Impact of improved land surface model physics on simulated climate variability and change

Resumen

The land is a pivotal component of the Earth and its climate system since many processes of natural variations in the climate system, which affect the environment and human society, are governed by the land surface. Hence, a good representation of the thermal and hydrological states of the land surface in climate models is important to have a realistic simulation of the coupling between the atmosphere and the lito-biosphere. An influencing factor for improving the realism of the ground energy and water balance in climate models is the depth of the land zero-flux Bottom Boundary Condition Placement (BBCP) that ensures energy preservation in the climate model. Despite recent improvements in modeling land surface processes in climate models, only limited attention has been directed toward the effect of the BBCP in Land Surface Models (LSMs) and its impact on the representation of terrestrial thermodynamics. Previous analytical and modeling studies suggest that the simulation of subsurface thermodynamics in current-generation climate models is not accurate due to the zero-heat-flux BBCP being imposed too close to the surface. Under conditions of climate change, it can be expected that the warming signal at the surface propagates into the ground deeper than the commonly used depth of 3¿10 m in most climate models. An insufficiently deep land component in current-generation climate models compromises the simulation of the terrestrial thermal state and can influence land-atmosphere interactions. Further improvements in LSMs relate to the representation and sensitivity of coupling processes between the ground thermodynamic and hydrological regimes. As moisture is one of the main drivers of near-surface climate interactions, the hydro-thermodynamic coupling is crucial for studying the impacts of perturbations caused by human activity. Under climate change conditions, some areas and ecosystems are more vulnerable to a rapidly warming world than others. Arctic regions, for example, experience a three times larger temperature increase than the rest of the globe, which makes them more susceptible to climate change. The rapid warming is caused by positive snow-albedo climate feedback that leads to arctic amplification and exacerbates the degradation of soil permafrost, which releases additional greenhouse gas emissions to the atmosphere from warming-induced soil microbial activity. Therefore, a more realistic thermodynamic and hydrodynamic simulation of land surface processes is desired to more accurately assess anthropogenically forced changes in warming-sensitive regions. The overarching objective of this thesis is to provide a deeper understanding of the hydro-thermodynamics of land surface processes that can be used to improve current-generation climate models.