Analysis of trends in surface energy fluxes under hot conditions using remote sensing products

Studying land-atmosphere interactions is important for understanding the mechanisms leading to changes in temperature and precipitation extremes. However, the non-conservation of energy and water in most products and their coarse spatial and temporal resolution hamper the study of land-atmosphere feedbacks. The combination of remote sensing data and modelling frameworks allows to greatly improve the spatial coverage and resolution of data products. Here, we investigate trends in surface fluxes over Europe using the new data product generated with the high-resolution land surface fluxes from satellite and reanalysis data (HOLAPS) framework. HOLAPS is a one dimensional modelling framework that solves the energy and water balance at the land surface, providing consistent surface and soil variables derived from remote sensing data and reanalysis products as forcings. The evaluation of the HOLAPS product against eddy covariance measurements shows slightly better results than other ET and H products at daily scales in summer (KGE > 0.0 for ET and KGE > -0.3 for H) and during hot extremes (KGE > -0.15 for ET and KGE >-0.7 for H), while the state-of-the-art products show KGE > -0.49 for ET and KGE > -1.2 for H in summer and KGE > -0.49 for ET and KGE > -1.5 for H during hot extremes. These results together with the 1D conservation of energy and water in the modeling framework makes this product the perfect tool for the analysis of trends in surface energy and water fluxes during the last decades. Preliminary results for the period 2001-2016 reveals a larger increase in the energy reaching the surface during the hottest month of the year than during summer over central Europe and the Mediterranean coast. This extra energy is released as sensible heat over dry areas during the hottest month of the year. In areas where soil water is available, the extra energy available during the hottest month is released as latent heat flux, adding it to the already large latent heat flux during summer. These results support previous analyses indicating an increase of latent heat flux during hot conditions at monthly scales. However, trends at higher temporal resolutions should be examined to improve the robustness of this conclusion.