Combining Remote Sensing and Space-Time Analysis for Desertification Monitoring in the Semiarid Dryland of Nigeria

Abstract

Desertification has been identified as the resultant effect of dryland loss. Desertification is catalysed by anthropogenic modifications and variations in environmental/climatic conditions. The situation in Nigeria is further exacerbated by the growing demand for land by the population. To this effect, this study performed a space-time analysis of vegetative cover between 2001 and 2020 to unravel patterns and trends across the semiarid region of the dryland system in Nigeria. The dynamics during the rainy season (May and September) were examined using the Terra Moderate Resolution Imaging Spectroradiometer (MODIS) dataset subjected to space-time analysis. Generalised Difference Vegetation Index (GDVI) was computed to the power of 2 to quantify vegetative cover across the study area. The results showed that the average of the GDVI ranges between -0.40 and 0.94, with a standard deviation of 0.11. Time series cluster analysis revealed that there are two temporal clusters: (1) no statistically significant trend (Statistics= 1.33, p-value = 0.18) and (2) statistically significant downtrend (Statistics = -2.37, p=0.02), with cluster 1 covering 95% of the areas examined. The most dominant (97% of the area) emerging space-time pattern was cold-spots (persistent, diminishing, sporadic, oscillating, and historical types). In conclusion, most of the areas showed no definite temporal pattern of vegetation pattern during the period, while more than 90% of the areas have witnessed a decline in vegetative cover. There is a need for a more coordinated approach to desertification control, constant monitoring is pertinent while new approaches to restoring degraded land are recommended.