Effects of the realistic vegetation cover on predictions at seasonal and decadal time scales

<p>Vegetation is a relevant and highly dynamic component of the Earth System controlling, amongst others, surface roughness, albedo and evapotranspiration; its variability shows changes in seasons, interannual, decadal and longer timescales. In this study, we investigate the effects of improved representation of vegetation dynamics on climate predictions at different timescales: seasonal and decadal. To this aim, the latest generation satellite datasets of vegetation characteristics have been exploited, and a novel and improved parameterization of the effective vegetation cover has been developed. The new parameterization is implemented in the land surface scheme HTESSEL shared by two state-of-the-art Earth system models: ECMWF SEAS5 and EC-Earth3. The former model is used for sensitivity at seasonal timescale, while the latter is used for sensitivity at decadal timescale.</p><p>Both seasonal and decadal experiments show considerable sensitivity of models' surface climate bias with large effects on December-January-February (DJF) T2M, mean sea level pressure and zonal wind over middle-to-high latitudes. Consistently, a significant improvement in the skill for DJF T2M is found, especially over Euro-Asian Boreal forests. In seasonal experiments, this improvement displays a strong interannual coupling with the local surface albedo.&#160;From the region with the most considerable T2M improvement, over Siberia, originates a large-scale effect on circulation encompassing Northern Hemisphere middle-to-high latitudes from Siberia to the North Atlantic. As a result, in seasonal experiments, the correlation between the model NAO index against the ERA5 NAO index improves significantly.</p><p>These results show a non-negligible effect of the vegetation cover on the general circulation, especially for the northern hemisphere and on the prediction skill.</p>