Linking buildings and cities: Advancing the InVEST Urban Cooling model with building height and local climate zones to assess urban overheating and nature-based benefits

Source abstract

Mitigating urban heat islands (UHIs), reducing heat stress, and improving thermal comfort are key objectives for cities facing heatwaves, especially as extreme heat events intensify due to climate change and urbanization. The InVEST Urban Cooling model (UCM), an open-source, spatially explicit ecosystem service model for assessing the cooling benefits of green infrastructure, provides a practical tool that captures the interactions among land cover, urban morphology, and biophysical cooling processes. Nonetheless, key urban morphological attributes—such as building height and local climate zones (LCZs)—have consistently been excluded from previous studies, and their impacts on model performance remain unclear. This research develops a reproducible workflow to simulate high-resolution (10-m) air temperature and cooling service distributions by integrating ground-based observations (iButton network and Meteorological Service Singapore data) with multi-source remote sensing data. Calibration using simulated annealing optimization, combined with leave-one-station-out cross-validation and comprehensive sensitivity analysis, offers a systematic and transparent basis for optimizing model parameters, assessing predictive performance, and characterizing parameter uncertainty. The results show that incorporating building height and LCZs substantially improves model accuracy in Singapore, with overall R 2 values ranging from 0.72 to 0.81 across diurnal periods, and this improvement is further validated in Hong Kong, London, and Zurich, where incorporating LCZs into the calibrated models increases R 2 by 6.67%–22.64%. Scenario simulations illustrate that vegetation and albedo interventions yield distinct, spatially variable cooling benefits. These findings highlight the effectiveness of integrating LCZ-based morphological information to enhance InVEST UCM performance and demonstrate the model’s reliability for evaluating nature-based cooling solutions and supporting climate-informed urban planning.