Representative meteorological data for long-term wind-driven rain obtained from Latin Hypercube Sampling – Application to impact analysis of climate change

Accurate estimation of wind-driven rain (WDR) load on building facades is paramount importance for the assessment moisture-induced damage risks. The response facade depends used meteorological data, which can show significant variation over time, especially considering climate change. In this study, a statistical approach based Latin Hypercube Sampling (LHS) to generate reduced samples, accurately represent long-term conditions WDR. Based cumulative distribution functions, generated samples with LHS are subset actual measured independent temporal information, and clustered around values highest frequency. Computational fluid dynamics (CFD) simulations WDR performed historical located in Victoria, BC, Canada previously validated methodology, determining parts receiving load. sensitivity study shows that sample size 200 LHS, corresponding 0.2% total data 4.1% during rainfall, sufficient replicate successfully spatial maximum discrepancy 7%. be easily modified model various scenarios respect change presented different terms rainfall intensity wind speed as predicted by future climatic conditions. results indicate highly conditions, even when kept constant only varied.