The soil is a key component of the global carbon cycle. Soil Organic
Carbon (SOC) stock is the biggest ecosystem carbon reservoirs in the world;
it plays a critical role in mitigating the greenhouse effect. The amount
of Organic Carbon (OC) contained in the worlds soils is comprised
between 1500 and 2000 PgC in the upper 100 cm (Eswaran
et al., 1993; Batjes, 1996). A good
estimation of carbon pools in the soil has been suggested as a mean to
help mitigate atmospheric CO2 increases and anticipated changes
in climate (Batjes and Sombroek, 1997; Lal
et al., 1998, 2000). Enhancement of
Soil Organic Carbon (SOC) and its good management is very important for
agriculture. It increases the fertility, as well as by carbon sequestration
we attain healthy environment and generally, organic matter in soil stabilizes
soil structure and makes it more resistant to degradation (Mbah
et al., 2007). Reliable inventories of SOC stocks are primordial
information to help countries in fulfilling their obligations under the
National United Nations Framework Convention on Climate Change (Tompkins
and Amundsen, 2008). Moreover, as SOC is central for ecosystem functioning,
this information is also relevant to the United Nations Conventions to
Combat Desertification as areas with low SOC content are generally synonymous
with land degradation, which is a major issue (Bai
et al., 2008). Most national evaluations of SOC stocks and
their repartition were performed in temperate (Arrouays
et al., 2001) or tropical regions (Bernoux
et al., 2002; Batjes, 2005) except
few notable exceptions in semi arid zones (Batjes,
2006; Al-Adamat et al., 2007). Even
if it is commonly admitted that SOC per unit area of drylands is low compared
with other terrestrial ecosystems, the large area of drylands turns carbon
sequestration potential of total drylands important. For Tunisia, it is
important to assess the pools of SOC for several reasons. Organic carbon
is one of the most important constituents of soils; it has a main agronomic
and environmental interest. Organic carbon improves soil structure, capacity
in affecting vegetation development and it mediates many chemical and
physical properties. Also, organic carbon storage in Tunisian soils reflects
the capacity of arid and semi-arid regions to sequester organic carbon.
This study provides a first estimate of SOC stocks for the 0 to 30 cm
and 0 to 100 cm depth in Tunisian soils.
MATERIALS AND METHODS
This study was conducted in Tunisia from 2005 to 2008. Tunisia (31°38°N;
7°12°E and 164 000 km2) is situated in North of Africa
and South of Mediterranean Sea (Fig. 1) and has a wide
range of natural regions.
||Location of Tunisia in the Mediterranean basin and soil
map with simplified legends
Three dominant climatic zones illustrate the country and reflect the
influence by the Sea and Sahara desert: (1) the Northern region is humid
(600-1200 mm year-1) occupied by rainforest still; (2) the
central region is semi-arid (200-600 mm year-1) steppe is here
dominant vegetation; (3) the Southern region is arid its a desert
(<200 mm year-1). Thus, soil present an important variability
Tunisian soil literature from 1960 to 2006 was searched for data on
soil profiles. Chosen profiles had variable depth, but they were usually
more than 1 m in depth. A database was built with previous analytical
results available in soil profile information reported for soils pits
surveyed by teams of Tunisian and IRD (ex-ORSTOM) pedologists. The database
was made of 1483 soil profiles, corresponding to 5024 soil horizons. Soils
profiles were classified according to the nine main soil groups used in
the soil map and their locations (Delegation and Governorate) were checked.
The SOC stocks from 0.3 to 1 m were calculated, (formulae is given at
section 2.5) when possible, per profile using data on soil bulk density,
proportion of organic carbon and volume of fraction>2 mm. The list
of soil properties available in the database and used in predictive equations
are given in Table 1. Gaps in the available data, mostly
for bulk density and C content for some deeper horizons were filled using
specific pedotransfer functions (Bernoux et al.,
1998). The procedure used is fully described by Bernoux
et al. (2002) (Table 1).
The 1:500,000-soil map of Tunisia (Tunisian Ministry
of Agriculture, 1973) was used as spatial component. In total, the
map is made of 34049 map units that were distributed into nine main soil
orders plus one non-soil (water surface and urban soil) category (Table
||List of soil properties available in the database and
used in predictive equations
||Soil categories and their relation to the original soil
classes of soil map
The map for the period between 2006-2007 was digitized. The number of
polygons or Soil Map Units (SMU) were 34049. Soil Map Units were classified
to the FAO-UNESCO classification (FAO, 1974). The
minimum surface represented is 0.34 ha, but the maximum surface represented
is 9777.77 ha.
The estimation of SOC stocks for a certain area requires estimations
of mean soil organic carbon concentration and bulk density (Db).
Generally, Db is not determined in most routine analyses
and for most of the Tunisian soil profiles in the database no Db
was reported. The Db of only 707 soil horizons from
the 5024 records have been measured and it is therefore, necessary to
estimate Dbs for the rest of the horizons. To
this end, so values have to be determined using Pedotransfer Functions
(PTF) (Batjes, 1996; Bernoux
et al., 1998). Using all the available parameters, results
showed that the OC content was always the best predictor of Db,
accounting for up to 34% of the variation, the best Multiple Linear Regression
(MLR) resulted in the following equation:
Subdividing soils into groups by soil depth resulted in more accurate
Db predictions for soil layers. The MLR for superficial layers
(≤30 cm) were:
and for deep horizons layers (>30 cm):
Procedure for Determining the Individual SOC Stocks
The way of calculating SOC stocks for a given depth consists of summing
SOC stocks by layer determined as a product of Db, OC
concentration and layer thickness (Eswaran et al.,
1993; Batjes, 1996; Bernoux
et al., 2002). For an individual profile with n layers, we
estimated the organic carbon stock by the following equation:
where, SOC is the soil organic carbon stocks (kgCm-2), Dbi
is the bulk density (Mg m-3) of layer i, Ci is the proportion
of organic carbon (gCg-1) in layer i and Di is the thickness
of this layer (cm).
Bernoux et al. (2002) used a similar procedure
for calculating carbon stocks in soil of Brazil. In the second step of
calculation, OC densities of each great group were multiplied with their
respective area (Batjes, 1996) for quantification
of soil organic and inorganic carbon stocks. Summation of individually
of carbon of the nine great soil groups gave total carbon stock in Tunisia.
RESULTS AND DISCUSSION
Distribution of SOC Density and SOC Storage in Tunisia
Statistical results (Table 3) based on big soil
orders, indicated that SOC density varied considerably across soil types.
Result further showed that in 0-30 and 0-100 cm depth, Luvisols have the
highest SOC densities of 7.16 and 15.92 kgCm-2, respectively,
but Lithosols have the lowest SOC densities, at 0-30 and 0-100 cm it have
1.84 and 4.04 kgCm-2. Given a total area of 15520249.8 ha of
soil in Tunisia, summation of all soil map units yielded a total SOC storage
of 405.44 TgC in the 0 to 30 cm soil depth and 1006.71 TgC in the upper
layer 0-100 cm and a mean SOC density of 2.612 and 6.486 kgCm-2 at
0-30 and 0-100 cm, respectively.
Elaboration of Maps of SOC Density
In order to appreciate the geographical distribution of SOC densities
and its pattern it is useful to create a map of SOC contents. The SOC
map was derived by intersection of soil map and soil database. Figure
2 showed that soil and climatic zone have different influences on
the organic carbon stock distribution, depending of the geographical localization.
For example, the regions with the highest organic carbon stock has a soil
influence marked by the presence of forest soils. On the contrary, the
Southern part of Tunisia had low carbon stock mostly because the arid
climate influences the vegetation and the soil organic matter decomposition
(Bernoux et al., 2002). The Northern region
is characterized by high carbon stock and showed an important climatic
influence, humid zone.
SOC is very spatially variable at the scale of the map. This could have
been easily anticipated, given the large spatial heterogeneity of climate
and geology, which determine the storage of organic carbon in soil (Fig.
These stocks are consistent with data for the world level (Batjes,
1996) derived from the WISE (World Inventory of Soil Emission Potentials)
soil database. Batjes (1996) reported worldwide
mean carbon stock values for the 0 to 30 cm layer of 3.1, 4.5 and 5.0
kgCm-2 for Regosols, Vertisols and Cambisols, respectively.
It accounted for 0 to 100 cm depth of 9.6, 11.1 and 9.6 kgCm-2 for
Kastanozems, Vertisols and Cambisols, respectively.
Independent of the method used, total SOC stocks for Tunisia were comparable.
The calculated SOC stocks to 0-30 cm were closed to the amount (498 TgC)
reported by Henry et al. (2009) for this
country but using a world database, however the estimated stocks to 1
m were higher than the result (727 TgC).
||Soil organic carbon (SOC) density and storage by soil
order in Tunisia
|n: Number of soil profiles existing in database. SD:
||Map of SOC density of Tunisia, (a) in 0-30 cm depth
and (b) in 0-100 cm depth
These stocks are comparable with data for the world level derived from
the FAO soils database. Henry et al. (2009)
used 407 soil Mapping Units (MU) for Tunisia it estimate 498 and 727 TgC
from 0-30 and 0-100 cm, respectively.
Soils in Tunisia stored 1006.7 TgC and a mean SOC density of 6.486 kgCm-2
within the 100 cm soil depth and 405.44 TgC in the upper layer 0-30
cm within a mean SOC density 2.612 kgCm-2. Soil organic carbon
is very spatially variable at the scale of the maps. This could have been
easily anticipated, given the large spatial heterogeneity of climate,
geology and land use in Tunisia, which determines inter alia the storage
of organic matter in soils. Zones with a high OC content correspond generally
with areas of high rainfall; natural forests and mountain ranges in the
North of Tunisia, whereas the South with low rainfall show little SOC
density. Due to application of the calculated profile values method for
estimating SOC density and linkage with soil map, the results of this
first study for estimation Tunisian SOC stock were accurate and reliable.
Thus, information obtained in this study about SOC storage and density
of all soil orders, will be a first step accurately estimating and monitoring
of the changes of SOC storage in Tunisia.
This study is part of the CORUS-2 n°6112 co-financed project Séquestration
du carbone et biodiversité dans les sols africains méditerranéens
et leurs vulnérabilités aux changements climatiques.
The authors thank the anonymous reviewers for their helpful comments and
suggestions for the development of the manuscript.