Evaporation-Precipitation Coupling in a Global Climate Model (EC-earth)

Abstract

State of the art climate models still render substantial present-day biases over land, most notably in summer, i.e. the mean of (most) global climate models render a strong underestimated of evaporation and precipitation in (semi- ) dry regions, while overestimating these fluxes in the wet regions. These biases in the hydrology strongly affect the surface temperature biases in summer. In order to improve the surface biases of global climate models, an accurate hydrology is crucial. This study emphasizes the importance of the correct evaporation-precipitation coupling to achieve the correct hydrology (soil water content, evaporation, precipitation) in all seasons. Furthermore, the study elucidates the non-local impact of evaporation on precipitation. In the first results section, we compare the hydrology of EC-earth versus the reanalysis ERA-Interim using a moisture tracking scheme. We implement a moisture tracking scheme to track all water from continental evaporation forward in time, which allowed us to analyse how precipitation and continental evaporation are linked. Our analysis indicates that the evaporation-precipitation coupling is stronger in EC-earth. Obviously, a too strong evaporation-precipitation coupling works in both ways. Thus, the precipitation response can feedback on an initial deficit or excess in evaporation. In the seasonal cycle context, this means that the precipitation response can amplify drying that is initiated in spring. In regions where evaporation is limited by the soil water content, precipitation will respond too strongly to the arising decrease in evaporation in (semi-) dry regions. Conversely, where continental evaporation is not limited by the soil water content, the evaporation will increase in spring and will lead to an overestimation of precipitation. In the second part of this research, we examine EC-earth’s response to a heterogeneous perturbation that increased the runoff in (semi-) dry regions. We find propagation of drought throughout summer, which is not directly forced by our perturbation. The direct impact of the increased runoff leads to drying, indirectly, the drying of the soil is affecting its surroundings, referred to as the quasi-local impact. Conclusively, the drier soils are further propagating drought indirectly through (1) the quasi-local impact of evaporation on precipitation and (2) the quasi-local decrease in clouds. In the third part, we investigated the link between source and destination of atmospheric water. We show where most of the continentally evaporated water is accumulating in the atmosphere, i.e. the destination. By decreasing the continental evaporation, we observed a substantial decrease in the (remote) destinations of continental water. Thereby proving there is a clear connection between source and destination of water. The link between source and destination of precipitation was found less robust because the atmosphere is responding to a decrease in (continental) water content. Hence, the atmospheric responds generally increases the amount of water in the column from oceanic origin. The total amount of water in the atmosphere even increased in some regions. Conclusively, there is a link between source of water and precipitation, but it can be diminished (or even reversed) due to the increase in (oceanic water) convergence.