A global assessment of forest surface albedo and its relationships with climate and atmospheric nitrogen deposition.

We present a global assessment of the relationships between the short-wave surface albedo of forests, derived from the MODIS satellite instrument product at 0.5° spatial resolution, with simulated atmospheric nitrogen deposition rates (Ndep ), and climatic variables (mean annual temperature Tm and total annual precipitation P), compiled at the same spatial resolution. The analysis was performed on the following five forest plant functional types (PFTs): evergreen needle-leaf forests (ENF); evergreen broad-leaf forests (EBF); deciduous needle-leaf forests (DNF); deciduous broad-leaf forests (DBF); and mixed-forests (MF). Generalized additive models (GAMs) were applied in the exploratory analysis to assess the functional nature of short-wave surface albedo relations to environmental variables. The analysis showed evident correlations of albedo with environmental predictors when data were pooled across PFTs: Tm and Ndep displayed a positive relationship with forest albedo, while a negative relationship was detected with P. These correlations are primarily due to surface albedo differences between conifer and broad-leaf species, and different species geographical distributions. However, the analysis performed within individual PFTs, strengthened by attempts to select 'pure' pixels in terms of species composition, showed significant correlations with annual precipitation and nitrogen deposition, pointing toward the potential effect of environmental variables on forest surface albedo at the ecosystem level. Overall, our global assessment emphasizes the importance of elucidating the ecological mechanisms that link environmental conditions and forest canopy properties for an improved parameterization of surface albedo in climate models.