Generating Future Weather Data for Building Energy Simulations: A Review of Methods, Applications and Challenges

With an increasing awareness of climate change and its effects on the built environment, climate change adaptation is changing traditional building design practices. Future weather data are essential for building energy simulation (BES) that informs a resilient and energy-efficient building design under climate change. While general circulation models (GCMs) provide future climate predictions, their outputs often require downscaling to improve spatial and temporal resolution and further methodological processing to generate weather data suitable for building-scale analysis. This study aims to examine the methods for generating and utilizing future weather data for BES, with a particular focus on bias correction and uncertainty quantification in GCM predictions. This study summarizes the prevailing methods for bias correction of GCM outputs and the generation of representative future weather data. The characterization of GCM uncertainty and its implications for BES results are discussed. It is shown that GCM outputs can be effectively used for BES to evaluate long-term effects of climate change under various climate scenarios, and the most cost-effective approach often involves a combination of statistical downscaling and adjustment of grid cell size, which balances the need for high-resolution, site-specific weather data with the demand for computational resources. In addition, key challenges are identified, including the selection of appropriate GCMs and climate scenarios, the trade-off between computational cost and representativeness, and the need to include both extreme and typical weather conditions. Furthermore, future research prospects are proposed. Through a synthesis of current advancements in future weather data generation methods, this study contributes to the robustness of climate-responsive building design.