Increase of surface energy fluxes due to warming climate in Alaska based on upscaling of eddy covariance measurements

In order to examine change of radiative forcing associated with surface energy budget from 2000 to 2011, sensible and latent heat fluxes (H and LE) derived from an eddy covariance network (21sites) at Alaska were upscaled by satellite remote sensing data, gridded climate data and a support vector regression (SVR) model. Regional total of annual H and LE had a weak positive trend of 0.46 W m-2 yr-1 (R2 = 0.21 and p = 0.13) and 0.35 W m-2 yr-1 (R2 = 0.11 and p = 0.27), respectively, during the study period. The trend in the energy fluxes showed highly heterogeneous distribution, but majority of the trend at non-disturbed ecosystems was positive, especially for H. Increase in both fluxes was most significant in boreal spring: 0.72 W m-2 yr-1 (R2 = 0.53 and p < 0.01) for H and 0.53 W m-2 yr-1 (R2 = 0.44 and p < 0.18) for LE. This positive trends in the energy fluxes coincided with warming air temperature in interior region, indicating that recent spring warming could induce a positive feedback through the change in surface energy budget of terrestrial ecosystems in Alaska.