cobalt act as sound indicator for the redox condition. Owing

to the short burial time and weak diagenetic influence, Iron is not an indicator to the

diagenetic processes. Likewise, Zirconium may be associated to fluvial clastic in relation to

the lithotypes but not necessarily linked to the aeolian sands relative to that in the pelagic

realm. Furthermore, Sulfur in lake sediments may denote the redox condition and Potassium

is more likely linked to the K-feldspar which is associated with the allogenic fluvial inputs.

The two sand layers corresponding to LGM and YD event were dominated by coarse sands.

In the Gobi Desert of Mongolia, these two laminations provide an opportunity to be regarded

as potential chronological benchmarks. As inferred from the biplot between Ca and Al/Si,

disparate source lithotypes may exist before and after Termination I. The Holocene appears to

be a distinct alkaline environment compared with the late Pleistocene. This may be ascribed to

strengthened fluvial morphodynamics of the riverine inflows and intensified erosion of source

rocks in the upper catchment of the Orog Nuur. In exceptionally arid realm, sands may exert

significant influence on bulk-geochemistry of the lake sediments. In summary, considering

the bulk-geochemistry of lacustrine sediments, paleoclimate would not be the solely

governing factor, the coupled alteration of precursor rocks and decomposition intensity may

also exert pivotal influences.

169

Zusammenfassung

Beachtliche Versuche sind unternommen worden, um die hygrische und thermische

Entwicklung des ariden Zentralasiens im Spätquartär zu entschlüsseln. Allerdings sind das

allgemeine Schema der hygrischen und thermischen Entwicklung des ariden Zentralasiens im

Quartär sowie die zugrundeliegende Mechanismen noch nicht umfassend verstanden.

Das Ejina-Becken in der Inneren Mongolei (China) hat unterschiedliche Sedimentarchive und

ist als eine der Hauptquellen für die Lössakkumulation auf dem Chinesischen Lössplateau

bekannt. Um die Mechanismen entlang der überregionalen Sedimentkaskaden zu verstehen,

ist es notwendig, die Archiveigenschaften und Bildungsprozesse zu entschlüsseln. Um diese

Fragen zu beantworten wurden fünf Profile in unterschiedlichen geomorphologischen

Kontexten ausgewählt. Die Untersuchungsergebnisse zeigen, dass die fünf Sedimentarchive

aus drei lithologischen Einheiten folgende geochemische Charakteristika aufweisen: (i)

äolische Sande haben höhere Zirconium- und Hafnium-Gehalte, wobei nur Hafnium als

verwertbarer Indikator für den Grobsandanteil betrachtet werden kann; (ii) sandiger Löss hat

höhere Kalzium- und Strontium-Gehalte, welche beide eine deutliche Korrelation mit Mittel-

und Grobschluffanteilen aufweisen; (iii) lakustrische Tone haben höhere Gehalte an

felsischen, ferromagnesiumhaltigen und Glimmer-Ausgangselementen, wie z.B. Kalium,

Eisen, Titan, Vanadium, und Nickel; (iv) fluviale Sande haben höhere Magnesium-, Chlor-

und Natrium-Gehalte, welche in Evaporitmineralen angereichert sein können; (v) alluviale

Kiese haben höhere Chrom-Gehalte, welche von nahe gelegenen Chrom-reichen Gestein

stammen könnten. Zeitliche Variationen können anhand von vier Faktoren veranschaulicht

werden: Verwitterungsintensität, Anteil an Silikathaltigen Mineralien, salzhaltiger Magnitude

und Ausmaß an quasi-konstantem äolischem Eintrag. Zusammenfassend kann festgehalten

werden, dass die Zusammensetzung der Sedimente aus dem späten Quartär in diesem ariden

Kontext durch die Reliefbedingungen der Sedimentquelles, durch eine schwache chemische

Verwitterung, authigener Minerale sowie äolischen Sandeintrags bestimmt werden.

Pedogenese und Diagenese üben nur einen geringen Einfluss aus.

Zwei parallele Kernbohrungen (ONW I, 6.00 m; ONW II, 13.35 m) vom Orog Nuur (Nuur =

See) See in der Gobi Wüste der Mongolei wurden näher analysiert. Es gibt viele Belege für

einen ausgeprägten Feuchtigkeitsimpuls während des marinen Isotopenstadiums 3 (MIS 3;

~36-~24 ka), welcher auch den maximalen letztglazialen Eisvorstoß im Khangai Gebirge

hervorgerufen haben könnte. Ein scharfer Übergang am Ende der letzten Eiszeit (~11 ka) wird

mittels geochemischen, palynologischen und Ostrakodendaten nachgewiesen. Das untere

170

Einzugsgebiet des Orog Nuur war im Pleistozän durch Artemisia Steppengesellschaften

dominiert und veränderte sich im Holozän allmählich zur Chenopodiaceae-Wüstensteppe. Das

frühe Holozän ist ebenso durch eine relativ feuchte Umwelt gekennzeichnet, wenn auch eine

gegenläufige Variabilität der hygrischen und Temperatursignale zwischen den

palynologischen und gesamt-geochemischen Signalen abgeleitet werden können. Der

holozäne Wasserkörper zeigt alkalische Bedingungen und war häufig allochthonen

Sedimenteinträgen sowie höherer limnischer Sauerstoffzufuhr als im späten Pleistozän

ausgesetzt. Dadurch wurden Phasen anoxischer Bedingungen im See verringert. Die zwei

genannten Feuchteimpulse könnten ein weit verbreitetes Phänomen im trockenen Zentralasien

sein, welches in zahlreichen Archiven dokumentiert wurde. Betrachtet man die

atmosphärische Kinetik, scheinen die gekoppelten atmosphärische Phänomene des

Ostasiatischen Sommermonsuns (East Asian Summer Monsoon; EASM) und eine verstärkte

Feuchtigkeitzufuhr durch die Westwinde die treibenden Mechanismen zu sein. Diese

gekoppelte Funktionsweise könnte durch die Schwankungen der Solareinstrahlung in borealen

Breiten moduliert worden sein. Des Weiteren wurden in den Bohrkernen vier aride Phasen,

repräsentiert durch Playa-Phasen, nachgewiesen: Um ~47 ka, ~37 ka, ~19 ka (gLGM) und

~13-~11 ka (Jüngere Dryas). Der geringere Feuchtetransport über die Westwinde sowie ein

abgeschwächter Ostasiatischer-Monsun könnte zur Ausprägung dieser Playa-Phasen

beigetragen haben.

Kontinuierliche, hochauflösende Elementhäufigkeiten wurden in 1 cm-Intervallen im

Bohrkern ONW II untersucht. Aufgrund der vorherrschenden Ton- oder schluffigen

Tonfraktion in den Sedimenten, zeigen Aluminium und Silizium weitgehend ein identisches

Muster. Die Verhaltensweisen von Kalzium können auf die Varianz des authigenen Kalzits

zurückgeführt werden. Mangan und Kobalt fungieren als aussagekräftige Indikatoren für

Redoxbedingungen. Wegen der kurzen Überdeckungszeit und der schwachen diagenetischen

Überprägung ist Eisen kein Indikator für die diagenetischen Prozesse. Zirkon kann

wahrscheinlich mit den fluvialen Sedimenten in Verbindung gebracht werden, wenn man sich

auf die Deckgesteine bezieht. Nicht aber notwendigerweise mit den äolischen Sanden.

Darüber hinaus könnte das Vorkommen von Schwefel in Seesedimenten die Redoxbedingung

anzeigen. Kalium ist eher mit K-Feldspat verknüpft, was mit allochthonen fluvialen Einträgen

in Zusammenhang gebracht werden kann. Die zwei mit dem letztglazialen Kältemaximum

(LGM) und der Jüngeren Tundrenzeit (YD = Younger Dryas) - Ereignissen einhergehenden

Sandschichten, werden durch grobkörnige Sande in den Bohrkern angezeigt. Wie aus dem

Biplot zwischen Ca und Al/Si abgeleitet werden kann, kommen unterschiedliche sedimentäre

171

Oberflächen als Quellen der lakustrinen Ablagerungen vor und nach dem Ende des

Pleistozäns in Frage. Im Holozän scheint im Vergleich zum späten Pleistozän ein deutlich

alkalisches Seesystem vorzuherrschen. Darauf deuten eine verstärkte fluviale

Morphodynamik und effektivere Erosion in den oberen Einzugsgebieten des Orog Nuur

während des frühen Holozäns hin. In ariden Gebieten, können Sande einen signifikanten

Einfluss auf die Gesamt-geochemie der Seesedimente ausüben. Zusammenfassend kann

festgehalten werden, dass angesichts der Gesamt-geochemie lakustrischer Sedimente, das

Paläoklima nicht der einzig bestimmende Faktor sein muss, sondern die Kopplung der

Einflüsse von Ausgangsgesteinen und die Verwitterungsintensität ebenso entscheidenden

Einfluss haben.

Examples from the Ejina Basin and Orog - RWTH Publications

Documents

Transcript of Examples from the Ejina Basin and Orog - RWTH Publications