Global sensitivity analysis of an urban microclimate system under uncertainty: Design and case study

Over the last decade, significant efforts have been made to develop sophisticated physics-based urban simulators, given the undeniable need for sustainable-city initiatives to consider the potential impacts of climate change and massive urban growth. Nevertheless, even as a growing number of researchers have expanded their scope to the urban realm, there remain many problems resulting from the complexity of the urban microclimate, such as the Urban Heat Island (UHI) effect. This study is initiated with the intention to account for uncertainty in developing more coherent and integrated strategies concerning the energy and environmental issues in an urban system. The analysis builds upon the previously reported and updated Urban Weather Generator (UWG) to present a deep look into an existing urban microclimate system in Abu Dhabi (UAE). The case-specific baseline information is generated for the UWG and a global regression-based sensitivity analysis using the Monte Carlo technique is performed. Based on 30 candidate inputs covering the meteorological factors, urban characteristics, vegetation variables, and building systems, the uncertainty analysis indicates that the UWG is a fairly robust simulator to approximate the urban thermal behavior in downtown Abu Dhabi for different seasons. The identified significant factors will be the subject of future research to gain a higher resolution of critical urban simulation inputs, thereby providing more informed domain knowledge of the underlying mechanisms driving the microclimatic effect on the energy and environmental performances.