Sitting in the dark- The impact of shading on evapotranspiration in complex urban landscapes

Evapotranspiration (ET) is a key parameter in the water exchange of atmosphere, plant and soil and was studied on many different scales. In urban environments, the estimation of evaporation is particularly difficult, as it is effected by complex patterns of shading, which varies on a very small scale as a function of street canyon layout and orientation. Moreover, shading varies not only in space but also in time due to the seasonal orientation and altitude of the sun. Therefore, for a correct ET assessment, the diurnal variations as well as the annual variations of shading must be taken into account. For this purpose, radiation is divided into direct and diffuse radiation; in case of complete shading only the diffuse radiation was used for ET estimation, reducing the direct radiation to zero. The diffuse radiation is further influenced by the amount of visible sky in a street canyon as a function of street widths, which can be derived using the sky view factor.
To reduce the uncertainty of ET estimation in the built environment, a process-based model was developed with an hourly resolution that takes into account the particularly heterogeneous spatial variability of urban surfaces. To assess the impact of shading on ET, six different shadow scenarios as well as two typical urban soil sealing scenarios for a wide and a narrow street canyon were analysed regarding differences and similarities of radiation and resulting actual and potential ET of a street tree as well as soil water dynamics.
The model scenarios showed that ET is highly influenced by shading. Furthermore, shadow scenarios affect actual ET (ETA) differently during the vegetation period: whereas in April the ETA is higher for fully exposed sites, this changes by June when less exposed sites periodically have higher ETA rates. This difference is directly connected to alteration of soil moisture dynamics, for a fully sun exposed site a soil moisture of 10 Vol% is already reached by June. For a shaded site the decrease to 10 Vol% takes two months longer.

In conclusion, the results highlight that it is essential to include the effects of shading in the quantification of vertical water fluxes in urban environments. Moreover, this new model approach will help to identify water shortage periods and critical locations for street trees.