Forest land is essential for mitigating climate change due to its substantial carbon sink capacity. Various factors, including climate, terrain, population, and road density, significantly influence forest land dynamics. This study examines the impact of 18 biophysical and socio-economic spatial parameters on forest land changes in Dutch municipalities from 2000 to 2022 using ordinary least squares (OLS), geographically weighted regression (GWR), and multiscale geographically weighted regression (MGWR) models. The results show that MGWR, which accommodates variable spatial scales, provided the best fit compared to GWR and OLS. The factors were categorized by effect scale: global (e.g., nighttime light), regional (e.g., road distance, population, address density, temperature, and precipitation), and local (e.g., total road length, slope). This categorization supports scale-appropriate policies and localized planning. The study underscores the effectiveness of MGWR in identifying spatially sensitive parameters, highlighting the importance of hierarchical policy-making and localized planning to enhance climate mitigation through forest management.