How does the representation of vegetation within a urban land surface model influence the distribution of energy between the sensible and latent heat fluxes over a number of sites with varying land cover?

Results from the first international urban land surface model comparison showed that representing vegetation is the dominant physical process that needs to be represented for urban areas, although it was not clear from the results if the way in which the vegetation is represented is important. However, the urban comparison was limited to a single site, with specific fractional coverage of vegetation. So it is not clear if these results would hold for other urban sites with varying vegetation fractions within the footprint of the observations.

In this study we use the JULES model (one of the better performing models from the urban comparison), which includes a representation of vegetation, to assess how its performance changes over sites of varying vegetation fractions. This particular model uses a tile approach to vegetation and makes no distinction between vegetation in a rural environment and vegetation within the urban environment. As such, a number of sensitivity experiments are also used to assess the impact on model performance from setting up the model with vegetation parameter values that may be more appropriate for the urban environment.

Results will be presented to show how the model performs for the urban sites with different fractions of vegetation, and to identify the limitations of the model. The sensitivity studies will be used to demonstrate how the model could be improved and the importance of applying urban specific parameter values to the vegetation surface. Recommendations will be made for future model development, whilst the conclusions will refer back to the outcomes of the urban model comparison.