Madagascar's forests are known for their rich biodiversity, and they are home to numerous endemic plant and animal species. However, these ecosystems face significant threats from human activities, leading to deforestation and forest degradation. Effective conservation efforts must focus on both protecting the remaining natural forests and restoring those already destroyed. To address these challenges, the Thünen Institute of Forestry, in collaboration with the German Society for International Cooperation (GIZ), conducted the project "Analysis of forest management and reforestation potential in the regions of Boeny and DIANA in Madagascar" (AFOB). The project analysed deforestation and forest degradation along with their drivers, and evaluated the success of reforestation programs implemented over the past decades. These past programs sought to foster sustainable charcoal production practices, thereby reducing reliance on natural forests. To achieve these objectives, we used a spatial analysis approach that relied on open-access remote sensing data products. For natural forests, we identified when and where potential forest degradation and deforestation occurred and demonstrated forest fragmentation dynamics, differentiating between protected and non-protected areas. Our findings revealed that between 2000-2023, deforestation and forest degradation have increased substantially, with degradation having been a more significant issue in the past than previously understood. Key drivers include the conversion of forests into agricultural land, fire, illegal charcoal production, and the illegal extraction of valuable timber. We showed that over the last two decades, cropland has expanded significantly, with an even faster rate of increase in recent years. Our findings suggest that deforestation and forest degradation should be viewed as distinct but interrelated threats, each driven by different underlying processes that require tailored strategies, respectively. In the reforestation areas, crown cover was initially low during the first rotation but showed an upward trend over the analysed period. However, more than half of the plots did not develop crown cover exceeding 20%, suggesting that they did not reach their full ecological potential. In a subsequent analysis, we found that environmental factors, such as topography and soil, consistently influenced crown cover development across regions and programs, while the influence of socio-economic factors varied by program and region. However, these factors had only a weak overall influence on crown cover. In contrast, our analysis suggested that unaccounted plot-level factors, which we could not include due to data limitations, had a significant impact on crown cover variation. These factors could potentially include initial planting densities, survival rates, presence of remnant vegetation, fire management, and management objectives of the landowner, though their exact nature remains speculative. This indicates that a uniform reforestation approach is ineffective; instead, site-specific adaptive management is crucial for success. In an additional analysis, we identified suitable areas with potentially high growth rates for reforestation, which can guide future restoration initiatives with a stronger focus on site-specific planning. The insights gained from the AFOB project will assist GIZ in planning and implementing targeted reforestation and conservation initiatives, ultimately supporting the preservation of Madagascar's unique ecosystems and maintaining ecological integrity.
