Comparing micro climate weather conditions from the weather research and forecasting (wrf) model with typical meteorological year (tmy) data for building energy simulations

Urban microclimates, characterized by localized weather conditions, play a crucial role in urban building energy simulations. Comparing urban microclimates with TMY3 (Typical Meteorological Year) weather files is essential for understanding the limitations of using representative climate data in building energy simulations. In this study, we utilized the Automatic Building Energy Modeling (AutoBEM) framework to compare simulation results using TMY3 and WRF-derived EnergyPlus Weather Files (EPWs) for Arizona’s building stock. A total of 20,760 grid cells, each at a 1km-by-1km resolution across Arizona, were simulated using WRF for the period from May 15 to May 31, 2020. The WRF outputs for EnergyPlus were then extracted, processed for all grid cells, and converted into the time series EPW format. This time slice was then morphed into an existing EPW for Maricopa for each WRF grid cell. Based on this comparison, we aim to evaluate how well TMY3 data represents real-world urban conditions and determine whether more detailed, location-specific data are needed to improve the accuracy of building consumption simulations in a typical urban building energy modeling process.

The results indicated a 1.93% overall increase in total energy consumption across 10 Arizona counties when using WRF EPWs instead of TMY3. The northernmost counties, Coconino, Mohave, and Navajo, experienced a decrease in total energy consumption with WRF EPWs, while the remaining seven counties saw increases. We also conducted a preliminary analysis to evaluate how urban density affects the simulation. In high urban density grid cells with more than 300 buildings, the simulated total building energy consumption for the tile was almost always higher when using data from WRF compared to TMY3 files. In contrast, simulations for less dense areas were more similar with the two classes of weather data, but still favored WRF in terms of higher total energy consumption. This finding suggests that WRF can represent urban weather conditions with greater fidelity because its detailed grid- based outputs more accurately modeled urban changes and their effects on the micro-climate conditions of the local built environment. Instead of using representative weather files for an entire region, grid- based high-resolution weather data could provide a more reliable understanding and evaluation of the urban micro-climate’s impact on building stock energy consumption.