Assessment of the spatial dynamics of sandy desertification using remote sensing in Nemamcha region (Algeria)

Abstract

Sandy Sandy desertification is considered one of the environmental problems in the pre-desert and steppe ranges of Algeria. In order to reveal the spatial dynamics of sandy desertification, a multivariate linear equation was used to assess the relationship between spectral data of Landsat images and ground-surveyed soil samples. The multivariate models were used to estimate sand, silt, and clay percent, and the USDA textural triangle classification was used to specify soil texture by applying the limits of the textural variables in a GIS environment. The statistical analysis showed a strong correlation between blue, SWIR 1, SWIR 2, sand, silt, and clay contents. The results showed that the sandy loam are the most widely distributed soil textural class 52% and 42% in 2013 and 2019, respectively. Furthermore, spatial change detection results between 2013 and 2019 illustrated that 12% of the entire study area have suffered signs of desertification that have a negative change of soils to sandy and loamy sand. The soil indices indicate a strong relationship with the top soil sand content, and the NDVI values varied between 0.06 and 0.44 in 2013 and −0.01 and 0.3 in 2019, indicating a severe regression in vegetation conditions over time. Thereafter, land cover and land use analysis show the severity of land degradation and the irreversibility of sandy desertification, agriculture land decreased from 31% in 2013 to 4% in 2019, pasture land decreased from 21% to 13%, bare land increased from 22% to 36% and sandy land increased dramatically from 26% to 47%.