Post on 22-Jan-2023
Cloud forest dynamics in the Mexican neotropics duringthe last 1300 years
B L A N C A L . F I G U E R O A - R A N G E L *, K AT H Y J . W I L L I S w z and M I G U E L O LV E R A - VA R G A S *
*Departamento de Ecologıa y Recursos Naturales-IMECBIO, Centro Universitario de la Costa Sur, Universidad de Guadalajara,
Apartado Postal 108, Autlan de Navarro, Jalisco, C.P. 48900, Mexico, wOxford Long-Term Ecology Laboratory, Oxford University
Centre for the Environment, School of Geography, South Parks Road, Oxford OX1 3QY, UK, zDepartment of Biology, University of
Bergen, N-5007 Bergen, Norway
Abstract
Key questions for understanding the resilience and variability of Mexican Neotropical
cloud forest assemblages in current and future climate change include: How have human
disturbances and climate change affected the dynamics of the cloud forest assemblage?
What are the predominant processes responsible for its present day composition and
distribution? Are the current conservation strategies for the cloud forest in accordance
with preserving its natural variability through time? In this study, the temporal dynamics
of the cloud forest in west-central Mexico over the last � 1300 years were reconstructed
using palaeoecological techniques. These included analyses of fossil pollen, microfossil
charcoal, and sediment geochemistry. Results indicated that a cloud forest assemblage
has been the predominant vegetation type in this region over the last � 1300 years.
During this time, however, there have been changes in the vegetation with an apparent
expansion of cloud forest from � 832 to 620 cal years BP and a decline from 1200 to 832 cal
years BP. Climate change (intervals of aridity) and human disturbances through anthro-
pogenic burning appear to have been the main factors influencing the dynamics of this
cloud forest. The spatial heterogeneity reported for high-altitude forests in this region, in
concert with high beta diversity, appears to be a manifestation of the high temporal
variability in species composition for these forests. Greater turnover in cloud forest taxa
occurred during intervals of increased humidity and is probably representative of a
higher temporal competition for resources among the cloud forest taxa. The present
results support the current protection scheme for cloud forests in west-central Mexico
where areas are kept in exclusion zones to avoid timber extraction, grazing, and
agriculture; this will maintain diversity within these forests, even if there are only a
few individuals per species, and enable the forests to retain some resilience to current
and future climate change.
Keywords: conservation, fire, forest ecology, logging, palaeoecology
Received 6 March 2009; revised version received 16 June 2009 and accepted 24 June 2009
Introduction
Cloud forest is recognized as the terrestrial ecosystem
with the highest diversity per unit area in Mexico and
includes 10% of the Mexican flora and 12% of terrestrial
vertebrates (Pineda & Halffter, 2004). This forest is
characterized by a complex biogeographical history that
combines temperate and tropical elements and has
specific environmental requirements including constant
levels of humidity, usually provided through the pre-
sence of fog, a precipitation range 1000 and 3000 mm,
and temperatures from 12 to 23 1C (Rzedowskii, 1981).
In terms of geographical distribution, cloud forest is
more continuous on the Atlantic than on the Pacific
slopes and it is mainly restricted to humid ravines
(Rzedowskii, 1996) located between 800 and 2700 m
a.s.l. The west-central region of Mexico, in particular,
contains several cloud forest blocks discontinuously
distributed within the Pacific part of the Serranias
Meridionales province (Luna Vega et al., 1999). Pre-
sently cloud forests in Mexico are extremely fragmentedCorrespondence: Blanca Figueroa-Rangel, tel. 1 317 38 25010 ext.
7045, fax 1 317 38 11425, e-mail: bfrangel@cucsur.udg.mx
Global Change Biology (2010) 16, 1689–1704, doi: 10.1111/j.1365-2486.2009.02024.x
r 2009 Blackwell Publishing Ltd 1689
(Luna Vega et al., 2000) covering o1% of the country.
Suggested causes of fragmentation are climate change
and human activities such as clearing for agriculture,
livestock grazing and logging (Bubb et al., 2004). How-
ever, very little data exists to determine the relative
importance of each of these on fragmentation of the
cloud forest.
Several stands of cloud forest comprising � 4500 ha
are located inside the Sierra de Manantlan Biosphere
Reserve (SMBR) in west-central Mexico, an area pro-
tected by government decree since 1987. Cloud forest
in the SMBR represents mature communities and con-
tains � 23% of all plant species in the area and the
highest taxonomic richness (Cuevas-Guzman et al.,
2004); these characteristics, in concert with its fragmen-
ted character, high beta diversity (Vazquez-Garcıa &
Givnish, 1998), complex vertical and horizontal struc-
ture and high endemism (Rzedowskii, 1996) make this
forest of high conservation priority.
Many of the tree types in the SMBR cloud forest
(e.g. Quercus, Tilia, Clethra, Dendropanax, Carpinus) have
generation times 450 years. In order to fully appreciate
the natural variability of this important vegetation
type and factors responsible for its fragmentation, a
longer-term perspective is essential. At present, the
longest perspective of cloud forest dynamics in
Mexico are data spanning no more than one decade
(Williams-Linera, 2002; Ramırez-Marcial, 2003; Alvarez-
Aquino et al., 2004; Suarez Guerrero & Equihua, 2005);
longer studies of several decades and centuries are still
very patchy in Mexico and centered in the Yucatan
Peninsula in lowland vegetation (Curtis et al., 1996;
Hodell et al., 2001, 2005a, b) or in central Mexico mainly
in Pine, Pine-Oak and Fir forests (Gonzalez-Quintero,
1986; Lozano-Garcıa et al., 1993; Metcalfe, 1995; Arnauld
et al., 1997; Almeida-Lenero et al., 2005; Metcalfe et al.,
2007).
The aim of this study, therefore, was to reconstruct
the long-term dynamics of the cloud forest in the SMBR
in order to address the following key questions: What
have been the long-term dynamics of the cloud forest in
this biosphere reserve? How have past human distur-
bances and climate change affected its present day
composition and distribution? Are conservation strate-
gies currently in place for the cloud forest in this region
in accordance with its long-term dynamics?
The temporal vegetation dynamics of the cloud
forest are reconstructed using fossil pollen, microfossil
charcoal and inorganic and organic sediments. The
data span the last � 1300 years covering Pre- and
Post-Hispanic occupation periods and a crucial time
of important climatic change in Mexico (Hodell
et al., 1995, 2001, 2005a, b; Curtis et al., 1996; Beach
et al., 2006).
Materials and methods
Study area
The cloud forest stand selected for this study is located
at 1913503200N, 10411605600W approximately 1820–1890 m
a.s.l. in the SMBR in west-central Mexico (Fig. 1). The
environmental characteristics most influential in deter-
mining the distribution of the cloud forest in the SMBR
are a temperate subhumid climate with temperatures
from 8 to 25 1C, rainfall range of 1000–1700 mm, and the
constancy of fog and humidity throughout the year
(Vazquez-Garcıa et al., 1995). The cloud forest in the
SMBR is a highly diverse community with at least 483
plant taxa identified (Cuevas-Guzman et al., 2004). Of
the arboreal component, predominant trees include
Carpinus tropicalis, Clethra vicentina, Cornus disciflora, Ilex
brandegeana, Magnolia iltisiana, Ostrya virginiana, Persea
hintonii, Pinus douglasiana, Quercus xalapensis, Quercus
candicans, Quercus castanea, Q. xalapensis, Tilia americana
var. mexicana and Zinowiewia concinna. In the understory
Parathesis villosa and Euphorbia sp. are the most common
shrubs (Vazquez-Garcıa et al., 1995). The nonarboreal
component of the cloud forest includes terrestrial
herbs, terrestrial and epiphytic ferns, climbers, mosses,
lichens, orchids and bromeliads with the following
families being the most representative: Adiantaceae,
Aspleniaceae, Grammitidaceae, Hymenophyllaceae,
Lycopodiaceae, Orchidaceae and Polypodiaceae.
In the early 1900s the cloud forests of the SMBR were
extensively logged until 1987 when they became pro-
tected in core zones of the biosphere reserve. This
protection scheme excluded most human activities such
as grazing, timber harvesting and agriculture (SEMAR-
NAP, 2000) from these zones.
Archaeology and culture. One of the nearest
archaeological sites to the cloud forest is located in
Autlan (1914600500N, 10412105900W) (Fig. 1) which, at
present, is the largest population centre close to the
SMBR (53,269 inhabitants). Kelly (1945, 1995) explored
43 archaeological sites in Autlan spanning from � AD
1200 to 1500. Human activities, mainly based on
agriculture, have been continuous in this region since
before the Spanish conquest in AD 1521; maize and chilli
were the main staples while clothing was manufactured
using cotton and maguey (Kelly, 1945).
Climate change. Most palaeo-reconstructions of climate
change in Mexico have focused on central Mexico and
the Yucatan Peninsula. From these studies based on
pollen records, stable isotopes, microfossil charcoal,
sedimentology, geomorphology and stratigraphy,
diatoms and historical records it is apparent that in
1690 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
central Mexico humid conditions prevailed from 4000 to
around 2500 BP (Gonzalez-Quintero, 1986; Lozano-
Garcıa et al., 1993; Metcalfe et al., 2000; Almeida-
Lenero et al., 2005) and arid conditions from 2500 to
1900 BP (O’Hara et al., 1993). Another period of
pronounced aridity also occurred around 1000 BP
(O’Hara et al., 1993; Metcalfe & Hales, 1994; Metcalfe,
1995; Arnauld et al., 1997). In the Yucatan Peninsula wet
conditions prevailed until 3000 BP and there is evidence
for an arid interval between � 1200 and 800 BP that
coincides with the collapse of the Mayans (Hodell
et al., 1995, 2001, 2005a; Curtis et al., 1996; Beach et al.,
2006). Furthermore, evidence from the study area
undertaken in pine-dominated forests revealed
intervals of aridity between � 4200 and 2500, 1200
and 850 and 500 and 200 cal years BP (Figueroa-Rangel
et al., 2008).
Coring and sediment sampling
A 96 cm sedimentary core of black organic material was
retrieved from a small forest hollow (approximately
11 m� 8 m diameter) in the cloud forest of the SMBR
at 1913503200N, 10411605600W (Fig. 1), using a Russian
corer (Glew et al., 2001). The core was extruded in the
field, wrapped, and returned to the Oxford Long-Term
Ecology laboratory for further analyses.
The core was sectioned in 1 cm3 subsamples for loss-
on-ignition (LOI) analysis to measure the contents of
Fig. 1 Map of the coring site in a cloud forest stand of the Sierra de Manantlan Biosphere Reserve, west-central Mexico.
C L O U D F O R E S T D Y N A M I C S I N M E X I C O 1691
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
carbonate and organic carbon in the sedimentary se-
quence (Dean, 1974).
Chronology
A chronology for the sedimentary sequence was estab-
lished by AMS radiocarbon dating from two labora-
tories (Poznan, Poland and East Kilbride, UK). Six
organic samples taken at regular intervals down
through the core were analysed and the results were
calibrated using the CALIB v5.02 and INTCAL04.14 pro-
grams (Reimer et al., 2004). The age–depth relationship
was modelled using linear interpolation (Bennett, 1994);
confidence intervals were reported using 1s range.
Fossil pollen, microfossil charcoal and geochemicalanalysis
A standard acetolysis method was used to extract
pollen (Bennett & Willis, 2001) from sedimentary sam-
ples 0.5 cm3 each. Sample processing included standard
sieving procedures with a maximum size fraction of
150 mm were undertaken. Tablets containing a known
quantity of Lycopodium spores were added to each
pollen samples in order to act as markers for the
estimation of pollen concentration (pollen grains
cm�3). To obtain a statistically significant sample size
at least 400 pollen grains was counted per sample
(Maher, 1972). The sequence was analysed every 2 cm
for a total of 48 samples. Pollen types were identified
using a pollen and spore reference collection of trees,
shrubs, herbs and terrestrial and epiphyte ferns col-
lected during the field season in the study area together
with the reference collection held in the Long-Term
Ecology Laboratory, Oxford University Centre for the
Environment.
Microfossil charcoal concentration of the sedimentary
sequence was determined using the point count method
(Clark, 1982). The same sampling interval was used as
for the pollen.
Geochemical analysis was undertaken at the same
sample resolution as for pollen and microfossil charcoal
analysis. A modified technique of the acid digestion
method (Bengtsson & Enell, 1986) was used whereby
soluble concentrations of different elements were deter-
mined using inductively coupled plasma mass spect-
ometer (ICP-MS) with a Perkin Elmer spectometer
(Perkin Elmer, Waltham, MA, USA) based in the Earth
Sciences Department of the University of Oxford.
Data handling
Pollen data were plotted and divided into zones by
optimal splitting and significant zones were determined
using a broken-stick model (Bennett, 1996). Rarefaction
analysis (Birks & Line, 1992) was applied to provide an
estimate of the palynological richness for each sample.
The rate-of-change analysis was undertaken to reveal
changes in the cloud forest assemblage over time (Ben-
nett & Humphry, 1995). The above-mentioned proce-
dures were undertaken using PSIMPOLL 4.25 and PSCOMB
1.03 software (Bennett, 2005). To detect significant dif-
ferences in the cloud forest assemblage and the envir-
onmental variables (geochemical elemental
concentrations, microfossil charcoal, pollen concentra-
tion, palynological richness, and rate-of-change) among
the zones determined by optimal splitting, a one-way
analysis of variance ANOVA was calculated using SPSS
v.16.0 [ SPSS, 2007; SPSS for Windows (version 16.0);
Science Inc., Chicago, IL, USA].
Results
Chronology
Radiocarbon dating of the sedimentary sequence indi-
cates that sediment accumulation began about
1540 � 30 cal years BP and continues to present with
no evidence for a hiatus in the sequence (Table 1). The
age–depth model (Fig. 2) indicates that approximately
1 cm of sediment accumulated in the basin every 16
years resulting in a pollen, microfossil and geochemical
sampling interval of one sample every 32 years starting
approximately at 1300 cal years BP. Pollen preservation
Table 1 Radiocarbon calibrated dates calculated from CALIB v5.02 and INTCAL04.14 programs (Reimer et al., 2004) used to develop
the chronology for cloud forest in west-central, Mexico
Depth (cm) Laboratory code 14C age (BP) Calibrated age (cal years BP)(1s)
16 SUERC-5872 207 � 24 162 � 10
40 SUERC-5873 586 � 21 615 � 20
60 SUERC-5874 743 � 24 678 � 10
75 SUERC-5877 925 � 21 879 � 30
82 Poz-10750 1100 � 30 1040 � 10
96 Poz-4747 1635 � 35 1540 � 30
1692 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
before this time was poor and did not allow preserva-
tion of a statistically significant sample size. The aver-
age content of organic material was � 18% while
carbonate content averaged � 0.02%.
This sequence therefore covers two important
historical episodes in Mexico, the pre- (before AD 1521
or � 465–436 cal years BP) and post-Hispanic (after AD
1521 or � 465–436 cal years BP) periods. These two
periods can be seen as contrasting epochs where human
occupation differed substantially in terms of natural
resource use and social interactions (Denevan, 1992;
Lentz, 2000). Pre-Hispanic (before AD 1521) includes
years before Spaniards conquered Mexico. Post-Hispa-
nic (after AD 1521) years enclose the Spanish conquest,
the Spanish Colony and recent times.
Cloud forest assemblage dynamics
The reconstructed cloud forest assemblage dynamics
from � 1300 cal years BP to present is displayed for all
pollen types, even those with scarce number of grains,
in order to portray the diversity of taxa through this
time (Fig. 3). Previous work on the modern pollen
signature has also indicated that, for the determination
of cloud forest, it is essential to include rare or very
infrequent taxa (B.L. Figueroa-Rangel, unpublished
data). Arboreal taxa indicative of cloud forest (Table 2)
are represented throughout the entire sequence
although with low abundances. However, as has been
previously demonstrated (Rzedowskii, 1996), epiphytic
fern spores such as Aspleniaceae, Hymenophyllaceae
and Polypodiaceae are good indicators of cloud forest
presence and their percentages are high at certain
intervals in the diagram (Fig. 3). According to the ANO-
VA results, pollen concentration and the rate-of-change
calculated through the chord distance are different
among the four zones while palynological richness is
similar (Table 3).
The four pollen zones determined by optimal split-
ting are named according to Mexican history; Pre-
Hispanic I, Pre-Hispanic II and Pre-Hispanic III repre-
sent years before AD 1521 and Post-Hispanic years after
AD 1521 (Fig. 3):
Pre-Hispanic I (� 1300–832 cal years BP)
In this zone arboreal taxa commonly found in present-
day cloud forest predominate. These include Pinus,
Quercus, Alnus, Magnolia, Carpinus and some which
are more abundant in the first half of the zone such as
Clethra, Lauraceae and Tilia (Fig. 3a). Although these
are taxa commonly found in present-day cloud forest
(Table 2), the sum including all cloud forest taxa
indicates low percentages when compare with the
following zone (Fig. 3b). Quercus and Pinus show simi-
lar percentages (from 20% to 30%) while Alnus displays
a high peak (� 15%) at the lowermost section of the
zone (Fig. 3a).
Ferns are evident through the presence of Adiantum,
Anogramma, Dryopteris, other Aspleniaceae, Poly-
podiaceae and Hymenophyllaceae. Low values for
Other Aspleniaceae (o10%) can be observed for this
zone. However, Polypodiaceae and Hymenophyllaceae
show their highest percentages (Fig. 3b). In terms of
herbaceous taxa, Poaceae and Asteraceae are predomi-
nant. The lowest pollen concentration occurred in this
zone with the minimum values around 1200 and 1100
cal years BP while the highest palynological richness is
observed in this zone (Fig. 4).
Pre-Hispanic II (� 832–620 cal years BP)
This zone is a flourishing stage for cloud forest as
revealed by the highest percentages in cloud forest taxa
(Fig. 3b) as well as the highest values in pollen concen-
tration (Fig. 4). Arboreal components with high percen-
tages in this zone are Quercus, Clethra, Meliosma,
Carpinus and Parathesis. Opposite to this pattern is Pinus
with the lowest percentages in the sequence (Fig. 3a).
Excepting Plantago with a peak around 678 � 10
cal years BP, the rest of the herbaceous taxa are scarce
in this zone (Fig. 3b). Other Aspleniaceae, Adiantum and
Anogramma are ferns showing highest percentages
along the sequence (Fig. 3b). Palynological richness is
fairly constant except for a decrease around 640 cal
years BP immediately followed by an increase in the
uppermost section of this zone (Fig. 4). The chord
distance curve, expressing the amount of change in
the cloud forest assemblage, is considerably higher for
Fig. 2 Age-depth model based on radiocarbon dates using
linear interpolation. The age is measured as calibrated years
before present (cal years BP).
C L O U D F O R E S T D Y N A M I C S I N M E X I C O 1693
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
(a)
(b)
1694 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
this zone suggesting a rapid change starting around
678 � 10 cal years BP and dropping in the following
zone (Fig. 4).
Pre-Hispanic III (� 620–480 cal years BP)
This is also a predominant cloud forest stage. Pinus
percentages rise in this zone coincident with a decline in
Quercus and Alnus (Fig. 3a). Main arboreal components
also include Clethra, Cleyera, Dendropanax, Fraxinus,
Hedyosmum, Ilex, Magnolia, Carpinus, Parathesis and Tilia
with very low percentages (Fig. 3a). Herbaceous taxa
such as Zea sp. and the families Chenopodiaceae,
Asteraceae and Poaceae are scarce or even absent
(Fig. 3b). Ferns present in this zone are Adiantum, Other
Aspleniaceae, Polypodiaceae and Hymenophyllaceae.
The highest drop in palynological richness can be ob-
served in this zone while chord distance decreases
reaching low values as well (Fig. 4).
Post-Hispanic (� 480 cal years BP to present)
Some cloud forest taxa decrease in this zone, in parti-
cular Quercus and Carpinus. At the same time, Pinus
performs the highest percentages of the sequence,
reaching 50% in some depths (Fig. 3a). Contrarily others
(e.g. Fraxinus and Tilia) increase only in this zone. In
general, the sum of all cloud forest taxa (ferns included)
decrease to the lowest percentages in the sequence
(Fig. 3b). The maximum number of rare pollen types
(e.g. Dalea, Polygonum, Apiaceae, Juncaceae and Lilia-
ceae) is present in this zone, although they are not
displayed graphically. The herbaceous element is pre-
sent with Crotalaria, Chenopodiacaeae, Asteraceae and
Poacaeae although all with low percentages (Fig. 3b).
Two peaks reaching 30% in Other Aspleniaceae are
apparent in the first section of the zone while Poly-
podiaceae reaches high values during the last 150 years.
There is a peak in pollen concentration around 162 � 10
cal years BP; chord distance is constant but low along
this zone (Fig. 4).
Environmental dynamics
Results from the geochemical and microfossil charcoal
analysis suggest that the environment surrounding the
cloud forest has been changing over the last 1300 years
in tandem with the pollen fluctuations (Figs. 3–5).
Burning has occurred irregularly over those years.
Although there are nonsignificant differences for char-
coal concentration among the zones (Table 3), the
Table 2 Present-day taxa representative of cloud forest in the
Sierra de Manantlan, Biosphere Reserve, west-central Mexico
Arboreal taxa (species) Non-Arboreal taxa (families)
Abies religiosa Araceae
Alnus jorullensis Apiaceae
Carpinus tropicalis Asclepiadaceae
Cinnamomum pachypodum Aspleniaceae
Citharexylum mocinnii Asteraceae
Clethra vicentina Begoniaceae
Cleyera integrifolia Bromeliaceae
Clusia salvinii Convolvulaceae
Conostegia volcanalis Cyperaceae
Cornus disciflora Dennstaedtiaceae
Dendropanax arboreus Euphorbiaceae
Fraxinus uhdei Fabaceae
Fuchsia arborescens Hymenophyllaceae
Garrya laurifolia Melastomataceae
Hedyosmum mexicanum Ophioglossaceae
Ilex brandegeana Orchidaceae
Inga vera subsp. eriocarpa Piperaceae
Juglans major Poaceae
Lippia umbellata Polypodiaceae
Magnolia iltisiana Rubiaceae
Matudaea trinervia Solanaceae
Meliosma dentata Thelypteridaceae
Oreopanax echinops
Ostrya virginiana
Parathesis villosa
Perrottetia longistylis
Persea hintonii
Pinus douglasiana
Prunus cortapico
Quercus candicans
Quercus salicifolia
Rhamnus hintonii
Rondeletia manantlanensis
Sebastiana hintonii
Styrax argenteus
Symplococarpon purpusii
Symplocos citrea
Tilia americana var. mexicana
Triumfetta barbosa
Vaccinium stenophyllum
Xylosma flexuosum
Zinowiewia concinna
Source: Cuevas-Guzman et al. (2004).
Fig. 3 (a) Pollen diagram of cloud forest trees and shrubs in west-central Mexico. Ages are calibrated years BP. Divisions are delineated
by zones according to the optimal splitting option available in PSIMPOLL 4.25. (b) Pollen diagram of cloud forest herbs and ferns in west-
central Mexico. The sum of all cloud forest taxa contributing to the total pollen sum is also displayed. Ages are calibrated years BP.
Divisions are delineated by zones according to the optimal splitting option available in PSIMPOLL 4.25.
C L O U D F O R E S T D Y N A M I C S I N M E X I C O 1695
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
greatest peak occurred in the Pre-Hispanic II and Post-
Hispanic zones coinciding with small decreases in the
sum of cloud forest taxa (Fig. 3b). Two high peaks are
centred at 678 � 10 cal years BP and a third one around
100 cal years BP while low values occurred indistinctly
in the four zones (Fig. 4). The ANOVA results show
statistically significant differences in some of the geo-
chemical elements (Table 3); an influx of some of these
elements (Fe, Cr, Co, Mg, K, Mn, Ba, Pb, U and Zn) can
be observed in the Pre-Hispanic I zone (� 1300–832 cal
years BP) with the highest values of the sequence
occurring in this zone. The following two zones are
represented by low values except for a distinct peak
around 500 cal years BP for all the elements. The Post-
Hispanic zone (� 480 cal years BP to present) experi-
ences a rise in some of the geochemical elements such
as Co, Mg, Mn, Ba and Pb (Fig. 5).
Discussion
What have been the long-term dynamics of the cloud forestin this biosphere reserve?
According to the zonation by optimal splitting, four
distinctive zones can be discriminated along the last
1300 years of forest dynamics in the present cloud forest
of the SMBR. These zones are particularly defined by
the alternate dominance of canopy arboreal elements
namely by Pinus, Quercus, Tilia, Carpinus, Fraxinus,
Magnolia, Clethra, Cleyera and Ilex. These taxa are com-
mon to most of the cloud forest studied in Mexico
(Alcantara & Luna, 2001; Cartujano et al., 2002;
Sanchez-Rodrıguez et al., 2003; Mejıa-Rodrıguez
et al., 2004). Their attainment of large heights and
diameters has enabled them to thrive in the upper
Table 3 Confidence intervals and ANOVA significant values (in bold) among the four zones delineated according to optimal
splitting (Bennett, 1996) for some of the most important vegetation and environmental variables in the sedimentary sequence
recovered in cloud forest of west-central Mexico
Variables
Zones
SignificancePre-I Pre-II Pre-III Post
Pollen concentration (# grains/cc) 6217.41 � 2432.9 52324.45 � 14564.7 69992.07 � 41705.0 39301.06 � 10763.5 0.000
Palynological richness E(T240) 17.76 � 1.4 16.73 � 0.9 14.90 � 6.0 15.90 � 1.1 0.093
Charcoal concentration (cm2/cm3) 1.01 � 04 1.84 � 0.9 0.84 � 1.6 1.93 � 0.8 0.272
Chord distance (cm�1) 0.87 � 0.21 3.51 � 0.81 2.53 � 2.92 1.26 � 0.22 0.000
Geochemical elements
g kg�1
Mg 4.02 � 1.1 0.78 � 0.1 1.04 � 0.9 1.73 � 0.6 0.000
Al 21.52 � 6.2 18.37 � 4.1 16.78 � 14.1 20.77 � 4.0 0.626
Fe 13.56 � 3.2 4.04 � 1.1 3.82 � 3.4 8.54 � 2.0 0.000
Ca 1.21 � 0.4 1.14 � 0.2 0.71 � 0.7 1.19 � 0.2 0.270
P 0.77 � 0.4 0.77 � 0.1 0.58 � 0.5 0.89 � 0.2 0.527
mg kg�1
Na 51.84 � 23.8 41.84 � 37.2 25.26 � 30.0 43.44 � 13.7 0.820
K 340.18 � 83.4 75.50 � 28.1 10.3.09 � 45.7 41.20 � 43.3 0.000
V 95.08 � 22.7 76.93 � 15.9 45.54 � 41.7 45.14 � 10.7 0.000
Cr 12.53 � 3.0 6.34 � 1.2 3.80 � 3.5 5.07 � 1.1 0.000
Mn 140.65 � 37.9 32.89 � 6.3 33.80 � 22.4 85.71 � 25.8 0.000
Co 8.61 � 2.2 1.82 � 0.4 2.37 � 1.8 4.39 � 1.1 0.000
Cu 32.98 � 8.3 22.97 � 4.3 22.22 � 17.3 25.20 � 5.3 0.081
Sr 7.44 � 1.8 10.26 � 2.0 7.28 � 6.3 11.14 � 2.1 0.047
Cd 0.06 � 0.1 0.10 � 0.0 0.06 � 0.1 0.09 � 0.0 0.032
Ba 97.19 � 23.9 52.18 � 8.7 45.72 � 37.3 85.71 � 19.0 0.000
La 9.97 � 2.5 8.41 � 1.3 7.65 � 6.0 10.53 � 2.0 0.203
Ce 23.22 � 6.3 18.66 � 3.2 18.28 � 14.9 22.95 � 4.5 0.307
Nd 15.40 � 4.0 12.84 � 2.2 13.18 � 11.3 16.62 � 3.1 0.223
Pb 3.97 � 1.1 2.30 � 0.4 2.41 � 2.1 4.54 � 1.0 0.000
Th 1.73 � 0.5 1.14 � 0.2 1.07 � 0.9 2.01 � 0.8 0.075
U 0.36 � 0.1 0.24 � 0.1 0.24 � 0.2 0.31 � 0.1 0.042
Zn 117.88 � 29.1 62.01 � 11.8 52.95 � 41.5 70.88 � 12.6 0.000
ANOVA, analysis of variance.
Bold values signify Po0.05.
1696 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
stratum. However, it is known from present-day studies
that cloud forest is much more diverse than this, usually
containing up to 100 different arboreal species (Cuevas-
Guzman et al., 2004).
Some of these arboreal elements, indicative of current
cloud forest in the SMBR (Table 2), are present in the
pollen sequence but the majority have low pollen
percentages or are absent from the record. This is most
likely a result of the pollination biology of the cloud
forest taxa which are mostly dispersed by animals,
mainly by insects and birds (Barbosa et al., 2006; Souba-
dra & Davidar, 2006; Deliso, 2008). In tropical forests
seventy percent of Nearctic forest taxa (e.g. Pinus, Alnus,
Quercus) are anemophilous while most of the Neotro-
pical taxa are entomophilous giving the possibility for
underrepresentation of pollen from Neotropical origin
and consequently overrepresentation of Nearctic pollen
(Bush, 2002). In view of the insufficient arboreal pollen
representing cloud forest, ferns are therefore regarded
as the main indicators for this community (Figueroa-
Rangel et al., 2008; B.L. Figueroa-Rangel, unpublished
data). Epiphytes embody the most diverse group (more
than 30% of all species) in the cloud forests of Mexico,
displaying a great endemism (Rzedowskii, 1996). In this
study Aspleniaceae, a family which includes a high
number of epiphytic species, presents percentages si-
milar to the most abundant trees (Pinus and Quercus)
while Polypodiaceae, considered the second most di-
verse family (47 genera) in cloud forest of Mexico
(Rzedowskii, 1996), shows abundances similar to that
of Carpinus, Tilia and Alnus. However, the rest of the
ferns, recognized to the genus level (Adiantum, Ano-
gramma, Botrichium, Cheilanthes, Dryopteris and Pteri-
dium), show limited numbers.
There has been change in the composition of the
cloud forest a number of times during the last 13
centuries as evidenced by the different taxa dominating
the pollen sum along the sequence. The rate-of-change,
calculated through the chord distance, expressed that
the main changes in the cloud forest assemblage
Fig. 4 Pollen concentration, palynological richness, chord distance and charcoal concentration for the cloud forest of west-central,
Mexico. Divisions are delineated by zones according to the optimal splitting option available in PSIMPOLL 4.25 undertaken on pollen data.
C L O U D F O R E S T D Y N A M I C S I N M E X I C O 1697
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
(a)
(b)
Fig. 5 Geochemical concentration for some elements in a cloud forest of west-central, Mexico. Divisions are delineated by zones
according to the optimal splitting option available in PSIMPOLL 4.25 undertaken on pollen data. Concentrations are expressed in mg kg�1
except Fe and Mg, which are in g kg�1.
1698 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
occurred during the time (Pre-Hispanic II � 832–620
cal years BP) when the cloud forest flourished the most.
During this period pollen concentration was also higher
than in the rest of the zones, moreover the sum of all
cloud forest taxa reached high percentages in this zone
summing up to 70% in some depths. Two of the most
important taxa whose changes in abundance varied
along the sequence were Pinus and Quercus, one ex-
panded at the same time the other contracted.
Of particular significance is a decline in the abun-
dance of Quercus at approximately 615 cal years BP
during the Pre-Hispanic III zone. This genus is one of
the canopy dominants in most cloud forests in Mexico
with high densities and basal areas (Sanchez-Rodrıguez
et al., 2003; Mejıa-Rodrıguez et al., 2004) and high
numbers of species (� 32 species for the study area)
even exceeding very diverse herbaceous families
(Alcantara & Luna, 2001). The Quercus decline could
therefore be an indication of a cloud forest decline from
the end of the Pre-Hispanic III to present.
Tilia became an important component of the cloud
forest from approximately 670 cal years BP and Fraxinus
from approximately 200 years ago. There is also an
important shift in dominance with Cleyera, Fraxinus, Ilex
and Tilia dominating in the Post-Hispanic zone (� 480
cal years BP to present) while others such as Carpinus,
Clethra and Meliosma, predominate in the Pre-Hispanic
II zone (� 832–620 cal years BP).
Differences in life-intervals for canopy dominants
have been reported in tropical forests (Bush & Colin-
vaux, 1994), with some trees needing a century or less to
become mature while others needing up to three cen-
turies. In addition, cloud forest taxa need particular
microhabitat conditions to survive (Hietz & Briones,
1998), so a temporal division of the resources is prob-
ably taking place through temporal niche partitioning.
It is suggested that this is playing an important role in
the alternating occurrence of the taxa; as a result there is
a heterogeneous canopy composition although with
low abundances per taxa across the 13 centuries that
this sequence represents.
In summary, it is difficult to identify a specific con-
stant cloud forest assemblage along the sequence. The
spatial heterogeneity reported for high-altitude forests
in this region (Olvera Vargas et al., 2006), in concert with
high beta diversity (Vazquez-Garcıa & Givnish, 1998), is
a manifestation of the high temporal variability in
species composition for these forests. This diverse as-
semblage has therefore remained remarkably resilient
during the last 1300 years as it is evidenced by the
nonsignificant differences in palynological richness
among the four zones, but its components are highly
dynamic as portrayed by the high differences in chord
distance among the zones (averages from 0.87 to
3.51 cm�1 and standard errors from 0.21 to 2.92 cm�1).
The different taxa seem to follow different trends ac-
cording to the zone, a tendency common to many
biological communities (Lovejoy, 2006) with each taxon
displaying an individualistic response. Therefore the
cloud forest taxa, particularly the arboreal component,
appear and disappear from the cloud forest assemblage
along the last 1300 years. The higher values for chord
distance in the Pre-Hispanic II zone (� 832–620 cal
years BP) is expressing that larger change in the cloud
forest assemblage was taking place during this interval
in time.
How have past human disturbances and climate changeaffected its composition?
Cloud forest conservation is critical as this forest shel-
ters arboreal dominants of different phytogeographical
affinities, as well as herbs and ferns with particular
requirements for their establishment. Results from this
study indicate that even though the cloud forest block
has remained remarkably resilient during the last 1300
years, the composition of the cloud forest has changed
possibly as individual species react to climate change or
to human disturbances, but to what extent are these
factors influencing its composition?
Climate change
Previous studies on the climate of Mexico reconstructed
from various fossil proxies has revealed that over the
past 1300 cal years BP broad-scale climate changes have
occurred in both temperature and precipitation (O’Hara
& Metcalfe, 1997; Metcalfe et al., 2000; Almeida-Lenero
et al., 2005). These changes were widespread and de-
tected in records from Bolson de Mapimi in Durango
and Coahuila (northern Mexico), Lake Zacapu in Mi-
choacan, lakes Zempoala, Quila and Texcoco in central
Mexico, the northern Yucatan Peninsula and the Pacific
Guatemalan coastline. The dry period (� 1200–832 cal
years BP) reported for the present research project coin-
cides with the period when drier conditions have been
reported for Mexico (AD 800–1000) (Metcalfe et al., 2000;
Hodell et al., 2001; Haug et al., 2003) and elsewhere (Chu
et al., 2002; Russell & Johnson, 2005; Neff et al., 2006).
For central Mexico an analogous dry period has been
also reported (Metcalfe & Hales, 1994; Metcalfe, 1995;
Arnauld et al., 1997; O’Hara & Metcalfe, 1997; Almeida-
Lenero et al., 2005; Ludlow-Wiechers et al., 2005; Met-
calfe et al., 2007). A general pattern that has emerged
from these regions is that during the past 1300 years,
there have been two intervals of widespread aridity
between 1200 and 850 and � 500 and 200 cal years BP
(Haug et al., 2003; Neff et al., 2006), probably driven by
C L O U D F O R E S T D Y N A M I C S I N M E X I C O 1699
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
changes in latitudinal migration of the intertropical
convergence zone (ITCZ).
Results from the geochemical records in this study
provide evidence for significant change occurring
around the basin over the past 1300 years, often tempo-
rally coinciding with the climate changes suggested
from other records. For example, in the Pre-Hispanic I
zone (� 1200–832 cal years BP) high values of geochem-
ical elements are detected. The excess of some geochem-
ical elements (i.e. Co, Mg, Mn, Ba and Pb) in the
sediment for that particular period could be related to
the scarcity of water in the soil, which hindered leach-
ing processes. A similar high concentration of some
geochemical elements has been related to climate fac-
tors, particularly to droughts, in other studies from
tropical (Brncic et al., 2007) and temperate regions
(Tipping et al., 2003). In contrast, during intervals of
increased humidity (� 832–550 cal years BP), assumed
by the increase in cloud forest taxa that rely on foggi-
ness to survive, lower values of inorganic geochemical
element (Fig. 5) are detected in this study coincident
with the interval of greatest turnover in the cloud forest.
So what is the cloud forest response to these intervals
of increased aridity? During the dry interval there is a
decline in pollen concentration and the sum of all cloud
forest taxa is also reduced, in particular some canopy
dominant decrease percentages such as Clethra, Tilia,
Cleyera, Fraxinus, Ilex and Meliosma. There is also an
increase in open ground taxa such as Poaceae as well as
in ferns Dryopteris, Hymenophyllaceae and Polypodia-
ceae. The increase in Poaceae is possibly related to open
canopies where the incidence of sunlight into the forest
is associated to an increase in aridity (Markgraf, 1993).
Hymenophyllaceae family is recognized for its high
tolerance to desiccation (Proctor, 2003), while Poly-
podiaceae family fluctuates enormously according to
saturation deficit (Benzing, 1998) with some species
reported as desiccation-resistant ferns (Muslin &
Homann, 1992; Andrade & Nobel, 1997; Proctor, 2003).
In contrast, during intervals of increased humidity (i.e.
� 832–620 cal years BP) there is evidence for increased
change in the cloud forest assemblage given by the
higher chord distance in that zone. An increase in
humidity probably provided the best conditions for
the reproduction of cloud forest taxa.
Human disturbances
Archaeological evidence suggests that there has been
human activity in the SMBR region over the past 1300
years. Archaeological remains from this region date
back to ca. AD 1200 suggesting human occupation before
the Spanish Conquest (AD 1521). In addition, archae-
ological excavations in Autlan (� 25 km from the coring
site), one of the localities closest to the SMBR (Fig. 1),
have revealed that by AD 1525 the area was an urban
population reliant upon intensive agriculture and irri-
gation (Kelly, 1945).
So what impact did these people have on the cloud
forest? There is certainly no evidence, in the pollen
record, of widespread clearance of the cloud forest but
it is highly likely that they would have carried out
small-scale agriculture, clearance through burning and
logging.
Similarities in the cloud forest assemblage as well as
in environmental dynamics between the Pre-Hispanic I
and the Post-Hispanic zones are obvious. The decline of
cloud forest and the consequent dominance of Pinus are
particularly similar between the two zones. Some of the
geochemical elements such as Mn, Ba and Pb, also reach
their highest values in these two zones. These coinci-
dences therefore seem to indicate that the changes are
the result of climate change events rather than to a
higher human disturbance.
However, human disturbance and climate change are
intricately linked in this part of the world and it is
possible that some changes occurring during these two
intervals are also indicative of human activity (Metcalfe
et al., 2000; Williams, 2003). It is possible that the
extraction of timber taxa such as Pinus and Quercus
has dictated some changes in the cloud forest assem-
blage. The decline of Quercus in the cloud forest of the
SMBR from �615 cal years BP to present may be related
to human influences such as logging, which resulted in
the simultaneous increase in Pinus. Similarly the expan-
sion of Pinus in southern Mexico in recent times has
been widely attributed to the influence of human activ-
ity, in particular to the extraction of Quercus trees
(Ramırez-Marcial et al., 2001).
Burning is constant through the whole sequence and
again it is possibly linked to the forest clearance and the
slash-and-burn system of agriculture. Agriculture in
Mexico is mainly developed through the slash and burn
system (Ellingson et al., 2000; Castellanos et al., 2001;
Ochoa-Gaona, 2001; Negreros-Castillo et al., 2003). This
consists of clearing the land followed by burning. This
land is cultivated and then left fallow for a certain
period of time before cultivating again (Pascual, 2005).
Evidence from our record also indicates the first
appearance of maize during the dry interval (� 1200–
832 cal years BP). It is likely that people in that time
started maize cultivation in the highlands as a conse-
quence of the scarcity of water in the lowlands. During
the subsequent wet periods (� 832 cal years BP to
present) reported in this study, maize is nearly absent
in the pollen spectra. We therefore suggest that the main
changes in the cloud forest assemblage of the SMBR
over the past 1300 years possibly indicate some evi-
1700 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
dence of human disturbances although these are small
scale and appear to be coupled with climate changes.
Implications of this record for predicting the impact offuture climate change and human use in the region
Results from this study corroborate the existence of a
dry period from 1200 to 800 cal years BP in mountain
forests of the region (B.L. Figueroa-Rangel, unpublished
data); in central Mexico (Metcalfe & Hales, 1994; Met-
calfe, 1995; Arnauld et al., 1997; O’Hara & Metcalfe,
1997; Almeida-Lenero et al., 2005; Ludlow-Wiechers
et al., 2005; Metcalfe et al., 2007); lowlands of the Yucatan
Peninsula (Hodell et al., 1995, 2001, 2005a, b) and the
Cariaco Basin in Venezuela (Haug et al., 2003). This
period is also an interval (1200–1000 cal years BP) when
global scale rapid climate change has been reported
(Metcalfe & Davies, 2007). The causes associated to this
phase of climate change have been attributed to solar
activity (Hodell et al., 2001; Haug et al., 2003), changes in
the latitudinal migration of the ITCZ (Metcalfe et al.,
2000; Hodell et al., 2005a, b; Berrio et al., 2006) and to the
ENSO variability (Metcalfe, 2006). During this time
interval the effect of climate change on human activities
has been observed, particularly for the Yucatan Penin-
sula (deMenocal, 2001).
The response of the cloud forest in this study to
previous intervals of climate change and human impact
reveals some interesting implications for the possible
impact of future climate change and human distur-
bances on this forest as follows: (i) during intervals of
aridity, cloud forest taxa tend to become reduced; (ii) in
contrast during intervals of increased humidity, the
cloud forest thrives (unsurprisingly) and there is also
a higher rate of change. We suggest that this is indica-
tive of a higher temporal competition for resources
among the cloud forest taxa during this more humid
interval due to greater availability of resources, parti-
cularly those linked to water; and (iii) The cloud forest
appears to be remarkably resilient to small-scale human
disturbances and burning, the main impact possibly
apparent through change in forest dominance. How-
ever, more extensive activity to include logging and
agriculture would be detrimental to the resilience of this
forest, due to a reduction in the overall taxonomic
richness and the expansion of cultural taxa.
Given the global climate change predictions for Neo-
tropical Mexico, where drier environments are ex-
pected, coupled with an intensified land use in the
search for water in the uplands of the region, it is
possible that the cloud forest will become species-poor,
its ecological memory for returning to a diverse assem-
blage in wetter conditions, lost and thus its overall
resilience greatly reduced.
Conservation strategies
At present the high beta diversity, few individuals per
species and the fragmented character of cloud forest
make it difficult to conceive the idea of a constant
assemblage over centuries. Results from our study
indicate, however, that while there may be no such
thing as a ‘constant’ assemblage of species in these
highly diverse forest, cloud forest ecosystems as a
whole have been remarkably resilient over time. The
composition of the cloud forest is constantly changing
through time and appears to be remarkably resilient to
climate change and small-scale human activities. In this
respect, it is therefore important to maintain diversity
within these forests, even if there are only a few
individuals per species and to avoid excessive logging
in these stands. Alternatively a selective extraction
scheme for timber species has to be designed depend-
ing on the composition of the canopy taxa and the
structure of the present-day vegetation in order to
maintain diversity so that long-term resilience of the
ecosystem may be retained. The present results fully
support the current protection scheme for cloud forests
in west-central Mexico where areas are kept in exclu-
sion zones, namely the core areas of the SMBR. In these
areas timber extraction, grazing and agricultural activ-
ities are excluded. Our results confirm that these are the
best conservation strategies for the long-term main-
tenance of the cloud forest dynamics.
Acknowledgements
We are greatly indebted to Martın Vazquez Lopez from theUniversity of Guadalajara for his valuable assistance duringthe fieldwork and to John Arden, from the Earth SciencesDepartment, University of Oxford for the ICP-MS analysis. Thefollowing institutions provided funds for the fieldwork: LinacreCollege, St Hugh’s College and The Vaughan Cornish Bequestfrom the University of Oxford. NERC Radiocarbon Laboratory atEast Kilbride, UK funded the radiocarbon dates by datingallocation no. 1048.1003. This research was supported by Con-acyt (Mexican National Council for Science and Technology) andPromep (SEP) through a Doctorate Scholarship to the firstauthor.
References
Alcantara O, Luna I (2001) Analisis florıstico de dos areas con
Bosque Mesofilo de Montana en el estado de Hidalgo, Mexico;
Eloxochitlan y Tlahuelompa. Acta Botanica Mexicana, 54, 51–87.
Almeida-Lenero L, Hooghiemstra H, Cleef AM, Van Geel B
(2005) Holocene climatic and environmental change from
pollen records of lakes Zempoala and Quila, central Mexican
highlands. Review of Palaeobotany and Palynology, 136, 63–92.
Alvarez-Aquino C, Williams-Linera G, Newton AC (2004) Ex-
perimental native tree seedling establishment for the restora-
tion of a Mexican cloud forest. Restoration Ecology, 12, 412–418.
C L O U D F O R E S T D Y N A M I C S I N M E X I C O 1701
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
Andrade JL, Nobel PS (1997) Microhabitats and water relations
of epiphytic cacti and ferns in a lowland neotropical forest.
Biotropica, 29, 261–270.
Arnauld C, Metcalfe S, Petrequin P (1997) Holocene climatic
change in the Zacapu Lake Basin, Michoacan: synthesis of
results. Quaternary International, 43/44, 173–179.
Barbosa L, Velazquez A, Mayorga-Saucedo R (2006) Solanaceae
composition, pollination and seed dispersal syndromes in Mex-
ican Mountain Cloud Forest. Acta Botanica Brasilia, 20, 599–614.
Beach T, Dunning N, Luzzadder-Beach S, Cook DE, Lohse J
(2006) Impacts of the ancient Maya on soils and soil erosion in
the central Maya Lowlands. Catena, 65, 166–178.
Bengtsson L, Enell M (1986) Chemical analysis. In: Handbook of
Holocene Palaeoecology and Paleohydrology (ed. Berglund BE),
pp. 423–455. John Wiley & Sons, New York, NY, USA.
Bennett KD (1994) Confidence intervals for age estimates and
deposition times in late-quaternary sediment sequences. The
Holocene, 4, 337–348.
Bennett KD (1996) Determination of the number of zones in a
biostratigraphical sequence. New Phytologist, 132, 155–170.
Bennett KD (2005) PSIMPOLL 4.25 and PSCOMB 1.03: C Programs
for Plotting Pollen Diagrams and Analysing Pollen Data. Uppsala
Universitet, Uppsala, Sweden. http://www.kv.geo.uu.se.
Bennett KD, Humphry RW (1995) Analysis of late-glacial
and Holocene rates of vegetational changes at two sites in
the British Isles. Review of Palaeobotany and Palynology, 85,
263–287.
Bennett KD, Willis K (2001) Pollen. In: Tracking Environmental
Change Using Lake Sediments. Volume 3: Terrestrial, Algal and
Siliceous Indicators (eds Smol. JP, Birks HJB, Last WM),
pp. 5–32. Kluwer Academic Publishers, Dordrecht, the Nether-
lands.
Benzing DH (1998) Vulnerabilities of tropical forests to climate
change: the significance of resident epiphytes. Climatic Change,
39, 519–540.
Berrio JC, Hooghiemstra H, van Geel B, Ludlow-Wiechers B
(2006) Environmental history of the dry forest biome of
Guerrero, Mexico, and human impact during the last c. 2700
years. The Holocene, 16, 63–80.
Birks HJB, Line JM (1992) The use of rarefaction analysis for
estimating palynological richness from quaternary pollen-
analytical data. The Holocene, 2, 1–10.
Brncic TM, Willis K, Harris DJ, Washington R (2007) Culture or
climate? The relative influences of past processes on the
composition of the lowland Congo rainforests. Phylosophical
Transactions of the Royal Society London B, 362, 229–242.
Bubb P, May I, Miles L, Sayer J (2004) Cloud Forest Agenda. UNEP-
WCMC, Cambridge, UK.
Bush MB (2002) On the interpretation of fossil Poaceae pollen in
the lowland humid neotropics. Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, 177, 5–17.
Bush MB, Colinvaux PA (1994) Tropical forest disturbance:
Palaeoecological records from Darien, Panama. Ecology, 75,
1761–1768.
Cartujano S, Zamudio S, Alcantara O, Luna I (2002) El Bosque
Mesofilo de Montana en el municipio de Landa, Matamoros,
Queretaro, Mexico. Boletın de la Sociedad Botanica Mexicana, 70,
13–43.
Castellanos J, Jaramillo VJ, Sanford RL, Kauffman JB (2001)
Slash-and-burn effects on fine root biomass and productivity
in a tropical dry forest ecosystem in Mexico. Forest Ecology and
Management, 148, 41–50.
Chu GQ, Liu JQ, Sun Q, Lu HY, Gu ZY, Wang WY, Liu TS
(2002) The ‘Mediaeval Warm Period’ drought recorded in
Lake Huguangyan, tropical South China. Holocene, 12,
511–516.
Clark RL (1982) Point count estimation of charcoal in pollen
preparations and thin sections of sediments. Pollen et spores, 24,
523–532.
Cuevas-Guzman R, Koch S, Garcıa-Moya E, Nunez-Lopez NM,
JardelPelaez E (2004) Flora vascular de la Estacion Cientıfica
Las Joyas. In: Flora y Vegetacion de la Estacion Cientıfica Las Joyas
(eds Cuevas-Guzman R, Jardel Pelaez E), pp. 117–176.
Universidad de Guadalajara, Autlan de Navarro, Jalisco, Mex-
ico.
Curtis JH, Hodell DA, Brenner M (1996) Climate variability on
the Yucatan Peninsula (Mexico) during the past 3500 years,
and implications for Maya cultural evolution. Quaternary
Research, 46, 37–47.
Dean WE (1974) Determination of carbonate and organic matter
in calcareous sediments and sedimentary rocks by loss on
ignition. Journal of Sedimentary Petrology, 44, 242–248.
Deliso E (2008) Climate Change and Hummingbirds of the Monte
Verde Cloud Forest. Centro Cientıfico Tropical, Costa Rica.
deMenocal PB (2001) Cultural responses to climate change
during the Late Holocene. Science, 292, 667–673.
Denevan WM (1992) The Pristine Myth: the landscape of the
Americas in 1492. Annals of the Association of American Geogra-
phers, 82, 369–385.
Ellingson LJ, Kauffman JB, Cummings DL, Sanford RL,
Jaramillo VJ (2000) Soil N dynamics associated with deforesta-
tion, biomass burning, and pasture conversion in a Mexican
tropical dry forest. Forest Ecology and Management, 137,
41–51.
Figueroa-Rangel BL, Willis K, Olvera-Vargas M (2008) 4000 years
of pine-dominated forest dynamics in the uplands of West-
Central Mexico: conserving a human or natural legacy?
Ecology, 89, 1893–1907.
Glew JR, Smol JP, Last WM (2001) Sediment core collection and
extrusion. In: Tracking Environmental Change Using Lake Sedi-
ments. Volume 1: Basin Analysis, Coring, and Chronological Tech-
niques (eds Last WM, Smol. JP), pp. 73–105. Kluwer Academic
Publishers, Dordrecht, the Netherlands.
Gonzalez-Quintero L (1986) Analisis polınicos de los sedimentos.
In: Tlapacoya: 35,000 anos de historia del Lago de Chalco
(eds Lorenzo JL, Mirambell L), pp. 157–166. Coleccion Cientıfica,
Serie Prehistoria. Instituto Nacional de Antropologıa e Historia,
Mexico, DF.
Haug GH, Gunther D, Peterson LC, Sigman DM, Hughen KA,
Aeschlimann B (2003) Climate and the collapse of Maya
civilization. Science, 299, 1731–1735.
Hietz P, Briones O (1998) Correlation between water relation and
within-canopy distribution of epiphytic ferns in a Mexican
cloud forest. Oecologia, 114, 305–316.
Hodell DA, Brenner M, Curtis JH (2005a) Terminal Classic
drought in the northern Maya lowlands inferred from multiple
1702 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
sediment cores in Lake Chichancanab (Mexico). Quaternary
Science Reviews, 24, 1413–1427.
Hodell DA, Brenner M, Curtis JH, Guilderson T (2001) Solar
forcing of drought frequency in the Maya Lowlands. Science,
292, 1367–1370.
Hodell DA, Brenner M, Curtis JH, Medina-Gonzalez R, Can EIC,
Albornaz-Pat A, Guilderson TP (2005b) Climate change on the
Yucatan Peninsula during the little ice age. Quaternary
Research, 63, 109–121.
Hodell DA, Curtis JH, Brenner M (1995) Possible role of climate in
the collapse of Classic Maya civilization. Nature, 375, 391–394.
Kelly I (1945) The Archaeology of the Autlan–Tuxcacuesco area
of Jalisco. I. The Autlan Zone. Ibero-Americana, 26, 1–93.
Lentz DL (2000) Imperfect Balance: Landscape Transformations in the
Precolumbian Americas. Columbia University Press, New York.
Lovejoy TE (2006) Protected areas: a prism for a changing world.
Trends in Ecology and Evolution, 21, 329–333.
Lozano-Garcıa MS, Ortega-Guerrero B, Caballero-Miranda M,
Urrutia-Fucugauchi J (1993) Late Pleistocene and Holocene
paleoenvironments of Chalco Lake, Central Mexico. Quatern-
ary Research, 40, 332–342.
Ludlow-Wiechers B, Almeida-Lenero L, Islebe G (2005) Paleoeco-
logical and climatic changes of the Upper Lerma Basin, Central
Mexico during the Holocene. Quaternary Research, 64, 318–332.
Luna Vega I, Alcantara Ayala O, Espinosa Organista D, Morrone
JJ (1999) Historical relationships of the Mexican cloud forests:
a preliminary vicariance model applying parsimony analysis
of endemicity to vascular plant taxa. Journal of Biogeography, 26,
1299–1305.
Luna Vega I, Alcantara Ayala O, Morrone JJ, Espinosa Organista
D (2000) Track analysis and conservation priorities in the
cloud forests of Hidalgo, Mexico. Diversity and Distributions,
6, 137–143.
Maher LJJ (1972) Absolute pollen diagram of Redrock Lake,
Boulder County, Colorado. Quaternary Research, 2, 531–553.
Markgraf V (1993) Younger Dryas in southernmost South Amer-
ica-un update. Quaternary Science Reviews, 12, 351–355.
Mejıa-Rodrıguez NR, Meave JA, Ruiz-Jimenez CA (2004) Anali-
sis estructural de un Bosque Mesofilo de Montana en el
extremo oriental de la Sierra Madre del Sur (Oaxaca), Mexico.
Boletın de la Sociedad Botanica Mexicana, 74, 13–29.
Metcalfe SE (1995) Holocene environmental change in the Zaca-
pu Basin, Mexico: a diatom-based record. The Holocene, 5,
196–208.
Metcalfe SE (2006) Late quaternary environments of the northern
deserts and central transvolcanic belt of Mexico. Annals of the
Missouri Botanical Garden, 93, 258–273.
Metcalfe SE, Davies S (2007) Deciphering recent climate change
in central Mexican lake records. Climatic Change, 83, 169–186.
Metcalfe SE, Davies SJ, Braisby JD, Leng MJ, Newton AJ, Terrett
NL, O’Hara SL (2007) Long-term changes in the Patzcuaro
Basin, central Mexico. Palaeogeography, Palaeoclimatology,
Palaeoecology, 247, 272–295.
Metcalfe SE, Hales PE (1994) Holocene diatoms from a Mexican
crater lake – La Piscina Yuriria. In: Proceedings of the 11th
International Diatom Symposium, San Francisco, USA, 1990,
Vol. 17, pp. 155–171. California Academy of Sciences, San
Francisco, CA, USA.
Metcalfe SE, O’Hara SL, Caballero M, Davies SJ (2000) Records of
Late Pleistocene–Holocene climatic change in Mexico – a
review. Quaternary Science Reviews, 19, 699–721.
Muslin EH, Homann PH (1992) Light as a Hazard for the
desiccation-resistant resurrection Fern Polypodium polypo-
dioides L. Plant, Cell, and Environment, 15, 81–89.
Neff H, Pearsall DM, Jones JG, de Pieters BA, Freidel DE (2006)
Climate change and population history in the Pacific Low-
lands of Southern Mesoamerica. Quaternary Research, 65,
390–400.
Negreros-Castillo P, Snook LK, Mize CW (2003) Regenerating
mahogany (Swietenia macrophylla) from seed in Quintana Roo,
Mexico: the effects of sowing method and clearing treatment.
Forest Ecology and Management, 183, 351–362.
O’Hara SL, Metcalfe SE (1997) The climate of Mexico since the
Aztec period. Quaternary International, 43-4, 25–31.
O’Hara SL, Street-Perrot FA, Burt TP (1993) Accelerated soil
erosion around a Mexican highland lake caused by prehispa-
nic agriculture. Nature, 362, 48–51.
Ochoa-Gaona S (2001) Traditional land-use systems and patterns
of forest fragmentation in the highlands of Chiapas, Mexico.
Environmental Management, 27, 571–586.
Olvera Vargas M, Figueroa-Rangel BL, Vazquez-Lopez JM,
Brown ND (2006) Dynamics and Silviculture of Montane
Mixed Oak forests in Western Mexico. In: Ecology and
Conservation of Neotropical Montane Oak Forests Vol. Ecological
Studies 185 (ed. Kappelle M), pp. 363–374. Springer-Verlag
Heidelberg, Berlin.
Pascual U (2005) Land use intensification potential in slash-and-
burn farming through improvements in technical efficiency.
Ecological Economics, 52, 497–511.
Pineda E, Halffter G (2004) Species diversity and habitat frag-
mentation: frogs in a tropical montane landscape in Mexico.
Biological Conservation, 117, 499–508.
Proctor MCF (2003) Comparative ecophysiological measure-
ments on the light responses, water relations and desiccation
tolerance of the filmy ferns Hymenophyllum wilsonii Hook and
H-tunbrigense (L.) Smith. Annals of Botany, 91, 717–727.
Ramırez-Marcial N (2003) Survival and growth of tree seedlings
in anthropogenically disturbed Mexican Montane rain forests.
Journal of Vegetation Science, 14, 881–890.
Ramırez-Marcial N, Gonzalez-Espinosa M, Williams-Linera G
(2001) Anthropogenic disturbance and tree diversity in Mon-
tane Rain Forests in Chiapas, Mexico. Forest Ecology and
Management, 154, 311–326.
Reimer PJ, Baillie MGL, Bard E et al. (2004) INTCAL04 terrestrial
radiocarbon age calibration, 0–26 cal kyr BP. Radiocarbon, 46,
1029–1058.
Russell JM, Johnson TC (2005) A high-resolution geochemical
record from Lake Edward, Uganda Congo and the timing and
causes of tropical African drought during the late Holocene.
Quaternary Science Reviews, 24, 1375–1389.
Rzedowskii J (1981) Vegetacion de Mexico. Limusa, Mexico, DF.
Rzedowskii J (1996) Analisis preliminar de la flora vascular de
los bosques mesofilos de montana de Mexico. Acta Botanica
Mexicana, 35, 25–44.
Sanchez-Rodrıguez EV, Lopez-Mata L, Garcıa-Moya E, Cuevas-
Guzman R (2003) Estructura, composicion florıstica y diversi-
C L O U D F O R E S T D Y N A M I C S I N M E X I C O 1703
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704
dad de especies de un bosque Mesofilo de Montana en la
Sierra de Manantlan, Jalisco. Boletın de la Sociedad Botanica
Mexicana, 73, 17–34.
SEMARNAP (2000) Programa de Manejo de la Reserva de la Biosfera
Sierra de Manantlan, Mexico. INE-SEMARNAP, Mexico, DF.
Soubadra M, Davidar P (2006) Breeding systems and pollination
modes of understorey shrubs in a medium elevation wet
evergreen forest, southern Western Ghats, India. Current
Science, 90, 838–842.
Suarez Guerrero AI, Equihua M (2005) Experimental tree assem-
blages on the ecological rehabilitation of a cloud forest in
Veracruz, Mexico. Forest Ecology and Management, 218, 329–341.
Tipping E, Smith EJ, Lawlor AJ, Hughes S, Stevens PA (2003)
Predicting the release of metals from ombrotrophic peat due to
drought-induced acidification. Environmental Pollution, 123,
239–253.
Vazquez-Garcıa JA, Cuevas-Guzman R, Cochrane TS, Iltis HH,
Santana-Michel FJ, Guzman-Hernandez L (1995) Flora de
Manantlan. SIDA-Botanical Miscellany, 13, 1–312.
Vazquez-Garcıa JA, Givnish TJ (1998) Altitudinal gradients in
tropical forest composition, structure, and diversity in the
Sierra de Manantlan. Journal of Ecology, 86, 999–1020.
Williams-Linera G (2002) Tree species richness complemen-
tary, disturbance and fragmentation in a Mexican tropical
Montane cloud forest. Biodiversity and Conservation, 11,
1825–1843.
Williams M (2003) Deforesting the Earth: from Prehistory to Global
Crisis. The University of Chicago Press, Chicago, USA.
1704 B . L . F I G U E R O A - R A N G E L et al.
r 2009 Blackwell Publishing Ltd, Global Change Biology, 16, 1689–1704