Extended Data Fig. 2: Alternative measures of edge effects.

a, Map of the Spearman’s rank correlation coefficient (ρ). Spearman correlation tests between log₁₀-transformed distance to the nearest forest edge and forest biomass density were conducted by each geographical grid cell with a scale of 100 km. Grid cells are shaded according to their ρ values. b, Map of \(\frac{\Delta AGB}{\Delta D}\) after filtering out points at the outermost (distance < 30 m) edge, which could be affected by mixed land cover. Spatial log-linear regression models, as described in Methods, are then applied for each grid cell to calculate \(\frac{\Delta AGB}{\Delta D}\). c, Map of \(\frac{\Delta AGB}{\Delta D}\) on log₁₀-transformed biomass density. A spatial log-linear regression model, as described in the Methods section, was conducted per grid cell with log₁₀-transformed biomass density as a dependent variable and log₁₀-transformed distance to the edge as an independent variable. Each geographical grid cell with a scale of 100 km is shaded according to the \(\frac{\Delta AGB}{\Delta D}\).