Mapping net primary production and related biophysical variables with remote sensing: Application to the BOREAS region

Maps of net and gross primary production, autotrophic respiration, biomass, and other biophysical variables were generated for 10(6) km(2) of boreal forest in central Canada (the Boreal Ecosystem-Atmosphere (BOREAS) region) using a production efficiency model (PEM) driven with remotely sensed observations at 1 km(2) spatial resolution. The PEM was based on carbon yields of absorbed photosynthetically active radiation for both gross and net primary production (GPP and NPP), accounting for environmental stress and autotrophic respiration (R-a). Physiological control was modeled using remotely sensed maps of air temperature, vapor pressure deficit, and soil moisture. The accuracy of the inferred variables was generally within 10-30% of point measurements at the surface and independent model results (both at the stand level). Biomass maps were derived from visible reflectance measurements and were also compared to independently derived maps. Area-averaged GPP was 604 g C m(-2) yr(-1) compared with average canopy respiration of 428 g C m(-2) yr(-1) and NPP of 235 g C m(-2) yr(-1). Net annual carbon uptake in net primary production for the region totaled 175 teragrams. Canopy carbon exchange (GPP and R-a) differed widely between land cover types even though the model does not use land cover information. Extensive areas of the least productive cover types (e.g., lowland needleleaf species) accounted for the greatest amount of NPP.