Assessing the impacts of physiography refinement on Stockholm summer urban temperature simulated with an offline land surface model

Urban climate information at high spatial resolution (tens or hundreds meter scale) is needed for local climate adaptation. Due to the high demand for computational resources by convection-permitting models (CPMs) that can resolve urban areas and the general lack of accurate urban physiography information, the downscaling of climate data to sub-km scale is challenging. This work explores the feasibility of simulating urban air temperatures using a much less computationally expensive offline land surface model (LSM) with refined physiography. The method is based on the SURFEX LSM downscaling to 300 m of a set of CPM simulations with an original grid spacing of 3 km. Results show that the influence of different physiography in the CPM atmospheric forcing on the SURFEX simulations decreases with the increase of forcing height, and it becomes statistically non-significant at a forcing height of 50 m or above. On the other hand, the impacts of physiography change on 2-m temperature (T2m) simulated by the offline LSM are highly correlated when using different physiography in the CPM simulations, regardless of the forcing height. The refined physiography dataset was produced using open-access products, which enables the proposed method to be used in other regions.