Evaluating Future Heat Health Emergencies in Philadelphia, PA: Blending Model Projections, Observations, and Downscaling Methods

Downscaled climate projections can play a key role in the development of local-scale adaptation and resiliency strategies for the current climate and for upcoming decades. One such example is city officials and their utilization of heat predictions and projections in Philadelphia, PA. The Philadelphia Department of Public Health defines Heat Health Emergencies (HHE) and Heat Cautions based on daily maximum heat index thresholds, duration of the exceedance of these thresholds, and time of year. Operationally, these extreme events are declared based on National Weather Service forecasts of temperatures and humidity at the Philadelphia International Airport weather station. When HHEs are declared, the public is informed via multiple communications channels and the city begins to take action, such as opening cooling centers and mobilizing other city services. Heat Caution conditions, which occur more frequently than HHE conditions, are communicated internally among city departments and partners, but are not broadcast to the general public. Using the same objective heat index criteria currently employed by the city, we examine the number of HHE and heat caution events simulated and observed during the historical period (1985-2014) and use Global Climate Model (GCM) data from four GCMs refined by two different bias correction/ statistical downscaling methods (QDM and MBC) to investigate how the occurrence of these extreme events may change in future decades. Using 400m resolution SURFEX model output focused on Philadelphia (Le Roy et al., in review), driven by a combination of reanalysis data and climate change signals from four CMIP6 climate models, we examine future projections of the occurrence of heat events, as well as the severity and intensity of those events. In addition to generating time-varying projections of the increased frequency of heat events meeting the current city of Philadelphia criteria, our experimental design allows us to evaluate the sensitivity of results to the choice of GCM, the choice of bias correction/downscaling method, and other methodological choices. When these uncertainty factors are communicated to and appreciated by practitioners involved with the delivery of climate services, it can promote better informed decision-making and contribute to the development of effective adaptation strategies.