Mesoscale models

Heat island map for Berlin
© Fraunhofer IBP
Urban heat island map for Berlin

Our team of experts use the Weather Research and Forecasting (WRF) model to generate coarse-grid climate predictions. This is a numerical weather prediction system designed to serve both atmospheric research and operational forecasting needs. However, by including special parameterizations, the WRF model can also be apllied tot he urban canopy layer. In addition, these parameterizations have recently been coupled with building energy models to enable the estimation of the buildings’ energy use. The results of a WRF simulation can be interpreted directly or provide boundary conditions for a micro-scale CFD model such as PALM-4U.

WRF incorporates, among other modules, a fully coupled land surface model to simulate column land surface processes: the results of this model in terms of atmospheric conditions, short-wave/long-wave radiation, and precipitation are used to update the land state, which in turn influences the WRF simulation of atmospheric conditions. Over urban areas, an urban canopy parameterization will be invoked to properly account for urban morphological and thermo-physical features. Given the critical importance of soil moisture conditions in most European cities, our researchers further improve the standard implementation of WRF by incorporating the uncoupled high-resolution land data assimilation system (HRLDAS) in order to initialize the land state variables. The HRLDAS simulation time required to reach equilibrium can exceed one year, which makes it practically impossible to couple the HRLDAS with the computationally intensive WRF atmospheric model.

Two off-shoots of WRF, significant to urban climate studies, are WRF-Chem and WRF-Hydro. WRF-Hydro, a physics-based hydro-meteorological model, can simulate floods, hydrological states and spatial distribution of water resources and provides satisfactory results for flood forecasting. Decision makers can thus obtain information on location, timing and duration of inundations while fully accounting for landscape dynamics. WRF-Chem is used to simulate meteorological conditions together with air pollutant concentrations, with a focus on Nitrogen oxides, particulate matter and Ozone (O3). Consequently, various gas-phase chemistry and aerosol mechanisms have been added to the base WRF model.

Application examples:

  • high-resolution spatial mapping of the current or future urban heat island (municipalities)
  • flood control (municipalities)
  • avoidance of heat stress or air pollution (public health authorities, city dwellers)
  • probabilistic forecasting of extreme events (public safety agencies, insurance companies)
  • combined effect of urban heat island and global warming (municipalities, national energy policy decision makers)


  • Vogel, J.; Afshari, A. (2020), Comparison of Urban Heat Island Intensity Estimation Methods Using Urbanized WRF in Berlin, Germany, Atmosphere, Volume 11, 1338,

Highlight projects