Bridging balances: water and energy in the urban climate

Summary

Urban environments experience substantial temperature enhancement due to among others increased imperviousness and reduced vegetation compared to the countryside. These landcover modifications exacerbate urban heat stress, necessitating strategies to mitigate this stress. One such strategy involves increasing evapotranspiration. Evapotranspiration appears as a term in both the surface water and surface energy balances, yet these balances are often studied separately. This research bridges the gap between the two by focusing on evapotranspiration, the shared flux, and how evapotranspiration influences urban climate dynamics.---The thesis explores how the water balance impacts the energy balance, using observations and model results from cities worldwide. The thesis aims to quantify the urban water storage capacity and its recession rate, evaluate and suggest improvements to the water balance representation in urban land surface models, and understand the role of surface cover in neighborhood-scale evapotranspiration.---A novel method was developed to quantify urban water storage directly from the decline of evapotranspiration using eddy-covariance observations. This eliminates the need to measure every contributing element. The method estimates water storage capacity without needing to measure every contributing element, using the decline of evapotranspiration during dry periods. Applied to 14 urban sites worldwide, this method showed that urban water storage capacities were at least five times smaller than those in natural environments, with recession timescales ranging from 1.8 to 20.1 days.---Next, 19 urban land surface models from the Urban-PLUMBER project were evaluated, revealing that none of the models fully closed the water balance. In total, 57% of model-site combinations showed mismatches in flux magnitudes. This highlights the need for improved runoff parameterizations and water balance closure in urban climate models. Therefore, the thesis examined how surface runoff could be captured better. The study concludes that most models do not include all necessary runoff processes leading to an underestimation of runoff. This underestimation was especially pronounced during heavy rainfall events, suggesting that including more runoff processes would improve model performance.---By analyzing eddy-covariance flux observations taken at two sites in Berlin, the mosaic of surface covers is linked to evapotranspiration. Vegetation was found to contribute more than proportional to its surface area. Impervious surfaces, while contributing less, still contribute meaningfully to the total evaporation. Numerical large-eddy simulations of the same sites highlighted the importance of sensor location and the observation "footprint" in accurately interpreting evapotranspiration data.---The thesis concludes that evapotranspiration is water-limited in urban areas, that urban land surface models need significant improvements, and that surface cover plays a vital role in evapotranspiration dynamics, influencing urban heat mitigation strategies.