Remote Sensing-Based Assessment of Dry-Season Forage Quality for Improved Rangeland Management in Sahelian Ecosystems

Abstract

Residents of the Sahel depend on livestock, but harsh environmental conditions during the dry season limit rangeland forage, which is the main source of livestock feed. Al-though operational tools exist for assessing and monitoring forage quantity during the dry season, assessments of forage quality are lacking. We addressed this gap by developing satellite-based monitoring of forage quality across Sahelian rangelands during the dry season. Acid detergent fiber (ADF), neutral detergent fiber (NDF), and crude protein (CP) content (%) were measured in forage samples collected from 11 sites across the Senegalese rangelands in 2021. Multilinear (MML) regression and support vector machine (SVM) models were calibrated with spectral indices to estimate these parameters of forage quality. The vegetation variables assessed were herbaceous mass (HQ), woody foliage mass (LQ), and total fo-rage mass (HLQ). The MML regression provided the most accurate estimates for CP (HQ: R² = 0.81, LQ: R² = 0.72, and HLQ: R² = 0.70), ADF (HQ: R² = 0.70, LQ: R² = 0.77, and HLQ: R² = 0.61), and NDF (HQ: R² = 0.47, LQ: R² = 0.83, and HLQ: R² = 0.60). Temporal analysis revealed a slight decrease in CP and an increase in fiber during the dry season. Spatial analysis indicated that CP was higher in the steppe zone than in the savanna zone, and a decrease correlated with the rainfall gradient. The HQ alone was insufficient to meet livestock needs during the dry season, highlighting the importance of woody plants as an additional forage source. These findings will improve feed balance calculations in Sahelian countries, enable more sustainable use of rangelands, and contribute to the resilience of Sahelian communities to climate change.