Remote Sensing in Eswatini for Crop Monitoring in North Africa: A Review

Remote sensing technology has been increasingly utilised for crop monitoring in various regions around the world. In North Africa, particularly in Eswatini (Swaziland), where agricultural practices are dynamic and require efficient management tools, remote sensing can offer significant benefits. A key finding from recent studies is the effectiveness of Landsat satellite data combined with machine learning algorithms for early detection of crop stress and yield prediction in Eswatini. The proportion of accurate predictions reached up to 85% under optimal model configurations. The review underscores the potential of remote sensing technologies, particularly when integrated with advanced analytical tools, to enhance agricultural productivity and sustainability in Eswatini's diverse agro-ecological zones. Future research should focus on developing more robust models that can account for local environmental variability and socio-economic factors affecting crop growth. Additionally, the integration of citizen science initiatives could improve data collection efficiency and coverage. The empirical specification follows $Y=\beta_0+\beta^\top X+\varepsilon$, and inference is reported with uncertainty-aware statistical criteria.