Dynamics of soil organic carbon stock under different types of savannah agrosystems in the Sudano-Sahelian zone of Cameroon

Article Info The aim of this study was to quantify the current soil organic carbon stock under different types of savannah agrosystems in the Sudano-Sahelian zone of Cameroon in the context of greenhouse gas emissions and land degradation. It is so crucial for combating climate change and improving ecological restoration. Random field sampling was carried out on 0-10, 10-20 and 20-30 cm depth, then were collected in four types of savannah agrosystems. Soil bulk density, pH, moisture content, CEC, exchangeable bases, particle size distribution and soil organic carbon were determined using standard laboratory procedures and calculations. The results of the study did not reveal a significant difference in soil organic carbon stock between different types of savannah agrosystems (P>0.05). Soils of Tamarindus indica savannah agrosystems in recorded higher values SCOS (36.03 ± 3.31 tC/ha), Prosopis africana (33.40 ± 3.27 tC/ha), Haematostaphis barterii (31.83 ± 3.21 tC/ha) and Detarium microcarpum (31.19 ± 3.19 tC/ha) savannah agrosystems. Similarly, SCOS decreased with soil depth in all types of savannah agrosystems. Results showed a positive and significant (P<0.05) correlation between soil organic carbon stock with basal area, biovolume, bulk density, moisture content, C/N ratio, Ca2+, Mg2+, OM; negative and significant (P<0.05) with Soil pH, Total Nitrogen, Na+ but negative and non-significant (P>0.05) with Density, K+, CEC, Sand %, Silt %, Clay %, Silt + Clay %. The results show the potential contribution of savannah agrosystems to improve soil organic carbon sequestration and environmental protection. Received : 01.05.2020 Accepted : 07.10.2020 Available online : 12.10.2020 Author(s) A.D. Victor * N.N. Valery


Introduction
Soil is the loose surface layer of the earth's crust. It is also defined as a natural environment with essentially dynamic properties, differentiated into horizons with mineral and/or organic constituents that are generally loose, resulting from the transformation of an underlying parent rock, under the influence of various chemical, physical and biological processes (FAO, 2017). It is the place where plant roots develop. The increase in population around the world is accompanied by an increase in agricultural production needs,

Field Description of the Study area
The study was carried out in the north region (Cameroon). The zone extends between 8° and 10° North latitude and between 12° and 16° East longitude, and is bounded to the North by the Far North region, to the South by the Adamawa region, to the East by the Republics of Chad and Central African Republic and to the West by the Federal Republic of Nigeria (Awé et al., 2019c). The north Cameroon region has a tropical climate of the Sudano-Sahelian type. Average monthly temperatures are between 25.4 and 32.5 °C. Each year, precipitation averages 1003 mm. The relief is a vast pediatric plain between the Mandara Mountains (1,442 m) in the North and the Adamawa Plateau in the South. The soil is of ferruginous type formed by degradation of sandstone from the Middle Cretaceous (Awé et al., 2020). The vegetation encountered is a shrubby Sudanian savannah with a clear and degraded savannah appearance (Awé et al., 2020). The fauna is rich and very diverse (Awé et al., 2019b). Economic activities concern: agriculture, animal husbandry, fishing, social economy and handicrafts, transport and trade. Agriculture is the main activity of the populations of the North region (Cameroon) (Figure 1).

Data collection
Transects 80 m long by 25 m wide were installed at each site and each transect is spaced 10 m apart. A total of 4 transects were installed for a total sampling area of 1 ha per site. Sampling strips were established using compass, tape measure, GPS and twine. At the ends of each strip, stakes were planted equidistantly 20 m apart. Along the transect, all woody trees of Dbh ≥ 10 cm were surveyed in the four selected savannah agrosystems (Detarium microcarpum, Haematostaphis barterii, Prosopis africana and Tamarindus indica). For the calculation of vegetation structure two parameters were taken into account: tree density, basal area and biovolume. For the density of woody plants, we applied the formula below: D= n/S with D: density (trees/ha), n: number of trees present on the area considered and S: area considered (ha). For the basal area, we applied the formulas below: S = π (Di²/4) with S: basal area (m²/ha) and Di: diameter (m). The biovolume is given by the formula of Dawkins (1959): Bv= 0,53 agi x hi x ni with gi: basal area (m 2 /ha); Hi: height of trees (m); ni: number of trees; Bv: biovolume (m 3 /ha). According to (Roger andRabarison, 2000 in Awé et al., 2019c), the biovolume is high when it is above 250 m 3 /ha, medium when it is between 50 and 250 m 3 /ha, and low when it is below 50 m 3 /ha. Soil samples are taken from January to March. In each 2000 m² survey, soil samples were taken in 0.25 m x 0.25 m frames. These samples are taken at 0-10 cm, 10-20 cm, 20-30 cm depth on the four elementary plots. Each level of soil depth was sampled using a machete and trowel and then immediately put in a closed bag in a cooler, in the shade to avoid evaporation. A total of 3 samples were taken per drilling unit, which corresponds to a total of 12 samples per site and then homogenized to obtain an aggregate sample. A total of 48 samples (4 sites x 3 depths x 4 replicates x 1 area) for all four sites were dug into the ground to a depth of 30 cm. Once all samples were collected, they were taken for laboratory analysis. The laboratory method consists of determining, evaluating or measuring the physico-chemical parameters of the soils:

Bulk density
The determination of the bulk density was carried out by sampling a defined volume of soil using a cylinder driven into the ground. After drying the sample in an oven at 105°C for 48 hours, it was weighed again. The dry weight of the sample P divided by the sample volume (V) gave the bulk density (Da) in g/cm 3 . It is calculated using the following formula Da=P/V; was done according to the NF ISO 11464 Standard (AFNOR, 2006).

Determination of pH
The pH measurement was carried out on a sol-water solution for the pH water and a sol-KCL solution for the pH in a ratio of 1/2.5 using a PH-meter with a glass electrode. The pH meter was previously calibrated using the standard solutions according to the NF ISO 10390 standard (AFNOR, 2005).

Determination of the moisture content at 105°C
The moisture content at 105°C which allows to estimate the water content was done according to the NF X15-110 standard (AFNOR, 1994). It consists in introducing 5 g of the fresh sample into a previously tared flask, then let the soil sample dry in the oven at 105°C for 24 h; then let it cool in a desiccator and weigh. The equivalent moisture is thus determined by the following formula: H= (P gross air-dried) -(P gross air-dried at 105°C) / (P net air-dried) x 100. Soil texture analysis Soil texture analysis was determined by the Robinson's pipette method on air-dried soil samples sieved at 2mm. The organic matter was previously destroyed by attack with hydrogen peroxide. The sol was then dispersed by rotary shaking in flasks after addition of sodium hexa-metaphosphate (NaPO3)6. The different particle size fractions were determined by pipetting for the clayey and silty fractions and by sieving for the sand (AFNOR, 2003).

Determination of Total Nitrogen
The total Nitrogen was obtained through the (Kjeldahl, 1883) method after heat treatment of the sample with a mixture of sulphuric acid (H2SO4) and salicylic acid (C6H4(COOH)(OH). The nitrates present in the sample were first fixed by the salicylic acid and then reduced to ammonia by the use of a catalyst consisting of copper sulphate (CuSO4). The distillate was captured in boric acid (H3BO3) and then titrated with sulphuric acid (H2SO4) according to the NF EN ISO 23470 Standard (AFNOR, 2011).

Determination of Exchangeable Bases
Exchangeable bases were extracted from the soil with a solution of Ammonium Acetate (C2H3O2NH4) at pH7. The concentrations were made by atomic absorption spectrometry (Magnesium) and by flame emission (Calcium, Potassium, Sodium) according to the NF X31-108 standard (AFNOR, 2002). The K, Mg, Na and Ca contents are converted into kg/ha. Cationic exchange capacity (CEC): This was done with ammonium acetate at pH7 and notably in three phases: saturation of the absorbent complex by NH4 + ions and extraction of the exchangeable bases; washing of the soil with alcohol in order to eliminate excess NH4 + ions; determination of NH4 + by Kjéldahl distillation after desorption from a KCL solution according to the NF EN ISO 23470 standards (AFNOR, 2011). Soil organic carbon was determined by (Walkley and Black, 1934) method, which is an oxidation with potassium bicarbonate (K2Cr2O7) in an acid medium (H2SO4) according to the NF ISO 14235 standard (AFNOR, 1998). The dosage was done by calorimetry. The organic matter content was obtained by multiplying the organic carbon rate by the Sprengel factor which is 1.724 for cultivated soils and 2 for uncultivated soils. Soil carbon (SCOS) (tC/ha) = Da. (% COS). S. P (Awé et al., 2020) with Da: bulk density in tones /m 3 ; COS%, organic carbon content of the soil; S: area in m 2 ; p: depth m.

Data analysis
The data were encoded in EXCEL software and then analyzed using STATGRAPHICS plus 5.0 and R software. Correlation and significance tests were performed using ANOVA and Duncan's 5 % test.

Soil physical characteristics
The highest density was recorded in Detarium microcarpum (310 ± 10.10 stems/ha) savannah agrosystems (Table 1). This high density means that the stems used to reconstitute the environment are shrubs. This result lies in the range 208 ± 8.57 -408 ± 11.12 individuals/ha found by (Awé et al., 2019c) in savannah agrosystems in the Sudano-Sahelian zone of Cameroon. The highest values of basal area (11.50 ± 1.65 m²/ha) and biovolume (48.65 ± 3.95 m 3 /ha) were recorded in the Tamarindus indica savannah agrosystems (Table 1). This indicates the existence of large specimen trees on the one hand and a significant timber potential due to their large diameters on the other. The basal area and biovolume values obtained in this work are respectively in the range, 2.94 ± 0.13 -11.56 ± 0.57 m²/ha and 32.94 ± 3.03 -116.78 ± 16.57 m 3 /ha found by (Awé et al., 2019c) in savannah agrosystems in the Sudano-Sahelian zone of Cameroon. The analysis of variance shows that there is no significant difference in density (P=0.321), basal area (P=0.123) and biovolume (P=0.532) between the different types of savannah agrosystems studied (Table 1). The granulometric distribution made it possible to distinguish 4 textural classes including clay-sandy, fine silt, clayey and clay-silt soils. The analysis of variance relating to soil textural fractions (Clay: P = 0.0268; Silt: P = 0.0000 and sand: P = 0.0004) show that there is a variation in soil textural composition according to the different savannah agrosystems studied (Table 2). In fact, clay soils have a more acidic pH than sandy soils (Carrier, 2003). Bulk density varies with depth. The highest values of the bulk density values were recorded at depths of 0-10 cm. The highest bulk density value was recorded to Tamarindus indica (1.61 ± 0.16 g/cm 3 ) savannah agrosystems. This may be due to soil compaction which is contrary to the other three in the savannah agrosystems studied where the soil is loosened due to fine root mat, microbial and arthropod activities leading to soil aeration. The analysis of variance shows that there is no significant difference between depths (P=0.085) on the one hand and between savannah agrosystems (P=0.065) on the other hand (Table 3). Moisture content varies with depth. The highest values of humidity were recorded at depths of 20-30 cm. Among savannah agrosystems, the highest value was recorded in Tamarindus indica (27.61 ± 2.39) savannah agrosystems. This may be influenced by the vegetation cover; Detarium microcarpum savannah agrosystems being much more exposed to solar radiation. The texture of these soils may also influence its moisture content. Indeed, a sandy soil allows water to pass easily while a clay soil retains water (Coudurier and Bourgogne, 2012). As for pH, it is more acidic in forest soils. Tree growth involves taking ions from the soil and releasing others with identical electrical charges in order to maintain their electrical balance (Munguakonkwa, 2018). Since they require more cations than anions, their growth releases many cations (often H + ) into the soil, making it more acidic (Ranger, 2018). The analysis of variance shows that there is no significant difference between depths (P=0.238) on the one hand and between savannah agrosystems (P=0.312) on the other hand (Table 4).

Soil chemical characteristics
Soil reaction (pH) varies with depths. The highest values of the soil reaction were recorded at depths of 0-10 cm. Among savannah agrosystems, the highest value was recorded in the Detarium microcarpum (pH = 6.42 ± 1.42) savannah agrosystems. This can be explained by clearing by burning which brings large amounts of ash to the soil which can increase the initial pH. This result is in the range 4.5 to 6.5 (Dabin, 1985). The analysis of variance shows that there is no significant difference between depths (P=0.505) on the one hand and between savannah agrosystems (P=0.192) on the other hand (Table 5). The analysis of variance did not show a significant difference in nitrogen content between savannah agrosystems (P=0.544). Soils in Detarium microcarpum savannah agrosystems had the highest values of nitrogen content (6.64 ± 1.31 kg/ha). At only 5%, analysis of variance (P=0.360) did not reveal a significant difference between the C/N ratios of the different savannah agrosystems studied. Soils in Tamarindus indica savannah agrosystems have the highest C/N ratios (12.76 ± 5.03) ( Table 6). Soils of Detarium microcarpum savannah agrosystems have a low biological activity (C/N greater than 12) and therefore a slow rate of OM decomposition, whereas for Detarium microcarpum, Haematostaphis barterii, Prosopis africana savannah agrosystems, this rate is higher with normal values (C/N between 8 and 12). Several other factors would explain these variations in C/N ratios such as particle size and pH (Decoopman et al., 2013).  Soil organic matter content in the savannah agrosystems studied decreased with depth of sampling. The highest values of soil organic matter content were observed at a depth of 0-10 cm. Detarium microcarpum Savannah agrosystems had the highest values of soil organic matter content (2.34 ± 0.13). The analysis of variance did not reveal any significant difference in soil organic matter content between depths on the one hand (P= 0.553) and between savannah agrosystems on the other (P= 0.548) ( Table 8).

Soil organic carbon stocks
The highest values of soil organic carbon stocks were observed at a depth of 0-10 cm. The analysis of variance did not reveal any significant difference in soil organic carbon stocks between depths on the one hand (P= 0.207) and between savannah agrosystems on the other hand (P= 0.261) ( Table 9). Soils in Tamarindus indica savannah agrosystems (36.03 ± 3.31 tC/ha) are those that store more carbon than those in other types of savannah agrosystems. This result is in the range 24.65 ± 2.51-42.97 ± 4.35 tC/ha reported by (Awé et al., 2019c) for Burkea africana savannah agrosystems in the sudano-sahelian zone of Cameroon. Vegetation types can alter soil carbon stocks due to several key factors, including litterfall and root turnover, soil chemistry, root exudates, and microclimate (Ontl and Schulte, 2012;Awé et al., 2019c). Low carbon stocks of Detarium microcarpum savannah agrosystems are explained by the fact that agricultural practices such as deforestation, turning and frequent tillage, etc., cause a decrease in soil carbon stock (Swiderski et al., 2012).

Relationship between soil organic carbon stock and soil pysico-chemical characteristis
Soils with high carbon stock are clay loam soils (36.03 ± 3.31 tC/ha) followed by clay soils (33.40 ± 3.27 tC/ha); fine loam soils (31.83 ± 3.21 tC/ha) and clay-sandy soils (31.19 ± 3.19 tC/ha). The analysis of variance did not reveal a significant difference in soil carbon stock between textural classes (P= 0.164) (Figure 2). Results showed a positive and significant correlation (P<0.05) between soil organic carbon stock with basal area, biovolume, bulk density, moisture content, C/N ratio, Ca 2+ , Mg 2+ , OM (Table 10). This marks a dependence effect in the variation of SCOS. Also the negative and significant (P<0.05) correlation with Soil pH, Total Nitrogen, Na + would show an inverse and dependence effect with SCOS (Table 10). Finally, the negative and non-significant correlation (P>0.05) of SCOS with Density, K + , CEC would not reflect any dependence effect (Table 10). Results showed a negative and non-significant (P>0.05) correlation between soil organic C stock with % Sand, % Silt, % Clay, % Silt + Clay according to the three depth ranges of 0-10 cm, 10-20 cm and 20-30 cm respectively (Table 10). Soil organic carbon stocks decreased with increasing depth in all types of savannah agrosystems, as indicated in several results (Agboadoh, 2011;Jiao et al., 2012;Bessah et al., 2016;Awé et al., 2019c). On the other hand, there is no correlation between the density of savannah agrosystems and the amount of organic carbon sequestered in the soil. This can be explained by the presence of large trees in the savannah agrosystems studied. Organic carbon stock depends on basal area and biovolume in savannah agrosystems.

Conclusion
This study gives us a better understanding of the soil organic carbon stock in the savannah agrosystems studied. Soil is a non-renewable resource whose quality must therefore be preserved for its environmental functions. The results show that the soil organic carbon stock is higher in Tamarindus indica savannah agrosystems. However, the evolution of COS stocks is more or less decreasing as the savannah agrosystems evolve. From all the soil physico-chemical parameters measured, only bulk density, moisture content, C/N ratio, Ca 2+ , Mg 2+ , OM show a strong and positive linear correlation with soil carbon stock among all the physico-chemical parameters measured. Soil physico-chemical parameters (texture, total nitrogen, C/N ratio, pH, soil bulk density, moisture content, CEC, exchangeable bases) also vary according to the types of savannah agrosystems.