Mesoscale Analysis of Climatic Conditions in a Central/Eastern European Urban Region Using WRF Simulations

One of the major anthropogenic effects can clearly be recognized in urban areas. The artificial surface covers significantly modify the surface energy exchange processes through modification of naturally covered lands resulting in altered local wind and temperature patterns because of the presence of buildings. The architectures’ three-dimensional extensions certainly affect the incoming radiation, the sky-view factors as well, as the wind fields, resulting in specific local microclimatic conditions within the extended agglomerations of large cities. The increased temperature in the central built-up areas and the cooler surrounding of the cities lead to the urban heat island phenomenon, which is widely studied both with observations and numerical models.

In this study the Weather Research and Forecasting (WRF) mesoscale model coupled to multilayer urban canopy parameterisation is used to investigate this phenomenon for Budapest and its surroundings located in Central/Eastern Europe. Before starting the WRF simulations, the detailed surface has to be defined according to the actual conditions, for which CORINE and OpenStreetMap databases are used, both including buildings, different land use categories, and water bodies. Our new land use dataset serving as input for WRF runs distinguishes five urban categories: (i) low-intensity residential, (ii) high-intensity residential, (iii) commercial/industrial, (iv) airport, and (v) greenbelt area. For the simulations the initial meteorological fields are derived from the publicly available GFS (Global Forecast System) outputs. Simulations are completed for one-week-long period with an intense heat wave in summer 2015, thus, the atmospheric conditions include weak wind and clear sky, which is ideal to use satellite data as reference.

In order to keep the stability of the simulations, the entire downscaling is carried out in several steps using gradually smaller domains embedded to each other. Thus, three embedded target areas have been determined for this modeling study, the 15 000 km² large external area covers the whole Carpathian region with 10 km horizontal resolution, whereas the 7200 km² large innermost domain covers Budapest (the capital of Hungary) and its surroundings with 1 km grid resolution. The calculated skin temperature over Budapest is compared to the surface temperature fields of Land Surface Temperature and Emissivity product (MOD11A1) derived from the remotely sensed measurements of sensor MODIS (Moderate Resolution Imaging Spectroradiometer) onboard satellites Aqua and Terra.

On the basis of our results, the patterns of surface urban heat island intensities are quite similar to the observations, but the daytime and nighttime intensities differ. The average root mean square error (RMSE) of the simulations compared to observations is lower at nighttime than at daytime. This can be partly explained by the fact that the outgoing longwave radiation is affected by the incoming shortwave radiation, which is present only during daytime with possible quick changes due to cloud cover changes.