Forest cover restoration analysis using remote sensing and machine learning in central Malawi

Abstract

Forests play a crucial role in maintaining ecological balance and supporting human well-being, yet they are increasingly threatened by anthropogenic pressures. Over the past two decades (2000–2020), Malawi has experienced a 21 % decline in forest cover, necessitating urgent and effective restoration strategies. Community-led forest regeneration initiatives have emerged as a viable solution to counteract this degradation. This study employs remote sensing and machine learning techniques to evaluate the effectiveness of such interventions in a village forest area in central Malawi. Utilizing a Support Vector Machine (SVM) classification algorithm applied to time-series Landsat and high-resolution imagery (2003–2023), we quantify land cover changes, while Normalized Difference Vegetation Index (NDVI) trends serve as indicators of ecological recovery. Our results reveal a significant transformation of the landscape, including a 61.2 % reduction in bare land and a 57.8 % decline in grassland, coupled with a remarkable 305.6 % increase in tree cover. NDVI values evolved from indicating degraded surfaces (-0.12 in 2003) to sustained positive indices (0.10–0.42 by 2023), signifying ecosystem revitalization. Nearly 58.4 % of previously bare land transitioned into grassland, while 60.28 % developed into tree cover, underscoring the synergy between natural ecological processes and community-driven conservation strategies, such as reforestation and the cessation of encroachment in protected forest areas. The integration of SVM demonstrated high classification accuracy (>92 %), confirming its reliability in monitoring landscape recovery. These findings emphasize the effectiveness of participatory governance and targeted policy enforcement in fostering forest restoration. To sustain these gains, we advocate for the broader implementation of community-based conservation models, enhanced by Geographic Information Technology (GIT), to harmonize socio-economic development with ecological resilience. The success of this study presents a scalable and replicable framework for forest restoration, highlighting the critical role of collaborative stewardship, advanced monitoring, and adaptive governance in mitigating deforestation-induced environmental challenges.