Remote sensing-based spatio-temporal modeling to predict biomass in Sahelian grazing ecosystem

Abstract

The Sahelian land is used for both grazing and agriculture in the vicinity of rare watering points. This situation leads to a fundamental issue of how to manage scarce natural resources influenced by harsh climatic conditions and anthropogenic factors. The challenge of studying the spatio-temporal variability of primary production and developing appropriate management methods for this fragile environment is to find simple, reliable, and applicable approaches. Most of predicting models of arid herbaceous biomass can only simulate a limited number of species. The main objective of this research is to implement an approach that would help the management of the Sahelian pastoral environment. This approach would also minimize data acquisition costs and take into account grazing capacity in assessing the state of the environment. After defining the parameters relevant to the productivity of Sahelian grazing ecosystems, we have analyzed and adapted Wu's physical model to this environment to simulate its productivity and evaluate its variations. This adaptation took into account remote sensing and socioeconomic data. With an accuracy of at least 80% in simulating biomass over a decade, the reference and adapted models are of interest in planning and managing the studied environment.