Assessment of the Soil Organic Carbon Stock (SOCS) of Foumban (West Cameroon)

doi:10.21203/rs.3.rs-3093484/v1

Download PDF

Research Article

Assessment of the Soil Organic Carbon Stock (SOCS) of Foumban (West Cameroon)

https://doi.org/10.21203/rs.3.rs-3093484/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

This study aims to assess the organic carbon stock and the amount of organic carbon sequestered in the soils of Foumban, in order to facilitate climate change mitigation policies. To achieve this objective, ninety soil samples were collected from a depth of 0-30cm from the reference soil groups identified in the study area and transported to the laboratory where bulk density and organic carbon analyses were carried out to estimate the SOC stock. The SOC stock and carbon quantity calculations showed that the carbon stock in the soils of the study area varies from 64.10 t/ha in FRAGISOLS (plaggic) cambic to 116.52 t/ha in UMBRISOLS pretic, with an overall average of 88.95 ± 18.61 t/ha. The total amount of organic carbon stored in the soils is estimated to be about 85.79×10^− 4 GT. It varies from 1.30×10^− 4 GT in cambic hortic FERRALSOLS (plaggic) on trachytes to 17.13×10^− 4 GT in chernic arenic UMBRISOLS. The variation in carbon stock in the study area would be due to the variation in the density of the vegetation cover. The input of organic amendments would contribute to the increase of organic carbon stocks in these soils.

Organic Carbone

Land use

Geostatistics

Spatial variability

Kriging

In recent years, research on climate change has advanced remarkably (Durot, 2013). The increase in CO₂ concentration has a direct influence on the effectiveness of greenhouse gases in inducing global warming and climate change (Muoghalu, 2014). The natural movement of carbon in the atmosphere, vegetation, soils and oceans is key to mitigating climate change due to increasing CO₂ (Chaikaew and Chavanitch, 2017). The organic carbon stored in the top few meters of soil in most terrestrial ecosystems is about 2–4 times the amount found in vegetation (1500 Pg carbon) (Hougthon, 2007; Jobbagy and Jackson, 2000). Vegetation plays a key role in SOC stocks through the contribution of decomposing plant biomass, particularly from the Arabian layers (Bui et al., 2009) and vegetation indices such as NDVI have been used to predict SOC stocks (Yang et al., 2016b). In Cameroon, anthropogenic action is marked by uncontrolled and anarchic deforestation resulting in part from slash-and-burn agriculture (Vroh et al., 2014).

In the locality of Foumban and its surroundings, agriculture remains the main activity contributing to the economy of the population. On a socio-economic level, this area is marked by a significant dynamic of land use change due to agricultural activities in addition to those linked to legal or illegal logging. Agriculture is almost extensive and is practiced on small family farms of three to six hectares (Kambire et al., 2015). This agricultural practice, based increasingly on intensification (monoculture, use of chemical fertilisers) in the long term, correlated with unsustainable management, cannot be without consequences for the environment (Dussault, 2008; Kambire et al., 2015). Organic carbon levels in the soil are very often influenced by many environmental factors (climate, topography, altitude, rock, vegetation cover, type of plants, land use, human activities...) (Lozano-Garcia and Alcantara, 2013; Parras-Alcantara et al., 2014). These factors can contribute effectively and in the short term to variations in soil SOC stock.

SOCS play an important role in carbon cycling and soil fertility management. The SOC parameter is normally used as an indicator of soil quality (Ruiz Sinoga et al., 2012) and can accumulate in the soil for decades. A better understanding of the spatial variation of SOCS is essential for sustainable management of soil resources. This study provides a solid baseline for land management and decision-making regarding soil quality, SOC accounting and SOC sequestration potential.

2 .1. Study area, land use and sampling points

The study area extends between the northern parallels 5°39' to 5°52' and the eastern meridians 10°42' to 11°00', covers an area of 793.27 km² (Fig. 1a), and is located in the central part of the Noun valley in western Cameroon. The study area has a humid tropical climate at high altitude. The annual rainfall is 1994.2 mm with a temperature between 22°C and 24°C. It is drained by two large catchment areas: the Mfu basin and the Nchi basin. Both show a sub-parallel hydrographic network. The morphology is composed of two main domains: the lower domain with altitudes between 750 and 1250 m, consisting of small rounded peaks with fairly wide U-shaped talwegs; the upper domain with altitudes between 1250 and 1850 m, dominated by the Nkogham massif.

Figure 2a below is a land use model of the study area obtained from Landsat 8 satellite images using Google Earth software, digitized and processed using ArcGis 10.1 software. This figure shows that forest cover covers 38.7% of the area, i.e. an area of approximately 306.99 km². It is more present in the north and east of the study area. In the west of the area, it is present at the top of the Nkogham massif. Bare soil (steppe, cultivated area) covers 42.9%, i.e. an area of about 340.31 km². They are more present in the center of the study area and to the west. Habitat represents only 18.4% of the sector, i.e. an area of approximately 145.96 km². It is most important in the center of Foumban, although some dwellings are observed in the surrounding villages.

The field campaign consisted of a soil survey on pits whose depth varied from 0 to 30cm in the arabian layers as indicated by Bui et al. (2009). Then, soil samples were taken from the walls of these pits along a cultural profile. The samples were collected by soil group using the soil map established by Lotse et al. (2022) from the WRB (FAO-ISRIC, updated 2022). These samples were collected considering lithology (parent material), morphology and reference soil group. In total, ninety (90) soil samples were taken randomly (Fig. 2b). That is, three from Cambic (plagic) FRAGISOLS, sixteen from Plagic (arenic) Hortic NITISOLS, twenty-one from Chernic arenic UMBRISOLS, seven from Umbric arenic FERRALSOLS, four from Mollic fragic NITISOLS, sixteen on cambic (plaggic) horticultural FERRALSOLS on basalts, ten on umbric (haplic) pisoplinthic PLINTHOSOLS, four on nitique (pretic) CHERNOSOLS, three on cambic (plaggic) horticultural FERRALSOLS on trachytes, six on pretic UMBRISOLS. It should be noted that the amount of sample taken per soil group depends on the surface area of the soil group, accessibility and land use (housing, vegetation and others). The sampling device is designed to meet geostatistical concerns. Once the sample is taken, it is transported to the laboratory where the physico-chemical analyses are carried out.

In the laboratory, the chemical analyses carried out were those of organic carbon. The physical analyses carried out were those of bulk density. Organic carbon is determined by the Walkley and Black (1994) method, which is an oxidation with potassium bicarbonate (K₂Cr₂O₇) in an acid medium (H₂SO₄). The determination is done by calorimetry. The determination of the bulk density is obtained from the sampling of a defined volume of soil with a cylinder driven into the soil. After drying the sample in an oven at 105°C for 48 hours, it is weighed again. The dry weight of the sample P divided by the volume of the sample V gave the bulk density (Da) in g/cm³; according to the formula:

Da = P/V……………………………………………………………………………….(1)

To assess the organic carbon stock of soils as a function of area, three parameters were taken into account to determine the amount of carbon available in the horizon (Belkacem et al., 1998; Peng et al., 2000; Evans et al., 2001; Cerri et al., 2007).

q(i) = 0.1 * Ei * da(i) * Ci............................................................................................ (2)

where q(i): organic carbon content in soil horizon i (t.ha-1), Ei: thickness of horizon i (cm), da(i): bulk density of the fine fraction (< 2mm) in horizon i (g.cm^− 3), Ci: the concentration of organic carbon in the fine soil of horizon i (g.kg-1) by the following relation:

1CO (%) = 10CO (g.kg^− 1)……………………………………………………………..(3)

To calculate the amount of organic carbon sequestered by a soil group, two parameters were taken into account. These are the amount of carbon in tonnes per hectare (t/ha) for a given area, and the total area in hectares (ha). The total amount is calculated from the following formula:

Qc = Q * Area................................................................................................................(4)

With Q, the amount of carbon in tonnes per hectare (t/ha), Area: area covered by the soil group in hectares (ha); and Qc: the amount of carbon expressed in tonnes (t) or in (Gt).

2.2 Statistical and geostatistical analyses

For the statistics, the data obtained after analysis of the samples in the laboratory were processed using Excel 2016 and IBM SPSS Statistic 20. A descriptive statistical analysis (mean, standard deviation etc.) was carried out to describe the soil properties involved in the SOCS estimation.

The spatial distribution of SOC and bulk density (BD) in the soil was made possible by the geostatistical method of variogram and kriging analysis using ArcGIS software. The main tool for this analysis is the semi-variogram, which describes the evolution of the semi-variance as a function of the distance between measurements and thus allows the spatial relationship between the data to be studied. The ordinary kriging method or point kriging is more suitable in this work.

In numerical mapping, model uncertainty has three sources of error: (i) measurement error on the variables introduced by operator bias and/or instrument measurement error; (ii) error induced by the structure of the models; (iii) error introduced by a weak relationship between the explanatory variables and the target variable (Cardenas and Malherbe, 2003; Heuvelink, 2013). There are different methods to quantify the importance of each of these errors in the model. The measurement error on the variable is often quantified by the standards governing the measurement methodology. The error induced by the models is quantifiable by model validation. Validation and measurement of final errors at the output of a quantitative model is based on the calculation of several statistical indicators: mean error (ME), mean absolute error (MAE), root mean square error (RMSE), coefficient of determination (R2) and Lin's concordance correlation coefficient (CCC). In the case of validation of qualitative models, for example the soil class at a point, the validation calculation will be based on the proportion of correctly classified points in relation to the calibration dataset used. In this work, the indicators used are the mean error, the root mean square error and the coefficient of determination.

The mean error is an indicator that provides information on possible model bias. It should be close to 0 for good quality models. It is calculated by the following equation:

$$ME=\frac{1}{n}{\sum }_{i=1}^{n}\left({P}_{i}+{O}_{i}\right)\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \left(5\right)$$

With n, the number of samples; Pi, the prediction of the variable and Oi, the sample value.

The root mean square is an indicator estimating the performance of the model and allowing the comparison of two different estimators, in particular the case where there is a bias by one or the other estimator. It is used to measure the accuracy of the model. The more reliable it is, the better the quality of the model. It is calculated according to the following equation:

$$RMSE=\sqrt{\frac{1}{n}{\sum }_{i=1}^{n}\left({P}_{i}+{O}_{i}\right)}\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \left(6\right)$$

With n, the number of samples; Pi, the prediction of the variable and Oi, the sample value.

The R² is an indicator that measures the proportion of variance explained by the model in relation to the total variance of the input data. The closer is to 1, the better the quality of the model. It is calculated according to the following equation:

$${R}^{2}=1-\frac{var res}{var tot}\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots .\dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \dots \left(7\right)$$

With Var res, the variance of the output residuals of the predictive model and Var tot, the variance of the sample series to be predicted.

The evaluation of the prediction uncertainties (errors) of the applied model (ordinary Kriging), allows to check the robustness of the results. This model is applied to 90 soil samples taken at random from a depth of 0 to 30 cm.

3.1 Summary statistics

Table 1 presents the statistics of three (03) soil variables obtained after data processing. Overall, all variables have a positive skewness ranging from 0.26 to 7.13, which means that the mean is higher than the median and the mode. When a variable has a skewness lower than − 1 or higher than 1, it means that the right tail of the distribution is longer than the left for positive skewness and the left tail is longer for negative skewness (Nguemezi et al., 2020 ). Kurtosis is also highly variable, with bulk density values well above 1; this deviation of skewness and kurtosis from zero means that most of these variables have a slightly anomalous distribution. The average organic carbon stock in the Foumban soils is estimated at about 88.12 t/ha. This value reflects a high organic carbon stock. The average bulk density value is 0.98 g/cm³, with values ranging from 0.79 g/cm³ to 2.46 g/cm³. This average value obtained reflects that. All the soils in Foumban have relatively low bulk density. According to Rodiguez Martin et al. (2016), soils rich in organic matter generally have a low bulk density.

Table 1

Summary statistics of variables
Statistique	SOC Stock	SOC	Da (g/cm³)
Number of values	90	90	90
Sum	7931,14	2698,80	88,91
Minimum	9,91	3,64	0,795
Maximum	201,47	64,30	2,46
Range	191,56	60,67	1,665
Mean	88,12	29,98	0,98
Standard deviation	37,08	12,57	0,17
Skew	0,294	0,26	7,13
Kurtosis	0,224	0,14	61,29

Table 2 below presents the results of the applied model (ordinary Kriging) on the variables: SOC (Soil Organic Carbon), bulk density (Da) and SOC Stock. It shows that the R² is very high overall (≥ 0.8). These high values indicate a good quality of the model. The values of the mean error (ME) are very low (ME ≤ 0.1), i.e. close to 0, which attest to a good quality of the model. The RMSE also has overall very low values, which also attests to a good quality of the model. These values obtained mean that ordinary Kriging shows better performance for the prediction of these parameters in the Foumban soils. This is in line with the work of Gomes et al. (2019), which shows that a prediction model performs better when the R² is high, the RMSE and ME are low.

Table 2

Prediction performance indicators for the different parameters of ordinary kriging
	R²	ME	RMSE
SOC Stock	0,91	0,01	5,33
SOC	0,9	-0,05	1,36
Da	0,8	0,01	0,25

3.2. Geostatistical analysis and spatial distribution of soil parameters in Foumban

Figure 3 shows the experimental semi-variogram of the different soil parameters and organic carbon stock. The spherical and Gaussian models were used to explain the semi-variogram. The spatial correlation range is wider in organic carbon stock (191.56) compared to the ranges of organic carbon and bulk density (66.67 and 1.66 respectively). The normalized semi-variograms of SOC, SOC stock and BD values show steps of the order of 170, 1380, 0.006 respectively. This shows the existence of finite variance and the dimensions of the study area sufficient to describe any spatial variability of these parameters (Arrouays et al., 1997). Except for bulk density which has a nugget effect of 0 and a range of 0.02, the other parameters (SOC and SOC stock) have respectively nugget effects of 51 and 450 with ranges of 0.035. This reflects a relatively high variability of these parameters at a scale smaller than the sampling scale.

As Delhomme in 1978, the spatial structures of these parameters were obtained by the ordinary kriging method. The Figs. 4a, 4b and 5 obtained made it possible to distinguish the isovalue zones of the various parameters studied, which could be considered in land management, atmospheric CO₂ reduction processes and the modulation of doses of certain fertility inputs, in order to comply with regional prescriptions and thresholds not to be exceeded (Juste, 1992b).

Figure 4a shows that low levels of soil organic carbon are very poorly represented in the study area. They appear as small pockets in the center and southwest and cover about 6.4% of the study area. The medium SOC content forms a broad band from the north to the center and covers about 28.8% of the study area. High to very high SOC levels are the most represented, covering about 34.6% and 30.2% of the study area respectively. The areas with very high SOC levels correspond to areas with high forest cover, as observed in Fig. 2a of the land use map. These results corroborate with the work of Wiesmeier et al., 2012; Albaladejo et al., 2013 which show that forest soils have higher SOC contents than other soils. The different amount of carbon deposition along the depths was largely influenced by vegetation (Chaikaev and Chavanitch, 2017). Areas of low content are indicative of their extensive use (Rodriguez Martin et al., 2016). It is important to note that the land use map shows that the soils with medium to low levels of SOC observed in Fig. 4a are occupied by housing.

Figure 4b shows that low bulk density soils are dominant and occupy almost the entire study area. They occupy about 72.3% of the total area. Soils with medium bulk density occupy about 24.1% of the study area and form pockets in the centre, north and east. Soils with a high bulk density are very rare. They occupy only 3.2% of the area and are located in the west. Low bulk density soils facilitate the circulation of water and air in the soil (Rodriguez Martin et al., 2016).

3.3. Stock and quantity of CO accumulated in Foumban soils

Carbon stock is the amount of carbon stored in a soil per unit area expressed in tonnes per hectare (t/ha). Soil organic carbon stocks result from the balance between the input and output of carbon in soils (Gomes et al., 2019). Several factors influence this balance depending on environmental conditions (Davidson and Jansens, 2006), local characteristics such as vegetation (Jobbagy and Jackson, 2000) and topography (Tang et al., 2017). The amount of carbon sequestered is the carbon stock contained in a given volume of soil. It is expressed in gigatons (GT). SOC stock and the amount of carbon stored have been assessed by soil group because they are associated with texture and structure, directly affecting SOC accumulation and stabilization processes (Harrison-Kirk et al., 2013; Lenka et al., 2013).

Table 3 presents the stocks and quantities of organic carbon in the different soil groups in the study area.

Table 3

Stock and quantity of CO in Foumban soils
Soil group	Surface area (ha)	SOC Stock (t/ha)					Quantity of C in GT
Soil group	Surface area (ha)	n	Max	Mean	min	sd	Quantity of C in GT
Nipha	13806	13	147.53	71.71	9.91	38.27	9.90 ×10^− 4
Umca	18023	20	162.43	95.08	41.94	38.07	17.13×10^− 4
Flua	8803	10	122.03	80.44	19.94	34.34	7.08 ×10^− 4
Nimfh	6053	4	96.80	91.50	75.30	8.90	5.53 ×10^− 4
FLchp B	14029	17	165.30	98.71	12.87	41.07	13.84 ×10^− 4
Pluph	7242	14	123.33	86.94	13.88	28.02	6.29 ×10^− 4
CHnp	3560	2	156.38	112.62	149.19	43.76	4 ×10^− 4
Ump	3964	4	201.47	116.52	70.38	73.67	4.61 ×10^− 4
FLchp T	1810	3	93.93	71.84	61.67	15.07	1.30 ×10^− 4
FGcp	2037	3	80.72	64.10	49.86	15.57	1.30 ×10^− 4
Totaux	79327	90	201.47	88.95	9.91	18.61	85.79×10^− 4

*n (number of samples); Plaggic hortic NITISOLS (arenic) = Nipha, Chernic arenic UMBRISOLS = Umca, Umbric arenic FERRALSOLS = Flua, Mollic fragic NITISOL (hortic) = Nimfh, Cambic hortic FERRALSOLS (plaggic) on Basalts = FLchp B, Umbric pisoplinthic PLINTHOSOLS (haplic) = Pluph, Nitic CHERNOSOLS (pretic) = CHnp, Cambic hortic FERRALSOLS (plaggic) on Trachytes = FLchp T, pretic UMBRISOLS = Ump, cambic FRAGISOLS (plaggic) = FGcp.

Table 3 above shows that the carbon stock in the soils of the study area varies from 64.10 t/ha in FRAGISOLS (plaggic) cambic (FGcp) to 116.52 t/ha in UMBRISOLS (Ump) pretic, with an overall average of 88.95 ± 18.61 t/ha. The total amount of organic carbon stored in the soils of the study area is estimated to be around 85.79×10 − 4 GT. It varies from 1.30×10^− 4 GT in cambic hortic FERRALSOLS (plaggic) (FLchp) on trachytes to 17.13×10^− 4 GT in chernic arenic UMBRISOLS (UMca). The amount of organic carbon sequestered depends mainly on the area per hectare and the density of the vegetation cover. The larger the area, the more organic carbon is sequestered in the soil.

3.4. Spatial distribution of organic carbon stock in Foumban soils

Four soil classes were defined in the study area according to their organic carbon stock. These are the class of soils with low organic carbon stock (SOCS < 30 t/ha), soils with medium organic carbon stock (30 < SOCS(t/ha) < 60), soils with high organic carbon stock (60 < SOCS(t/ha) < 90) and soils with very high organic carbon stock (SOCS > 90 t/ha) (Fig. 5).

Figure 5 shows that soils with a low organic carbon stock are very poorly represented and occupy only 8.3% of the study area. They form small pockets in the center and south of the study area. Soils with medium organic carbon stocks form large bands in the center and north of the study area. They occupy about 23.2% of the study area. The high organic carbon stocks are more observed in the center and west of the study area. They occupy about 36.7% of the total area of the study area. Very high organic carbon stocks are observed in the west, east and north. They form small pockets in the center and a broad band running from the south to the center of the study area. They occupy about 31.8% of the total area of the study area. In the west of the study area, the very high organic carbon stocks are explained by the presence of the Nkoghan Massif which creates higher altitudes (1800m) favouring the development of grasslands and consequently high percentages of organic matter. This is in accordance with the work of Gomes et al. (2019), who showed that there is a clear trend towards higher values in the milder climates of the mountainous regions, where salient, moderate and humic A horizons coexist, thus favouring the accumulation of organic carbon per unit area.

3.5. Distribution of NDVI (Normalized Difference Vegetation Index) in the study area

The NDVI rendering of the study area based on Landsat 8 images from the ArCgis software, as observed in Fig. 6, shows that the extreme values of the vegetation index range from − 1 to + 1. The total percentage of chlorophyll deficient areas (index < 0), calculated on the basis of the corresponding pixels is 24.82% of the total image area, while the chlorophyll surplus areas (index > 0) represent 75.18%.

Figure 6 shows 5 NDVI classes, namely:

Class 1, with its red colouring, represents the very sparse vegetation and is more present in the center of the study area, more precisely in Foumban, thus representing the urban center. This class is typical of habitats and waterways. It occupies 8.06% of the study area.
Class 2, with its dark yellow colour, represents rare vegetation. It is found directly around the red areas. It is distributed over most of the study area in the form of patches, and is found more in the center and south. This class is typical of bare soil. It covers 16.76% of the study area.
Class 3, which is lime green in colour and represents weak vegetation, is found directly around the dark yellow areas. It is represented throughout the image area in the form of small patches. These are intermediate areas between chlorophyll deficient and chlorophyll excess areas. It occupies 10.63% of the study area.
Class 4, which is light green in colour and represents medium-sized vegetation, is very abundant in the study area and forms interconnected networks. It is most important in the north and east of the study area. These are areas with a surplus of chlorophyll. This class is typical of soils covered with savannah or rocky steppe. It occupies 40.31% of the study area.
Class 5, which is dark green in colour and represents the strong vegetation, is observed directly around the dark green areas. It is more concentrated in the north and east of the study area. In the west it dominates the top of the Nkoghan massif. The area covered by this class increases with the altitude gradient. This class is typical of forest and agroforest soils. It occupies 24.24% of the study area.

Analysis of the maps of organic carbon stock distribution in the soils of the study area, NDVI distribution and land use, shows that the summits of Mount Nkoghan are characterized by soils rich in organic carbon. This corresponds to an area of vegetation cover or forest as shown on the NDVI map. The richness of these soils in organic carbon is due to the decrease in the rate of mineralization of soil organic matter due to low temperatures, high altitude and unfavourable humidity for the activity of microorganisms (Mainlay et al., 2007; Tsozué et al., 2019). According to Schlesinger and Palmer (2000), Amundson (2001), soil organic carbon or organic matter stocks result from a balance between inputs (mostly plant residues) and losses (mostly through microbial decomposition and transfer by erosion). According to the work of Kooke et al. (2019), the carbon stock also varies according to soil type and climatic conditions. Montagnini and Nair (2004) state that the amount of carbon sequestered is a function of tree species, geographical regions (climate, soil), planting densities and system management. High values of organic carbon stock in the study area are recorded in pretic UMBRISOLS (116.52 t/ha), nitic CHERNOSOLS (pretic) (112.00 t/ha), Cambic hortic FERRALSOLS (plaggic) on Basalts (98.71 t/ha), Mollic fragic NITISOL (hortic) (91, 50 t/ha), chernic arenic UMBRISOLS (95,08 t/ha), because of their diagnostic properties in horizons rich in organic matter (organic matter content higher than 2.5% and horizon thickness higher than 15 cm) according to WRB (FAO-ISRIC, update 2022). Lower values are recorded in soils with ferralitic characteristics such as Cambic hortic FERRALSOLS (plaggic) on Trachytes (71.84 t/ha) and Cambic FRAGISOLS (plaggic) (64.10 t/ha). These results are in accordance with the works of Boulmane et al. (2013), Diatta et al. (2016), Bello et al. (2017) and Kooke et al. (2019); which show that soils with a ferralitic character store less organic carbon due to their low organic matter and/or high bulk density (Rodiguez Martin et al., 2016). The carbon stock is therefore higher in the above-ground biomass than in the root biomass of these soil groups. The soils in the study area have overall higher organic carbon stock values than those estimated by the IPCC (2003), which is 31 t/ha for dry tropical areas, and those obtained by Palm et al. (2000) in a cocoa-based agroforestry system, which is 42 t/ha. These values are rather close to those of Albrecht and Kandji (2003) which are 95 t/ha for the average value and 228 t/ha for the maximum value in Acacia auriculiformus agroforests. This means that the predominant agroforest in the Foumban area and its surroundings is essentially made up of trees and shrubs with the characteristics of the acacia. The variation in carbon stock in the study area would be due to the variation in the density of the vegetation cover because the quantity of carbon sequestered by the agroforestry system depends on the species planted, their density, and the structure and function of the latter. In an agricultural system, the organic carbon stock would be lower than in an agroforestry system. This makes it possible to affirm that the agroforestry system is a carbon sink compared to an agricultural system. Nevertheless, agroforestry, for social and cultural reasons, such as land management, seems difficult to promote. It will therefore be a less important contributor to carbon sequestration in the study area than expected. Anthropogenic activity marked by the addition of organic amendments (manure, compost, plant debris, etc.) would contribute to the increase of organic carbon stocks in the soil. There are different soil management practices to increase the organic matter content of the soil such as: increasing productivity and biomass (varieties, fertilisation and irrigation).

The present study focused on the evaluation of the carbon stock in the soils of Foumban in West Cameroon. To this end, ninety (90) soil samples were collected in the field and taken to the laboratory where organic carbon content and bulk density analyses were carried out. Organic carbon stocks and stored carbon quantities were calculated for each reference soil group identified by the study area. The results obtained were processed by statistical and geostatic methods. As a result, R² values higher than 0.8 and RMSE values close to 0, attest to a better performance of ordinary kriging for the prediction of soil parameters studied in the Foumban area. The spherical and Gaussian models were used to explain the experimental semi-variogram of the different soil parameters (bulk density and SOC) and their spatial structures were obtained by the ordinary kriging method. Calculations of SOC stock and carbon quantity showed that the carbon stock in the soils of the study area varies from 64.10 t/ha in FRAGISOLS (plaggic) cambic (FGcp) to 116.52 t/ha in UMBRISOLS (Ump) pretic, with an overall average of 88.95 ± 18.61 t/ha. The total amount of organic carbon stored in the soils of the study area is estimated to be around 85.79×10^− 4 GT. It varies from 1.30×10^− 4 GT in cambic hortic FERRALSOLS (plaggic) (FLchp) on trachytes to 17.13×10^− 4 GT in chernic arenic UMBRISOLS (UMca). The NDVI rendering of the study area shows that the extreme values of the vegetation index range from − 1 to + 1. The total percentage of chlorophyll deficient areas (index < 0), calculated on the basis of the corresponding pixels is 24.82% of the total image area, while the chlorophyll surplus areas (index > 0) represent 75.18%. The variation of the carbon stock in the study area would be due to the variation of the density of the vegetation cover. Anthropogenic activity marked by the addition of organic amendments (manure, compost, plant debris, etc.) would contribute to the increase in organic carbon stocks in the soils of the study area.

Albaladejo J., Ortiz, R., Garcia-Franco N., Navarro A., Almagro M., Pintado, J., Martínez-Mena M. (2013). Land use and climate change impacts on soil organic carbon stocks in semi-arid Spain. J. Soils Sediments 13, 265–277.
Albrecht A., et Kandji S.T. (2003) Carbon sequestration in tropical agroforestry systems. Agriculture, Ecosystems and Environment, 99: 15–27.
Amundson R. (2001). The soil carbon cycle. Annual Reviews of Earth and Planetary Sciences, 29: 535–562.
Arrouays D., Isabelle Vion Jolivet C.I., Dominique Guyon, Couturier A., Wilbert J. (1997). Variabilité interparcellaire de quelques propriétés des sols sableux des Landes de Gascogne (France). Etude et Gestion des sols, 4, 1, 1997, pages 05–16,
Belkacem S., Nys C., Dupouey J.L. (1998). Evaluation des stocks de carbone dans les sols forestiers: importance de la sylviculture et du milieu sur la variabilité, INRA/DPE, Agriges, 68P.
Bello O.D., Saidou A., Ahoton E.L., Ezin A.V., Avaligbé J.F.Y., Akponikpè P.B.I., and Aho N. (2017) Assessment of organic carbon stock in cashew plantations (Anacardium occidentale L.) in Benin (West Africa). International Journal of Agriculture and Environmental Research, 3(4): 3471–3495.
Boulmane M., Regina I.S., Khia A., Oubrahim H. (2013) Estimation du stock de carbone dans les ilicaies du Haut Atlas Marocain. Ann. Rech. For. Maroc, 20(2): 275–286.
Bui E., Henderson B., Viergever K. (2009). Using knowledge discovery with data mining from the Australian Soil Resource Information System database to inform soil carbon mapping in Australia. Glob. Biogeochem. Cycles 23 (4) (n/a-n/a).
Cardenas G., Malherbe L. (2003). Evaluation des incertitudes associées aux méthodes géostatistique. s.n.2(1).
Chaikaew Pasicha et Chavanitch Suchana, (2017). Variabilité spatiale et relation entre la matière organique du sol de la mangrove et le carbone organique. Applied and Environnemental soil sciences. Vol 2017. Article ID 4010381, 9 pages.
Cerri C.E.P., Easter M., Paustian K., Killian K., Coleman K., Bernoux M., Falloon P., Powlson D.S., Batjes N., Milne E., Cerri C.C. (2007). Predicted soil organic stocks and changes in the Brazilian Amazon between 2000 and 2030, Agr. Ecosyst. Environ. 15, 112–127.
Davidson E.A., Janssens I.A. (2006). Temperature sensitivity of soil carbon decomposition and feedbacks to climate change.
Delhomme P. (1978). Kriging in hydrosciences. Adv. Water Res., 1: 251–266.
Diatta A.A., Ndour N., Manga A., Sambou B., Faye C.S., Diatta L., Goudiaby A., Mbow C., Dieng S.D. (2016) Services écosystemiques du parc agroforestier à Cordyla pinnata (Lepr. Ex A. Rich.) Arachidier (Sénégal). Int. J. Biol. Chem. Sci., 10(6): 2511–2525.
Durot C. (2013) Evaluation et comparaison des stocks de carbone des systèmes agroforestiers à base de cacaoyers du centre Cameroun: Cas de l’arrondissement de Bokito. Mémoire d’ingénieur agricole, Institut supérieur d’Agriculture de Lille. 83p.
Dussault Frenette C. (2008) L’agroforesterie comme outil de développement durable dans les pays en voie de developpement. Domaine de Savoir UdeS, Dépôt Institutionnel.101. http//hdl.handle.net/1114.
Evans J., Fernandez I.J., Rustad L.E., Norton S.A. (2001). Methods for evaluating fractions in forest soil, Techn. Bull., 178, 1–44.
GIEC (2013). Résumé à l’intention des décideurs, Changements Climatiques 2013: Les éléments scientifiques. Contribution du Groupe de travail I au cinquième Rapport d’évaluation du Groupe d’experts intergouvernemental sur l’évolution du climat. Cambridge, Royaume-Uni et New York : Cambridge University Press.
Gomes C.L., Faria R.M., Eliana de Souza, Veloso G.V., Schaefer Carlos E.G.R., Filho E.I.F. (2019). Modelling and mapping soil organic carbon stocks in Brazil. Geoderma 340 (2019) 337–350.
Harrison-Kirk T., Beare M.H., Meenken E.D., Condron L.M. (2013). Soil organic matter and texture affect responses to dry/wet cycles: effects on carbon dioxide and nitrous oxide emissions. Soil Biol. Biochem. 57, 43–55.
Heuvelink G.B. (2013). Uncertainty quantification of GlobalSoilMap products. Glob. Basis Glob. Spat. Soil Inf. Syst. 327–332.
Houghton R.A., (2007). Balancing the global carbon budget. Annu. Rev. Earth Planet. Sci. 35, 313–347.
IPCC (Intergovernmental Panel on Climate Change) (2003). Good Pratice Guidance for Land Use, Land-Use Change and Forestry (LULUCF). Institute for Global Environmental Strategies, Hayama Japan.
Jobbagy E.G., Jackson R.B. (2000). The vertical distribution of soil organic carbon and its relation to climate and vegetation. Ecol. Appl. 10, 423–436.
Juste C. (1992b). Acidité et chaulage des sols sableux. In: Etude et mise en valeur des sols sableux de Pologne et de France. INRA Editions, pp. 129–134.
Kambire H.W., Djenontin I.N.S., Kabore A., Djoudi H., Balinga M.P., Zida M., Assembe-Mvondo S. (2015). La REDD + et l’adaptation aux changements climatiques au Burkina Faso: Causes, agents et institutions (Vol. 123). CIFOR.
Kooke G.X., Mandus Ali R.K.F, Djossou J.M., Toko Imorou I. (2019). Estimation du stock de carbone organique dans les plantations de Acacia auriculiformus A. Cunn. ex Benth. des forests classées de Pahou et de Ouèdo au Sud du Bénin. Int. J. Biol. Chem. Sci. 13(1): 277–293.
Lenka N.K., Lal R. (2013). Soil aggregation and greenhouse gas flux after 15 years of wheat straw and fertilizer management in a no-till system. Soil Tillage Res. 126, 78–89.
Lotse T.V.P., Mbog M.B., Ngon Ngon G.F., Edzoa R.C., Tassongwa B., Bitom D., Etame J., (2022). Physico-chemical properties and fertility assessment of soils in Foumban (West-Cameroon). Applied and Environnemental soil sciences. Vol 2022. Article ID 7889430, 14 pages.
Lozano-Garcia B., and Parras-Alcantara (2013). Short-term effects of olive mill by-products on soil organic carbon, total N, C:N ratio and stratification ratios in a Mediterranean olive grove. Agriculture, ecosystems and environment 165, 68–73.
Mainlay R., Feller C., Swift M.J. (2007). Historical evolution of soil organic matter concepts and their relationships with the fertility and sustainability of cropping systems. Agric. Ecosyst. Environ. 119, 217–233.
Montagnini F., et Nair P. (2004). Carbon sequestration: An underexploited environmental benefit of agroforestry systems. Agroforestry Systems, (61): 281–295.
Muoghalu I.J. (2014). Vulnérabilité des systèmes biophysiques et socioéconomiques des savanes et formations boisees d’Afrique occidentale et centrale au changement climatique. African Forest Forum, 2(14): 35p. https//www.tandfoline.com/doi/full/10.
Ngueguim J.R., Zapfack L., Noumi V.N., Onana D., Betti J.L., Riera B. (2015). Expériences sylvicoles au Cameroun: Croissance, mortalité et adaptabilité des espèces de bois d’œuvre dans la station forestière de Mangombé (1964–2010). Int. J. Biol. Chem. Sci., 9(6) : 2789–2807.
Nguemezi C., Tematio P., Yemefack M., Tsozue D., Silatsa T.B.F. (2020). Soil quality and soil fertility statuts in major soil groups at the Tombel area, South-West Cameroon. Heliyon 6. DOI
Parras-Alcantara L, Diaz-Jaimes L, Lozano-Garcia B, Rebollo P,F, (2014). Organic farming has little effect on carbon stock in a Mediterranean dehesa (Southern Spain). Catena 113, 9–17.
Peng C., Liu J., Apps M., Dang Q., Kurz W. (2000). Quantifiying Ontario’s forest carbon budget. 1 Carbon stock and flux of forest écosystems in 1990, For. Res. Rep., 158, 123–137.
Rodiguez Martin J.A., Alvaro-Fuentes J., Gonzalo J. Gil C., Ramos-Miras J.J., Grau Corbi J.M., Boluda R., (2016). Assessment of the soil organic carbon stock in Spain. Geoderma 264 (2016) 117–125.
Ruiz Sinoga J.D., Pariente S., Diaz A.R., Martinez Murillo J.F., (2012). Variability of relationships between soil organic carbon and some soil properties in Mediterranean rangelands under different climatic conditions (south of Spain). Catena 94, 17–25.
Tsozué D., Nghonda J., Tematio P., Djakba Basga S. (2019). Changes in soil properties and soil organic carbon stocks along an elevation gradient at Mount Bambouto, Central Africa. Catena 175, 251–262.
Saïdou A., Dossa A.F.E., Gnanglè P.C., Balogoum I., Aho N. (2012). Evaluation du stock de carbone dans les systemes agroforestiers à karité (Vitellaria paradoxa C.F. Gaertn) et à néré (Parkia biglobosa Jacq. G. Don) en zone Soudanienne du Bénin. Bulletin de la Recherche Agronomique du Bénin (BRAB). 9p. http://www.slire.net/document/1737
Schlesinger P. et Palmer Winkler J. (2000). «Soil and the global carbon cycle.» In Wigley T.M.L. et Schimel D.S. (éd.): The carbon cycle, Cambridge University Press, Cambridge.
Tang X., Xia M., Pérez-Cruzado C., Guan F., Fan S. (2017). Spatial distribution of soil organic carbon stock in Moso bamboo forests in subtropical China. Sci. Rep. 7, 42640.
Vroh B.T.A., Tiebre M.S., N’guessan K.E. (2014). Diversité végétale urbaine et estimation du stock de carbone: cas de la commune du Plateau Abidjan, Cote d’ivoire. Afrique Science, 10(3): 329–340.
Yang R.M., et al. (2016a). Comparison of boosted regression tree and random forest models for mapping topsoil organic carbon concentration in an alpine ecosystem. Ecol. Indic. 60, 870–878.
Walkley A and Black IA (1994). An examination of the Degtjareff method for determining soil organic matter and a proposed modification of the chromic acid titration method. SoilSci. 37, pp29-37.
Wiesmeier M., Spörlein P., Geuß U., Hangen E., Haug S., Reischl A., Schilling B., Von Lützow M., Kögel-Knabner I. (2012). Soil organic carbon stocks in southeast Germany (Bavaria) as affected by land use, soil type and sampling depth. Glob. Chang. Biol. 18, 2233–2245.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Assessment of the Soil Organic Carbon Stock (SOCS) of Foumban (West Cameroon)

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

2 .1. Study area, land use and sampling points

Results and discussion

3.1 Summary statistics

3.3. Stock and quantity of CO accumulated in Foumban soils

3.4. Spatial distribution of organic carbon stock in Foumban soils

Conclusion

References

Additional Declarations

Status:

Version 1