Dynamiques contrastées de reverdissement et dégradation de la couverture végétale au Sénégal révélées par analyse de série temporelle du NDVI MODIS

Abstract

This study seeks to analyze changes in vegetation cover in the ecoregions of Senegal over the past two decades (2000 to 2019) using a time series of 16-day MODIS NDVI images with a spatial resolution of 250m. The study method is a combination of remote sensing and statistics. The data processing followed three main steps which are: first, the Main Component Analysis (PCA) on the 2000-2019 time series followed by the unsupervised classification using the PCA result; second, the calculation of the average NDVI of the 23 16-day periods; third, the calculation of the Kendall correlation or “tau” (τ) associated with the p 0.1 significance test. The results indicate changes in the phenological rhythm of the classes corresponding to lowland crops, floodable valleys, mangroves, dense forests, and wooded and wooded savannah. Some modifications are similar to degradation, and others to a regreening Similarly, there was a time lag and a later green season. Kendall’s correlation indicated an increase in NDVI between 2000-2009 primarily in the eastern, southwest and estuarine (ZDE) ecoregions with a significant trend. During the 2010-2019 decade, a set of small areas of progress was noted. On the other hand, the negative correlations of NDVI values, which indicate a decrease in vegetative activity, are observed mainly in the northern and central-western ecoregions of the country, although with a non-significant trend.

Keywords

• remote sensing
• Senegal
• degradation
• vegetation cover
• re-greening
• MODIS-NDVI