Sensitivity and Uncertainty Analysis on Urban Heat Island Intensity Using the Local Climate Zone (LCZ) Schema: the Case Study of Athens

The Urban Heat Island (UHI) phenomenon commonly occurs in cities worldwide and may affect the health of millions of people, especially during heatwaves. To design and propose mitigation strategies, accurate urban microclimate simulation tools are needed but retrieving information for creating reliable models is often challenging. Sensitivity (SA) and uncertainty (UA) analysis can help to achieve valuable information and to produce informed decisions even in an uncertain scenario. However, to date, that are few studies that perform SA and UA on microclimate models. In the present study, the Urban Weather Generator (UWG) software has been used to apply SA and UA to the city of Athens, Greece. The urban model is established by considering the variation of input parameters that may occur within three of the most diffused Local Climate Zones (LCZ) in city centers. Results showed that the highest average Urban Heat Island Intensity (UHII) is obtained in LCZ2 followed by LCZ3 and LCZ5 (2.92, 2.78 and 2.60 degrees C, respectively). However, due to the overlapping of UHII output values among different LCZs, it is not possible to state with absolute certainty that a "lower" LCZ corresponds to a higher UHII in a specific city. The SA confirmed that LCZ parameters such as building surface fraction, aspect ratio and mean height of buildings are the most influential ones in this case study. The percentage of buildings with an operating Air Conditioning system is also a key parameter in LCZ2, which however is also among the most uncertain ones. Thus, this value is required to be well established in future research.