Selection of Weather Files and Their Importance for Building Performance Simulations in the Light of Climate Change and Urban Heat Islands

Abstract

Building performance predictions and their reliability rely heavily on weather data inputs. Climate is affected by spatial and temporal differences related to climate change and urban heat island effects, but the weather files used in building performance simulations (BPS) often remain unchanged and may represent weather observations generated from inadequate space and time for their application. This study investigated Swedish weather data using statistical methods and analysed i) the local differences related to rural and urban microclimates and ii) the country-wide differences linked to climate change; by comparing recent observation data to the respective EnergyPlus Weather (EPW) files. The findings reveal that there are significant differences between rural and urban temperature means, and that outdated model years of weather data files make them unsuitable for BPS. The impact of using an inadequate weather file based on changes in recent climate in Sweden can lead to an overestimation of heating demand by 6.5 % on average, while the impact is higher for warmer climates – up to 12 %. The combined impact including climate change and urban heat island effects can lead to a heating energy overestimation by 12 % to 19 %, based on the Stockholm example. On the other hand, it was found that although the global radiation means saw a slightly increasing trend, its impact on the BPS remains inconclusive. The study highlights the importance of selection of adequate weather data for BPS keeping in mind the spatial and temporal influencing factors.