| Summary: | Rangeland degradation represents a significant environmental and socio-economic challenge in dryland ecosystems, particularly within Sub-Saharan Africa, where pastoral livelihoods depend on fragile landscapes.The proliferation of Prosopis juliflora, an aggressive woody invasive species, has emerged as a critical ecological and management concern. Accurate monitoring of Prosopis juliflora invasion is imperative for effective management. However, traditional field-based surveys, while valuable, are hindered by high costs, logistical challenges, and limited spatial coverage, rendering them inadequate for tracking invasions across extensive rangelands. Recent advancements in remote sensing and machine learning present a transformative alternative.The study employs Random Forest classification across multiple satellite datasets to assess sensor performance in detecting Prosopis juliflora along with community validation using PGIS. It further utilizes ground-truth observations to predict the spread of this invasive species across the region. This integrative approach bolsters the evidence base for decision-making, thereby supporting adaptive, inclusive, and sustainable management of rangelands.
|
|---|