Energy-saving potential of cool roofs at the urban scale: A case study of Xiamen city

Abstract

Urban-scale energy modeling of cool roof retrofits requires detailed knowledge of rooftop. This study developed a high-resolution Urban Building Energy Modeling (UBEM) framework that integrates roof reflectance data for Xiamen City, China. First, building types were classified using Geographic Information System (GIS) and machine learning techniques. Then, historical satellite imagery was analyzed to estimate the building year of each building. Roof color and reflectance were subsequently evaluated using Sentinel-2 and high-resolution imagery. This process resulted in the creation of a multi-source geospatial dataset comprising 37,132 buildings in Xiamen. A physics-based UBEM was then constructed using AutoBPS for 31,608 buildings to simulate energy performance under three scenarios: baseline, measured reflectance, and cool roof retrofit. Compared to the default reflectance value of 20 %, actual roof reflectance in Xiamen reduced energy consumption by 1.8 % (0.94 GWh). Furthermore, three types of cool roof technologies, spray coatings, membranes, and metal roofs, achieved annual energy savings of 4.42 % (2.32 GWh), 3.58% (1.88 GWh), and 2.27 % (1.19 GWh), respectively. Their corresponding peak-load reductions were 10.7 %, 8.97 %, and 5.5 %. Among the 31,608 buildings modelled, 3790 buildings (approximately 12 %) met the economic feasibility criterion of a payback period under 10 years. Of these, 1298 buildings were more suitable for membrane application and 2492 for spray coating. Over their life cycle, these buildings would require an initial investment of 35.3 million CNY and maintenance costs of 39.3 million CNY, yielding electricity savings of 151.67 GWh, equivalent to 89.6 million CNY, and resulting in net savings of 15.0 million CNY.