Geospatial Assessment of Soil Organic Matter Variability at Sidi Bennour District in Doukkala Plain in Morocco
More details
Hide details
1
Department of Geology, Faculty of Sciences, Chouaib Doukkali University, BP.20, 24000, El Jadida, Morocco
Publication date: 2021-12-02
Corresponding author
Abdelkrim Bouasria
Department of Geology, Faculty of Sciences, Chouaib Doukkali University, BP.20, 24000, El Jadida, Morocco
J. Ecol. Eng. 2021; 22(11):120-130
KEYWORDS
TOPICS
ABSTRACT
Understanding the spatial variability of soil organic matter (SOM) is critical for studying and assessing soil fertility and quality. This knowledge is important for soil management, which results in high crop yields at a reduced cost of crop production and helps to protect the environment. The benefits of an accurate interpolation of SOM spatial distribution are well known at the agricultural, economic, and ecological levels. It has been conducted this study for comparing and analyze different spatial interpolation methods to evaluate the spatial distribution of SOM in Sidi Bennour District, which is a semi-arid area in the irrigated scheme of the Doukkala Plain, Morocco. For conding this study, it was collected 368 representative soil samples at a depth of 0-30 cm. A portable global positioning system was used to obtain the location coordinates of soil sampling sites. The SOM spatial distribution was performed using four interpolation methods: inverse distance weighted and local polynomial interpolation as deterministic methods, and ordinary kriging and empirical Bayesian kriging as geostatistical methods. High SOM levels were concentrated in vertisols, and low SOM levels were observed in immature soils. The average SOM value was 1.346%, with moderate to high variability (CV = 35.720%). A low SOM concentration indicates a continuous possibility of soil degradation in the future. Ordinary kriging yielded better results than the other interpolation methods (RMSE = 0.395) with a semivariogram fitted by an exponential model, followed by inverse distance weighted (RMSE = 0.397), empirical Bayesian kriging (RMSE = 0.400), and local polynomial interpolation (RMSE = 0.412). Soil genetics and the strong influence of human activity are the major sources of SOM spatial dependence, which is moderate to low. Low SOM content levels (< 1.15%) were present in the southwestern and eastern parts of the digital map. This situation calls for the implementation of specific soil recovery measures.