SEBU: A novel fully automated Google Earth Engine surface energy balance model for urban areas

Abstract

Understanding the diverse energy exchange within the city's boundaries would enable better design of future urban planning and environmental policies and update current ones. More importantly, it would facilitate the establishment of smart cities, where human health and wellbeing is a top priority. In this regard, several energy balance models have been proposed for urban areas. Only recently, remote sensing images have been extensively used in these areas, but are still difficult to implement due to the large inputs' sources and types required as well as their complexity. Their accuracy can also be improved. Thus, in this paper, the Surface Energy Balance Model for Urban areas (SEBU) is proposed based on the 100-m Landsat images. It uses other sources as well such as Sentinel-1, MODIS and ERA5. SEBU is based on the innovative hot/cold pixels approach widely known in the agricultural-based models, but also includes several dynamic internal calibrations. It generates monthly turbulent sensible (Qh) and latent (Qe) heat values over a 100-m spatial resolution. When applied over seven different regions (i.e., Denver, New Hampshire, Basel, Heraklion, Singapore, Phoenix and Vancouver) in four contrasting climates (i.e., cold, arid, warm and equatorial) and when compared to local flux tower measurements, absolute mean error varied between 6.13 W m−2 month−1 for Qe and 14.46 W m−2 month−1 for Qh. More importantly, the novelty of SEBU does not lie only in providing reliable accuracy and full reliance on the remote sensing database, but also on its open-source nature and easy-accessibility over the Google Earth Engine (GEE) platform. Thus, SEBU has the potential to be scalable using the massive power and huge database found in GEE, where users need to specify the required date only. This would certainly assist researchers to access urban cilmate information in a timely manner. Policy makers and even local dwellers would then benefit from their findings. Furthermore, SEBU can be improved to accommodate current and future needs of its users, while ultimately, enhancing the urban surface energy models and related science.