Integrating Remote Sensing and Positive Deviance to Identify Scalable Forest Conservation Practices in Northern Ghana

While the role of trees in supporting climate resilience is well established, less is known about why some rural communities manage to maintain forest cover despite facing similar environmental and socio-economic constraints. This study addresses that gap by applying the Data-Powered Positive Deviance (DPPD) framework to identify and examine locally embedded forest conservation practices in Northern Ghana. Using satellite-derived vegetation indices, we identified communities that performed better or worse than expected in terms of forest-cover change and paired them for comparative analysis, complemented by qualitative fieldwork exploring the social, institutional, and behavioral factors shaping these outcomes. The results point to a consistent set of practices associated with higher forest retention: the economic use of tree products, the integration of trees into land boundary demarcation, and the presence of functioning community-level forest governance structures. Together, these elements, referred to as a "golden mix," illustrate how everyday livelihood strategies and local institutions can reinforce forest conservation. By combining geospatial analysis with community-level insights, the study demonstrates a practical approach for identifying contextually grounded conservation strategies. Although the findings are specific to Northern Ghana, they provide a basis for exploring similar, locally driven pathways to forest conservation and climate resilience in comparable semi-arid settings.