Generating fine resolution leaf area index maps for boreal forests of Finland

In addition to global-scale LAI products, higher resolution maps are needed to better depict the spatial heterogeneity of LAI. In this study, we applied empirical and forest reflectance model based LAI estimation methods to generate fine-resolution LAI maps for boreal forests of Finland. The countrywide satellite image mosaics at 25 m resolution for the years 2000 and 2006 were used as input satellite data. The Reduced Simple Ratio (RSR) vegetation index based approach performed better in estimating effective LAI (L_e) than forest reflectance model based approach. The final L_e estimates were corrected for shoot-level clumping in coniferous and mixed forests according to the land cover maps. The L_e estimates show good agreement with LAI field measurements over the modeling sites although the largest values are slightly underestimated. MODIS LAI V005 estimates are higher than the upscaled 1 km resolution forest LAI estimates for 2006.