Health-informed predictive regression for statistical-simulation decision-making in urban heat mitigation

Abstract

An integrated environmental, economic, and health-informed approach was developed as a decision making framework to assess the associated benefits of enriching the urban greenery cover to provide city cooling potential. The framework used an evidence-based statistical-simulation approach to support decision-making processes associated with heat mitigation strategies. A novel health-informed statistical modeling was developed by introducing log-linear Poisson regression based on clustering factors to increase model fit, control dataset overdispersion, and enhance model prediction. The statistical model was utilized to predict mortality records and emergency department visits based on changes in heat behavior. Microclimate simulations were developed utilizing an updated version of the Urban Weather Generator to assess the impacts of increasing greenery cover in York Region, Southern Ontario, Canada on ambient temperature, outdoor heat stress, and buildings' energy consumption. The heat-based health predictions of the York Region community confirmed the possible hazardous impact of climate change on health and the impact of extreme heat on mortality records and health system use. By intensifying the urban greenery cover, the results demonstrated significant reductions in ambient temperature, outdoor heat stress, neighborhood average daily energy use, expected mortality counts, and emergency department visits. The economic module has reported the expected benefits in terms of lower visits to emergency departments, avoiding premature mortality, reduced energy consumption, and reduced productivity losses. The proposed framework was developed as a flexible decision-making tool for policy-makers and stakeholders to assess environmental, economic, and health benefits of heat mitigation strategies within urban contexts.