Global variability of carbon use efficiency in terrestrial ecosystems

Vegetation carbon use efficiency (CUE) is a key measure of carbon transfer from the atmosphere to terrestrial biomass. Global temporal and spatial variability of CUE has been largely ignored. Here, we show that failure to account for such variability can induce up to 40% bias of anthropogenic CO2 emissions. We delineated the temporal and spatial variability of CUE and its driving factors based on a global database of site-year CUE that we developed from 188 field observational sites in five types of ecosystems (forest, grassland, wetland, crop and tundra). Globally, we found two prominent CUE gradients in ecosystem types and latitude, that is, CUE varied with ecosystem types, with the highest in wetlands and the lowest in grassland, and CUE decreased with latitude with the lowest CUE in tropics, and the highest CUE in higher latitude regions. Based on our global map of CUE, we find that global MODIS NPP alone was underestimated by 4.7 Pg C, which means that global terrestrial ecosystems as a carbon sink can actually fix up to 9% more carbon than previous estimation.