Statistical Downscaling for urban meteorology at hectometric scale

Predicting the weather in urban environments is a complex task because of the highly heterogeneous nature of the urban structure. However, there are many issues inherent in urban meteorology, such as thermal comfort and building’s energy consumption. Those stakes are linked to highly heterogeneous meteorological variables within the city such as temperature, humidity, wind, net radiative flux and city characteristics, including building uses and characteristics. State-of-the-art meteorological models with hectometric resolution, like the Meso-NH (Lac et al. 2018) research model, coupled with advanced urban canopy model such as TEB (Masson 2000), can provide accurate forecasts of urban meteorology. However, they require too much computing power to be deployed operationally.

Statistical downscaling techniques are machine learning methods enabling the estimation of a fine resolution field based on one or several lower resolution fields. ARPEGE is the operational planetary model of Météo-France and operates at a resolution of 5km over France. Using Meso-NH simulations covering Paris and the Île-de-France region, a statistical downscaling has been carried out aiming to obtain a temperature field at 300m resolution using simulation outputs from the ARPEGE model.

Several scores have been computed in order to evaluate the work. The model is able to represent fine scale temperature heterogeneities over the targeted domain, as well as the diurnal cycle. Particularly the statistical model is able to make the difference between night and day conditions and to reproduce the urban heat island effect. Sensitivity tests to different inputs of the model such as input resolution were conducted and enable to put forward the main characteristics impacting the downscaling model results.

 

Lac et al. 2018 : https://doi.org/10.5194/gmd-11-1929-2018

Masson 2000: https://doi.org/10.1023/A:1002463829265