Soil carbon and nitrogen stocks under chronosequence of farm and traditional agroforestry land uses...

ORIGINAL ARTICLE

Soil carbon and nitrogen stocks under chronosequenceof farm and traditional agroforestry land uses in GamboDistrict, Southern Ethiopia

Ambachew Demessie • Bal Ram Singh •

Rattan Lal

Received: 5 December 2012 / Accepted: 1 July 2013 / Published online: 10 July 2013

� Springer Science+Business Media Dordrecht 2013

Abstract Conversion of forests to farm lands with-

out trees and farm lands with scattered trees (tradi-

tional agroforestry systems) may lead to decline of soil

organic carbon (SOC) and N stocks provided that they

have similar original status. This study was conducted

on soils with the age chronosequences of 12, 20, 30,

40, 50 years of farm (F), traditional agroforestry (AF)

and the adjacent natural forest (NF) lands. We studied

the changes in the concentration and stocks of SOC,

total N and their distribution in the soil profile of an

Andic Paleustalfs in Gambo District, Southern Ethi-

opia. Soil samples were collected at 10, 20, 40, 60,

100 cm depth interval from pits of 1 m depth in all

land use types and they were analyzed for their SOC

and N stock. The results showed that the greater

proportion of SOC and N was concentrated in

0–20 cm depth and that their concentration in AF and

F land uses was significantly lower than that under the

NF. Soils in traditional agroforestry land use showed a

trend of higher SOC stocks in all chronosequences

than those under the corresponding farm lands. The

SOC stock under the chronosequence of 12–50 years

of AF and F land uses varied from 28.2 to

98.9 Mg ha-1 or 12 to 43 % of the stock under the

NF. The SOC was less by 6.2 Mg ha-1 year-1 for

AF12 and 0.9 Mg ha-1 year-1 for AF50 compared

with NF. The corresponding values for farm lands

were 6.6 and 1.3 Mg ha-1 year-1. The N values of all

land uses were also less than that of the NF. The SOC

and N stocks tended to be less in farm lands than in the

traditional agroforestry. However, the SOC stocks

were not significantly higher with AF compared with F

suggesting that the parkland systems as practiced is

not sufficient to overcome other effects of cultivation.

Keywords Agroforestry �Carbon stocks � Farm land �Natural forest � Soil nitrogen � Soil organic carbon

Introduction

Global warming caused by the emission of greenhouse

gases (GHGs) including carbon dioxide (CO2), meth-

ane (CH4) and nitrous oxide (N2O), is a major concern.

The conversion of natural forests and peat land

ecosystems to farm and other land uses contributes

to GHGs emission (Kirby and Potvin 2007).

A. Demessie

School of Plant and Horticultural Sciences, College of

Agriculture, Hawassa University, P.Box 5, Hawassa,

Ethiopia

B. R. Singh (&)

Department of Plant and Environmental Sciences,

Norwegian University of Life Sciences (UMB),

P.Box 5003, 1432 As, Norway

e-mail: [email protected]

R. Lal

Carbon Management and Sequestration Center, Ohio

State University, 2021 Coffey Road, Columbus,

OH 43210, USA

123

Nutr Cycl Agroecosyst (2013) 95:365–375

DOI 10.1007/s10705-013-9570-0

In agricultural landscapes, agroforestry systems can

be economically and environmentally sound. The C

sequestration potential of agroforestry systems is

estimated to be between 12 and 228 Mg C ha-1 over

the life of the system (Albrecht and Kandji 2003;

Dixon 1995).

This potential of soil carbon sequestration in

agroforestry systems depends on the prevailing envi-

ronmental condition, the species used and the man-

agement practices involved. For instance the biomass

C stock varied from 0.7 to 54.0 Mg C ha-1 and total C

stock (biomass C ? soil C, to 1 m depth) varied from

28.7 to 87.3 Mg C ha-1 under traditional parkland

agroforestry systems with Faidherbia albida and

Vitellaria paradoxa, improved agroforestry systems

(live fence, fodder banks) and an abandoned previ-

ously cultivated land in the West African Sahel,

(Takimoto et al. 2008). The traditional parkland

agroforestry systems had relatively larger C stock

than the other systems for this particular study. Using

the 13C isotopic ratio as an indicator of relative

contribution of trees (C3 plants) and crops (C4 plants)

to soil C, more tree-origin C was found in larger

particle size and surface soil and indicated that long-

term tree presence promoted storage of protected C in

deeper soil (Takimoto et al. 2009).

Sequestering C in soils, not only removes excess

CO2 from the air per-se, but also improving soil

organic matter (SOM) and nutrient supply (Janzen

2006). Agricultural soils have a significant CO2 sink

capacity provided that judicious management of farm

land is practiced (Paustian et al. 1997). The input of

crop residues is essential to increase SOC concentra-

tion (Duiker and Lal 2000; Loveland and Webb 2003).

In Gambo district, Southern Ethiopia, extensive

deforestation, overgrazing and conversion of natural

ecosystem into arable land are rampant (Solomon et al.

2002a; Lemenih and Itanna 2004; Ashagrie et al.

2005). However, remnant trees left on agricultural

lands created a traditional (parkland) agroforestry

(AF) system with unique configuration, mode of

establishment and management. Very little crops

residue is returned to the soil in both AF and F land

uses as these are used for animal feed and household

fuel. This type of residue management does not permit

the replenishment of depleted organic matter (OM)

and thus can lead to decline in soil quality (Yimer et al.

2007; Blanco-Canqui and Lal 2007). Soil C stock were

40.3 Mg C ha-1 and 5.3 Mg N ha-1 in semi-arid

woodland in the Rift Valley of Ethiopia, and

234.6 Mg C and 20.2 Mg N ha-1 in Podocarpus

falcatus forest soil, while C and total N stocks in

cropland soils were less (Solomon et al. 2002b;

Lemenih and Itanna 2004; Yimer et al. 2007).

However, little is known of changes in soil C and N

stocks following conversion of natural forest to either

agroforestry or farmland in Gambo district, Southern

Ethiopia and throughout sub-Saharan Africa. It is

hypothesized that SOC and N stocks in traditional

agroforestry systems differ compared to treeless

farming. The objectives of the study were to deter-

mine: (1) the changes in the concentration and stocks

of SOC and N in soils during 12, 20, 30, 40 and

50 years after conversion of natural forest to agrofor-

estry and farm lands under the prevailing farmer’s

practice, and (2) the distribution of SOC and N in the

soil profile to 1 m depth of these land use systems.

Materials and methods

Description of the study site

The study was conducted at three sites namely, Ashoka,

Leye and Beseko, which are adjacent to the natural forest

in Gambo district, Southern Ethiopia that covers an area

positioned on the lower fringe of the western escarpment

of the southeastern highlands (Fig. 1). The escarpment

extends from about 2,100 to 3,200 m a.s.l and the plain

descends gradually to the Rift Valley lakes at about

1,600 m a.s.l. The Ashoka, Leye and Beseko sites are

adjacent to each other, lie within 7�170N and 7�200N and

38�480E and 38�490E, at about 240 km southeast of Addis

Ababa. The altitude ranges from 2,137 to 2,215 m a.s.l,

and the slope from 4 to 11 % (Table 1).

Rainfall is bimodal with mean annual precipitation

of 973 mm, of which 50.3 % falls in July–September.

Temperature ranged between the mean annual max-

imum of 26.6 �C and mean of 10.4 �C for the period

from 1999 to 2007.

The soil parent materials of Gambo District are of

volcanic origin, principally trachytes and basalts with

ignimbrites and pumices at the Rift Valley floor

(Solomon et al. 2002a). The soil textural class ranges

from silt loam to clay (Table 2) and are classified as

Andic Paleustalfs with thick argillic horizon and some

Andic soil material in the upper soil layers (Soil

Survey Staff 1999).

366 Nutr Cycl Agroecosyst (2013) 95:365–375

123

The dominant species of the upper storey vegeta-

tion in the natural forest and agricultural landscapes

are P. falcatus Thunb.ex Mirb., Croton macrostachys

Hochst.ex Rich., Prunus africana (Hook.F.) Kalkm,

Schefflera abyssinica (Hochst. Ex A. Rich) Harms.,

Syzygium guineense (Willd.) DC and Ficus sure

(Forsk). The under story vegetation of the natural

forest is dominated by: Lannea schimperi (A.Rich.)

Engl. Rytigynia neglecta (Hiern) Robyns, Maytenus

arbutifolia (Hochst.ex A.Rich.) Wilezek, Bersama

abyssinica Fres, and Psydrax schimperiana (A. Rich).

The most abundant herb in the forest floor is Hypoestes

furskaolii (Vahl.)R. Br.

The trees in the traditional agroforestry system of

all age chronosequences are trees left deliberately

during the clearance of the natural forest and trees that

emerged in crop fields (Table 1). Farmers deliberately

kept the number of trees very small (median = 5/ha)

in their farm lands to avoid excessive shading on their

crops. These trees are mostly semi-deciduous trees

that contribute leaf litter and recycle soil nutrients.

Branches are harvested for fuel, fencing and timber for

construction, and for sale.

The main crops are maize (Zea mays L.), wheat

(Triticum aestivum L.), sorghum (Sorghum bicolor L.)

and potato (Solanum tuberrosum Linnaeus). Crops are

grown during the rainy season and management is

similar for both AF and F land uses. After harvest,

farmers often remove or heap and burn the crop

residues several weeks before the land preparation for

the next growing season.

Soil sampling and analysis

Due to lack of permanent plots for monitoring

temporal changes in C and N stocks at the study sites,

a spatial analogue and chronosequence approach was

used for soil sampling in the designated land use

systems (Sa et al. 2001; Turner et al. 2005; Liao et al.

2006). The Ashoka, Leye and Beseko study sites have

an age chronosequence of agroforestry and farm lands

within similar soil and climate (Fig. 1). The

Fig. 1 Maps showing the sites of the natural forest (NF) and the chronosequence of Aroforestry (AF) and farm (F) land uses in Gambo

District, Southern Ethiopia

Nutr Cycl Agroecosyst (2013) 95:365–375 367

123

Table 1 Sample field location and characteristics for two land uses and three locations in Southern Ethiopia

Site Land use Age

(years)

T. species in

AF land use

D1.3 (cm) H(M) Plt no Elevation

(m a.s.l)

Coordinate of soil sampling pits Slope

%Longitude Latitude

Ashoka Agroforestry 12 P f 212 42.2 1(4) 2205 07�170N 038�480E 11

Pa 134 41.1 2(4) 2208 07�170N 038�480E 11

Pa 120 36.2 3(4) 2212 07�170N 038�480E 11

Pa 116 35.7 4(4) 2203 07�170N 038�480E 11

Pa 105 27.0

Ashoka Agroforestry 20 F.s 115 12.6 1(4) 2178 07�180N 038�480E 8

F.s 103 12.6 2(4) 2183 07�180N 038�480E 8

C.m 103 30.6 3(4) 2181 07�180N 038�480E 8

C.m 98 12.6 4(4) 2187 07�180N 038�480E 8

C.m 88 16.8

C.m 84 13.8

C.m 80 18.0

C.m 74 25.2

C.m 66 25.0

C.m 65 26.8

Ashoka Agroforestry 30 S.a 129 20.0 1(4) 2177 07�180N 038�480E 6

Pa 112 29.4 2(4) 2182 07�180N 038�480E 6

P f 86 24.6 3(4) 2179 07�180N 038�480E 6

Pa 72 20.8 4(4) 2159 07�180N 038�480E 6

P f 60 20.5

Beseko Agroforestry 40 P f 133 30.9 1(4) 2149 07�190N 038�490E 7

P f 130 30.6 2(4) 2151 07�190N 038�490E 7

P f 124 32.7 3(4) 2157 07�190N 038�490E 7

P f 124 22.6 4(4) 2166 07�190N 038�490E 7

P f 121 34.8

P f 101 31.8

P f 98 29.4

P f 94 28.3

Leye Agroforestry 50 P f 232 53.1 1(4) 2163 07�180N 038�480E 8

P f 200 40.2 2(4) 2168 07�180N 038�480E 8

3(4) 2165 07�180N 038�480E 8

4(4) 2167 07�180N 038�480E 8

Ashoka Farm 12 1(4) 2205 07�170N 038�480E 11

2(4) 2209 07�170N 038�480E 11

3(4) 2215 07�170N 038�480E 11

4(4) 2207 07�170N 038�480E 11

Ashoka Farm 20 1(4) 2182 07�180N 038�480E 4

2(4) 2190 07�170N 038�480E 4

3(4) 2185 07�170N 038�480E 4

4(4) 2184 07�170N 038�480E 4


123

chronosequences of AF and F land uses were of 12, 20,

30, 40 and 50 years since their conversion from the

natural forest. The natural forest at Ashoka was the

reference land use. The information of the time since

conversion was obtained from elderly key informants,

farmers owning the farm and cross cheeked with the

Forest Department archive.

The sampling was done with four replicates

(Table 1). The sampling points were within rectangu-

lar plots of 20 9 20 m. The maximum distance in

which the tree litter observed in the field was 20 m

from the tree trunk. The size of the plots was then

determined by taking into account that the trees

influence on the accrual of C and N in the AF land use

can be captured up to a distance of 20 m. Within

approximately the same contour line, two trees with a

minimum of 60 cm diameter growth and a minimum

of 40 m distance between them were selected for each

age chronosequence of AF land uses. Then two of the

20 9 20 m plots were assigned and laid against the

slope on the east and western side of each tree.

Similarly 4 sampling plots in a row were laid out in the

east west direction against the slop gradient for each

age chronosequence in crop fields. However, the land

use types and their respective chronosequences were

not replicated in different sites of the study area. Lack

of land uses with similar chronosequences in the study

area makes their replication difficult.

For each land use a 1 9 1 9 1 m soil pit was dug in

each corner of the plots of which two were very close

to the tree (4 m from the trunk) and two were at 20 m

distance. The soil samples were collected from 0 to 10,

10 to 20, 20 to 40, 40 to 60, 60 to 100 cm depth

increments. The samples were then bulked corre-

sponding to the depth increments. Accordingly, 4*5

bulked samples were obtained from each land use.

Hence, in both AF and F land uses, a total of 200

bulked soil samples were collected. The number of

samples collected from the NF is as described in

Demessie et al. (2011). These samples were air-dried,

ground, passed through a 2 mm sieve and analyzed for

physical and chemical properties. In addition to these

pits, representative bigger pits, up to a depth of 3 m,

were dug across the study sites for profile description

and soil texture determination.

Soil core samples were collected (by forcing a sharp

edged steel cylinder) from the same direction and

depth increments in each peat for bulk density (BD)

determination (Tan 2005). In all, 200 core samples

were collected. Soil samples were then oven-dried at

105 �C for 24 h and weighed using electronic balance.

The dry soil was crushed and passed through 2 mm

sieve for making correction in soil BD for gravel

content. The pH was determined by potentiometric

method (Tan 2005). The concentrations of N were

measured using a LECO CHN-1000 Carbon and

Table 1 continued

Site Land use Age

(years)

T. species in

AF land use

D1.3 (cm) H(M) Plt no Elevation

(m a.s.l)

Coordinate of soil sampling pits Slope

%Longitude Latitude

Ashoka Farm 30 1(4) 2181 07�180N 038�480E 6

2(4) 2176 07�180N 038�480E 6

3(4) 2175 07�180N 038�480E 6

4(4) 2186 07�180N 038�480E 6

Beseko Farm 40 1(4) 2139 07�200N 038�490E 6

2(4) 2137 07�200N 038�490E 6

3(4) 2137 07�200N 038�490E 6

4(4) 2139 07�200N 038�490E 6

Leye Farm 50 1(4) 2159 07�180N 038�480E 9

2(4) 2158 07�180N 038�480E 9

3(4) 2163 07�180N 038�480E 9

4(4) 2157 07�180N 038�480E 9

Values in the parenthesis indicate number of pits/plot

Where T = Tree, AF = Traditional agroforestry system, Plt = Plot No, D1.3 = Diameter at breast height, H = Height, Pf = P.

falcatus, Pa = P. africana, Fs = F.sure, Cm = C.machrostachus, Sa = S. abyssinica


123

Nitrogen Analyzer. SOM was analyzed using titri-

metric method (Walkley and Black 1934).The SOC

was obtained by dividing the SOM concentration by a

factor of 1.724 (Tan 2005). Texture was measured by

the hydrometer method for the selected profiles of the

study sites (Table 2).

Measurements and calculations

The SOC stock (Mg ha-1) for each sampled depth was

calculated using equivalent soil mass by adjusting the

thickness of the soil as described by Ellert et al.

(2001); to correct error that may be introduced due to

variation in BD.

SOC stock Mg ha�1� �

¼ Cconc: � BD � T

� 10000 m2ha�1

� 0:001 Mg kg�1

� CFcoarse ð1Þ

where Cconc. = concentration (kg Mg -1), BD = bulk

density (Mg m-3)

T ¼ depth thickness mð Þ; CFcoarse

¼ correction factor�1�

�Gravel %

þ Stone %�=100

�

The soil mass was calculated using Eq. (2) and

thickness of the soil depths were adjusted using Eq. (3).

Msoil ¼ BD � T � 10; 000 m2ha�1 ð2Þ

where Msoil = soil mass per unit area (Mg ha-1)

Tadd ¼ Msoil;equiv � Msoil;layer

� �

� 0:0001 ha m�2=BD subsurface ð3Þ

where Tadd = additional thickness of sub surface layer

required to attain the equivalent soil mass (m)

Msoil;equiv ¼ equivalent soil mass

¼ mass of heaviest soil layer at each

1m depth Mg ha�1� �

BD subsurface

¼ bulk density of subsurface layer Mg m�3� �

Table 2 Soil characteristics of three selected profiles and two locations in Southern Ethiopia

Sites Horizon Depth

(cm)

pH(H2O)

1:2.5

BD

(g cm-3)

Particle size Textural

class

Oxalate extraction Allophane

%Sand

%

Silt

%

Clay

%

Alox

%

Siox

%

Alp

(%)

Leye 1A 5 7.6 0.9 44 37.3 18.7 L 0.21 0.1 0.1 0.7

1A2 10 6.9 0.9 54 33.3 12.7 SL 0.23 0.1 0.1 0.6

1A3 25 7.8 1.0 38 31.3 30.7 CL 0.19 0.1 0.2 0.5

AB 85 5.8 1.1 20 31.3 48.7 C 0.21 0.1 1.0 0.2

Bt1 125 5.1 1.2 22 19.3 58.7 C 0.19 0.1 1.4 0.1

Bt2 300 5.1 ND 12 19.3 68.7 C 0.20 0.1 2.5 0.0

Ashoka(NF) 1A 9 7.5 0.7 34 31.3 34.7 CL 0.18 0.1 0.3 0.4

E 18 7.8 0.9 55.3 21.3 23.4 SCL 0.16 0.1 0.2 0.4

2A 40 7.6 1.1 37.3 29.3 33.4 CL 0.17 0.1 0.2 0.4

AB 85 7.5 1.2 27.3 35.3 37.4 CL 0.16 0.1 0.8 0.2

B 210 7.4 1.3 21.3 11.3 67.4 C 0.19 0.1 2.0 0.1

Ashoka 1Ap 15 6.5 1.0 49.3 35.3 15.4 L 0.17 0.1 0.1 0.5

2A1 35 6.6 1.0 47.3 31.3 21.4 C 0.19 0.1 0.1 0.5

3Bt1 90 5.3 1.0 21.3 11.3 67.4 C 0.24 0.1 1.7 0.0

3Bt2 160 5.1 1.1 23.3 11.3 65.4 C 0.25 0.1 2.9 0.0

4BC 180 4.7 1.2 37.3 13.3 49.4 C 0.24 0.1 1.7 0.1

BD = Bulk density, BS = Base saturation, L = loam, SL = Silt loam, CL = Clay loam, C = Clay, SCL = Silt clay loam,

Jp = Juniperusprocera, AF = agroforestry, NF = natural forest, Alox. and Siox = Acid oxalate extractable Al, SI,

Alp = Pyrophosphate extractable Al, P = phosphate retention (Blakemore et al. 1987), % Allophane = 100(% Sio/(23.4–5.1x))

where x = (Alox-Alp)/Siox representing the atomic ratio of Al and Si in allophane


123

Allophane contents of three selected profiles of 3 m

depth were calculated according to Mizota and Van

Reeuwijk (1989).

% allophone ¼ 100 % Sio= 23:4� 5:1xð Þð Þ ð4Þ

where x = (Alo–Alp)/Sio representing the atomic ratio

of Al and Si in allophane

Alo ¼ oxalate extractable Al; Alp

¼ pyrophosphate extractable Al

Total stock of N (Mg ha-1) for each sampling depth

was also computed using the equations as for SOC

stock. The total soil SOC and N stocks were calculated

by adding the stock of the prescribed soil layer to 1 m

depth. The change of SOC stocks over time was

computed by subtracting the stocks of chronosequence

of AF and F land uses from that of NF and dividing it

by the number of years of the chronosequence.

Statistical analysis

The data obtained were analyzed using SAS software

(SAS Inc. 2003).The effect of land uses on SOC, N, pH

and BD (dependent variables) and their distribution in

each of the 1 m depth interval were subjected to one way

analysis of variance using the general linear model

procedures of SAS. Land uses and depth were the

independent variables. The mean square error (R) was

obtained by adding the square of differences between

the observed and the mean of the dependant variables

and dividing it to residual degree of freedom, i.e.

R =P

(y–�y)2/Residual DF. The Multiple compari-

son of means for each class variable was carried out

using the Student–Newman–Keuls (SNK) test at

a = 0.05.

Results

The physical and chemical soil characteristics of three

selected profiles at Ashoka and Lye sites are shown in

Table 2. Statistics is not provided as the 3 m depth

profiles were not replicated. However, the measured

value of the clay content in the surface soils at Ashoka

tended to be higher than that at Lye site but the allophone

in the surface layers of 0–10 cm depth tended to be

higher in soils at Lye compared to Ashoka sites.

Soil pH was low in the surface 0–10 cm depth

compared with 10–100 cm depth in all land uses

(Table 3). Soil pH was higher for NF, compared with

F12 for the 0–10 cm depth and most year systems for the

10–20 cm depth. BD in the 0–10 cm depth under NF was

0.7 compared with a mean of 1.0 g cm-3 for cultivated

fields. In 10–20 cm, it was 1.2 under F50 compared with

0.9 kg m-3 under NF. The BD at 20–40 cm under AF50

and F50 was high compared with AF40. Apart from few

exceptions at 100 cm depth, no significant variations in

BD was observed below 40 cm depth (Table 3). In

general, BD increased with soil depth.

In the 0–10 cm soil depth, SOC concentration

under NF was about 48–66 % higher compared with

the mean for all other and similar among F and AF

chronosequence (Figs. 2 and 3). In the 10–20 cm

depth, SOC concentration under F20 was low com-

pared with that under NF, AF40, AF50 and F40. In the

20–40 and 40–60 cm depth, the SOC concentration

did not differ (P \ 0.05) among land uses. In the

60–100 cm depth, the SOC under NF was higher with

F30, and F20 (P \ 0.05) and higher for AF40 and AF50

compared with F20 (Figs. 2 and 3). Soil N concentra-

tion in the surface layers of 0–10 cm depth under NF

was significantly high compared with F and AF land

uses (P \ 0.05) but it did not differ by land uses for

depth below 10 cm (Table 3).

The SOC stock under land uses of all age chron-

osequences with the exception of AF40, was low

compared with NF. The SOC stock F compared with

AF did not differ within time under cultivation

(Fig. 4). Land uses did not affect N stock except for

more under AF40 compared with AF20 (Fig. 5).

Conversion of the relict NF into arable land caused

decline in SOC stocks for all chronosequences of AF

and F land uses but the N stock did not decline. The

SOC stock in F and AF land uses of 12, 20 and

30 years of cultivation was 130.8–156.4 Mg ha-1 or

32–43 % lower compared with NF while the corre-

sponding SOC stock under the chronosequence of 40

and 50 years of both land uses varied from 164.5 to

201.5 Mg ha-1 or 12 to 28 % lower than that under

the NF. The SOC stock under the age chronosequence

of 12, 20, 30, 40 and 50 years of F and AF land uses

varied from 28.2 to 98.9 Mg ha-1 or 12 to 43 %.

Generally, the decline of SOC stocks was higher for

the first 12–20 years but approached to a steady state

after 40–50 years under both land uses of age chron-

osequences (Fig. 6). The SOC decline was lower

under AF (0.7 to 6.2 Mg ha-1 year-1) than that of

under F land use (1.3–6.6 Mg ha-1 year-1).


123

Discussion

Higher BD in the upper layer of AF and F land uses as

compared to NF may be attributed to lower SOC

content and soil compaction caused by cultivation

practices in the F land use (Murty et al. 2002). The

SOC and N under all land uses decreased with soil

depth. Most of the OM added through leaf fall, crop

residues and root litter were retained in the upper

layers. Secondly, the transport of dissolved organic

carbon (DOC) to sub-soil can also be limited in this

environment as evapotranspiration exceeds the pre-

cipitation in most months of the year (Hawando 1997).

Table 3 Distribution of pH, bulk density and N for AF and F land uses and three locations in Southern Ethiopia

Depth NF AF12 AF20 AF30 AF40 AF50 F12 F20 F30 F40 F50

pH-H2o (1:2.5)

10 6.4a 5.8ab 5.8ab 5.9ab 6.1ab 6.0ab 5.7b 5.8ab 5.8ab 6.3ab 6.0ab

20 6.7a 6.0b 6.0b 6.1b 6.3b 6.1b 6.1b 6.2b 6.0b 6.4ab 6.2b

Bulk density (g cm-3)

10 0.7b 0.9a 1.1a 1.0a 1.1a 1.0a 1.0a 1.0a 1.1a 1.0a 1.1a

20 0.9b 0.9b 1.0ab 1.1ab 1.0b 1.0ab 0.9b 0.9b 1.1ab 1.1ab 1.2a

40 1.1ab 1.1ab 1.0ab 1.0ab 1.0b 1.2a 1.1ab 1.1ab 1.0ab 1.1a 1.2a

60 1.2a 1.2a 1.3a 1.2a 1.1a 1.2a 1.2a 1.2a 1.2a 1.2a 1.2a

100 1.3a 1.3a 1.3a 1.2a 1.1b 1.1b 1.3a 1.3a 1.2a 1.2a 1.1b

Total nitrogen (%)

0–10 0.9a 0.5b 0.4b 0.4b 0.5b 0.3b 0.5b 0.4b 0.1b 0.5b 0.3b

Means followed by the same letter (s) in row under NF, AF and F are not significantly different (P = 0.05), Where

AF = Agroforestry land use system of 12, 20, 30, 40, 50 years since the conversion of natural forest, F = Farm land of 12, 20, 30,

40, 50 years since the conversion of the natural forest

0

20

40

60

80

100

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0

Dep

th (c

m)

SOC (%)

AF12 AF20 AF30 AF40 AF50 NF

P<0.0001P<0.0024

P<0.0011

Fig. 2 Vertical distribution of SOC in the 1 m depth under

natural forest (NF) and duration of the 12, 20, 30, 40 and

50 years since clearing and in agroforestry (AF) land uses. Error

bars represent standard errors of mean (n = 4)

0

20

40

60

80

100

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0

Dep

th (

cm)

SOC (%)

F12 F20 F30 F40 F50 NF

P<0.001P<0.024

P<0.0011

Fig. 3 Vertical distribution of SOC in the 1 m depth under

natural forest (NF) and duration of 12, 20, 30, 40 and 50 years

since clearing and in farmland (F) land uses. Error bars represent

standard errors of mean (n = 4)

0

50

100

150

200

250

0 10 20 30 40 50 60

Years

SO

C s

tock

(M

g h

a-1)

NFSOC

AFSOC

FSOC

a a a a a

bcd bcdcd

abbc

cdd cd

bcdbcd

Fig. 4 Total SOC stock in the 1 m depth in the duration of 12,

20, 30, 40 and 50 years since clearing and in agroforestry (AF)

and farm (F) land uses, Means followed by the same lower case

letter(s) are not Significantly different at (P \ 0.05)


123

The findings of Lemenih and Itanna (2004) showed

that in the Rift valley of Ethiopia, 50 % of soil C was

retained in the upper 20 cm of the soil. Similar results

were also reported by Awasthi et al. (2005) in the

Mardi watershed of Nepal that corroborates with our

finding.

The concentrations of SOC, and N were higher

under the reference NF than under either AF or F land

uses, probably due to the continuous and greater litter

input from diverse vegetation community compared

with annual crops and little or no soil perturbation

from cultivation. This trend of soil C distribution was

also observed in several other studies (Awasthi et al.

2005; Godsey et al. 2007; Jimqnez et al. 2007). In

general the SOM concentration in the cultivated soils

is less protected than that in the uncultivated soils (Fu

et al. 2004). Soil mixing due to tillage and increase in

soil temperature due to soil exposure and the enhanced

biological activity may lead to accelerated decompo-

sition of SOM.

The higher concentration of the SOC under AF40

compared with F12, F20 and F30 may be due to the

continuous litter input from relatively larger popula-

tion of P. falcatus a widespread coniferous trees

species in and around the AF40 sites. The result is

consistent with the finding reported by Demessie et al.

(2011, 2012a) that the SOC and N stock sequestered in

soils under the coniferous species in the same study

area was higher than the Eucalyptus plantations.

Wheat is often grown in the AF40 compared with

other fields where sorghum and maize are more

frequently produced. Hence the most of root growth

and leftover of crop residue of wheat that remains in

the field after the bulk of it is transported to the house-

holds for cattle feed may also contribute its part for the

higher value of SOC stock. The sorghum and maize

stalks are either burnt in place or are removed for use

in the house-holds during land preparation for the next

growing season. Such management during land prep-

aration process suggests that the input of crop residues

is minimal and not adequate to replace the depleted

SOC during the growing period. The low return of

crop residues can adversely affect soil physical and

chemical properties (Blanco-Canqui and Lal 2008),

and reduce agronomic productivity.

Tree density in AF was not the sole determinant of

SOC stock. Field AF20 had the greatest tree density but

it did not have high SOC compared with other AF

fields. Tree size, species as well as the starting

condition of the field may also be important for the

accrual of SOC stock (Demessie et al. 2012b).

Soil type also influence organic matter turnover due

to differences in soil clay content. Clay is assumed to

protect OM against decomposition and some of the

mechanisms proposed to explain stabilization of SOC

are adsorption of organics onto surfaces of clays or

organic complexes (Oades 1988) and entrapment of

organic particles in aggregates (Van Veen and Kuik-

man 1990). Parent material with high base status and

or the presence of substantial content of Al and Fe

oxides has also positive influence on stabilizing SOM

(Zunino et al. 982; Percival et al. 2000).

Cultivated land had less SOC compared with NF.

Soils have a C sink (storage) capacity, i.e. equilibrium

SOC content depending on the nature of vegetation,

climatic conditions and management (Gupta and Rao

1994). The C flux equilibrium is affected by land use

change until a new equilibrium is established (Guo and

Gifford 2002). The clearance of forests and the

0

5

10

15

20

25

0 10 20 30 40 50 60Years

N s

tock

(M

g h

a-1)

NF AF F

ab ab

a

ab

b

ab ab

ab ab ab

ab

ab

ab

ab

ab

Fig. 5 Total N stock of in the 1 m depth in the duration of 12,

20, 30, 40 and 50 years since clearing and in agroforestry (AF)

and farm (F) land uses, Means followed by the same lower case

letter(s) are not significantly different at (P \ 0.05)

0

1

2

3

4

5

6

7

AF12 AF20 AF30 AF40 AF50 F12 F20 F30 F40 F50Land use

SO

C lo

ss (

Mg

ha-1

yr-1

)

Fig. 6 Loses of soil organic carbon (SOC) for the period since

the conversion of (NF) natural forest to AF (agroforestry) and F

(Farm) land uses in the duration of 12, 20, 30, 40 and 50 years

since clearing


123

subsequent tillage practices affect soil C sink capacity

by: (1) the dwindling supply of litter to compensate for

decomposition, (2) reduced soil capacity to physically

protect SOC from decomposition due to the destruc-

tion of soil aggregates and (3) the enhanced leaching

of DOC or loss of particulate organic C by runoff and

erosion (Post and Kwon 2000; Lal 2002).

In cultivated soils of small land holders such as in

the study region, the amount of crop residues returned

to the soil is only a small portion of the total amount

produced. Even if the crop residue is fully returned to

the soil, it may not balance the quantity and quality of

litter supply that would increase the SOC to its

potential level as it used to be before the conversion of

the NF. This trend may be partly due to high rate of

mineralization (Singh and Lal 2005) and partly

because of the less diverse and lack of continuous

input from crops (Elberling et al. 2003). Most of

cultivated soils in the Midwestern USA have lost

25–40 Mg C ha-1, and their SOC content is below the

potential levels for perennial vegetation (Lal 2002).

Despite the initial SOC stock in AF and F land use

of 12 years of cultivation is not known, the low stocks

compared with the NF suggests that there was rapid

reduction of SOC stock during the early years of

cultivation. There is no evidence of further decline in

SOC stock with more than 12 years of cultivation

indicating a quasi equilibrium for cultivated lands.

The SOC stocks were not higher with AF compared

with F suggesting that the parkland systems as

practiced is not sufficient to shift the SOC equilibrium.

Acknowledgments The financial support to this study through

NORAD to Hawassa University, Ethiopia is gratefully

acknowledged. Our special thanks go to Line Tau Strand, Jan

Mulder and Tore Krogstad for their inspiring comment during

the write up of the manuscript. Our special thanks go to Mr

Teramage Tesfaye for his enormous assistance during the entire

period of data collection and practical work in the field.

References

Albrecht A, Kandji ST (2003) Carbon sequestration in tropical

agroforestry systems. Agric Ecosyst Environ 99:15–27

Ashagrie Y, Zech W, Guggenberger G (2005) Transformation of

a Podocarpus falcatus dominated natural forest into a

monoculture Eucalyptus globulus plantation at Munesa,

Ethiopia: soil organic C, N and S dynamics in primary

particle and aggregate-size fractions. Agric Ecosyst Envi-

ron 106:89–98

Awasthi KD, Singh BR, Sitaula BK (2005) Profile carbon and

nutrient levels and management effect on soil quality

indicators in the Mardi watershed of Nepal. Acta Agric

Scand B 55:192–204

Blakemore LC, Searle PL, Daly BK (1987) Methods for

chemical analysis of soils. New Zealand soil bureau sci-

entific report 80. Soil Bureau, Lower Hutt, New Zealand

Blanco-Canqui H, Lal R (2007) Soil and crop response to har-

vesting corn residues for biofuel production. Geoderma

141:355–362

Blanco-Canqui H, Lal R (2008) Corn stover removal impacts on

micro-scale soil physical properties. Geoderma 145:335–346

Demessie A, Singh BR, Lal R (2011) Soil carbon and nitrogen

stocks under plantations in Gambo District, Southern

Ethiopia. Sustain For 30(6):496–517

Demessie A, Singh BR, Lal R, Børresen T (2012a) Effects of

eucalyptus and coniferous plantations on soil properties in

Gambo District, Southern Ethiopia. Acta Agric Scand Sect

B Plant Soil Sci 62:467–476

Demessie A, Singh BR, Lal R, Børresen T (2012b) Leaf litter

fall and litter decomposition under Eucalyptus and conif-

erous plantations in Gambo District, Southern Ethiopia.

Acta Agric Scand Sect B Plant Soil Sci 62:467–476

Dixon RK (1995) Agroforestry systems: sources of sinks of

greenhouse gases? Agrofor Syst 31:99–116

Duiker SW, Lal R (2000) Carbon budget study using CO2 flux

measurements from a no till system in central Ohio. Soil

Tillage Res 54:21–30

Elberling B, Toure A, Rasmussen K (2003) Changes in soil

organic matter following groundnut-millet cropping at

three locations in semi-arid Senegal, West Africa. Agric

Ecosyst Environ 96:37–47

Ellert BH, Janzen HH, McConkey BG (2001) Measuring and

comparing soil carbon storage. In: Lal R, Kimble JM,

Follet RF, Stewart BA (eds) Assessment methods for soil

carbon. Lewis imprint of the CRC Press, Boca Raton, FL,

pp 131–146

Fu BJ, Liu SL, Chen LD, Lu YH, Qiu J (2004) Soil quality

regime in relation to land cover and slope position across a

highly modified slope landscape. Ecol Res 19:111–118

Godsey CB, Pierzynski GM, Mengel DB, Lamond RE (2007)

Changes in soil pH, organic carbon, and extractable alu-

minum from crop rotation and tillage. Soil Sci Soc Am J

71:1038–1044

Guo LB, Gifford RM (2002) Soil carbon stocks and land use

change: a meta analysis. Glob Change Biol 8:345–360

Gupta RK, Rao DLN (1994) Potential of wastelands for

sequestering carbon by reforestation. Curr Sci Bangalore

65:378–380

Hawando T (1997) Desertification in Ethiopian highlands. Rala

Report No. 200

Janzen HH (2006) The soil carbon dilemma: shall we hoard it or

use it? Soil Biol Biochem 38:419–424

Jimqnez JJ, Lal R, Leblanc HA, Russo RO (2007) Soil organic

carbon pool under native tree plantations in the Caribbean

lowlands of Costa Rica. For Ecol Manage 241:134–144

Kirby KR, Potvin C (2007) Variation in carbon storage among

tree species: implications for the management of a small-

scale carbon sink project. For Ecol Manage 246:208–221

Lal R (2002) Soil carbon dynamics in cropland and rangeland.

Environ Pollut 116:353–362


123

Lemenih M, Itanna F (2004) Soil carbon stocks and turnovers in

various vegetation types and arable lands along an eleva-

tion gradient in Southern Ethiopia. Geoderma

123:177–188

Liao JD, Boutton TW, Jastrow JD (2006) Storage and dynamics

of carbon and nitrogen in soil physical fractions following

woody plant invasion of grassland. Soil Biol Biochem

38:3184–3196

Loveland P, Webb J (2003) Is there a critical level of organic

matter in the agricultural soils of temperate regions: a

review. Soil Tillage Res 70:1–18

Mizota C Van Reeuwijk LP (1989) Clay mineralogy and

chemistry of soils formed in volcanic materials in di-

verseclimatic regions. Soil Monograph 2, Wageningen

ISRIC, pp 186

Murty D, Kirschbaum MUF, Mcmurtrie RE, Mcgilvray H

(2002) Does conversion of forest to agricultural land

change soil carbon and nitrogen? A review of the literature.

Glob Change Biol 8:105–123

Oades JM (1988) The retention of organic matter in soils. Bio-

geochemistry 5:35–70

Paustian K, Andren O, Janzen HH, Lal R, Smith P, Tian G,

Tiessen H, Van Noordwijk M, Woomer PL (1997) Agri-

cultural soils as a sink to mitigate CO2 emissions. Soil Use

Manag 13:230–244

Percival HJ, Parfitt RL, Scott NA (2000) Factors controlling soil

carbon levels in New Zealand grasslands: is clay content

important? Soil Sci Soc Am J 64:1623

Post WM, Kwon KC (2000) Soil carbon sequestration and land-

use change: processes and potential. Glob Change Biol

6:317–327

Sa JCD, Cerri CC, Dick WA, Lal R, Venske SP, Piccolo MC,

Feigl BE (2001) Organic matter dynamics and carbon

sequestration rates for a tillage chronosequence in a Bra-

zilian Oxisol. Soil Sci Soc Am J 65:1486–1499

SAS (2003) SAS/Procedure Guide, Release SAS 9.1.3, Service

Pack 1. SAS Iistitute Inc, Cary, NC, USA

Singh BR, Lal R (2005) The potential of soil carbon seques-

tration through improved management practices in Nor-

way. Environ Dev Sustain 7(161–184):2001

Soil Survey Staff (1999) Soil taxonomy: a basic system of

soil classification for making and interpreting soil surveys,

2nd ed. United States Department of Agriculture, Natural

resource Conservation Service, Washington, DC, 20402

Solomon D, Fritzsche F, Lehmann J, Tekalign M, Zech W

(2002a) Soil organic matter dynamics in the subhumid

agroecosystems of the Ethiopian highlands: evidence from

natural C-13 abundance and particle-size fractionation.

Soil Sci Soc Am J 66:969–978

Solomon D, Lehmann J, Mamo T, Fritzsche F, Zech W (2002b)

Phosphorus forms and dynamics as influenced by land use

changes in the sub-humid Ethiopian highlands. Geoderma

105:21–48

Takimoto A, Nair PKR, Nair VD (2008) Carbon stock and

sequestration potential of traditional and improved agro-

forestry systems in the West African Sahel. Agric Ecosyst

Environ 125:159–166

Takimoto A, Nair PKR, Nair VD (2009) Contribution of trees to

soil carbon sequestration under agroforestry systems in the

West African Sahel. Agrofor Syst 76:11–25

Tan KH (2005) Soil sampling, preparation, and analysis, 2nd

edn. Tailor & Francis Group, Boca Raton

Turner J, Lambert MJ, Johnson DW (2005) Experience with

patterns of change in soil carbon resulting from forest

plantation establishment in eastern Australia. For Ecol

Manage 220:259–269

Van Veen JA, Kuikman PJ (1990) Soil structural aspects of

decomposition of organic matter by micro-organisms.

Biogeochemistry 11:213–233

Walkley A, Black IA (1934) An examination of the Degtjareff

method for determining soil organic matter, and a proposed

modification of the chromic acid titration method. Soil Sci

37:29

Yimer F, Ledin S, Abdelkadir A (2007) Changes in soil organic

carbon and total nitrogen contents in three adjacent land

use types in the Bale Mountains, south-eastern highlands of

Ethiopia. For Ecol Manage 242:337–342

Zunino H, Borie F, Aguilera S, Martin JP, Haider K (1982)

Decomposition of 14C-labeled glucose, plant and micro-

bial products and phenols in volcanic ash-derived soils of

Chile. Soil Biol Biochem 14:37–43


123

Soil carbon and nitrogen stocks under chronosequence of farm and traditional agroforestry land uses...

Documents

Transcript of Soil carbon and nitrogen stocks under chronosequence of farm and traditional agroforestry land uses...