Questions Arising While Co-producing Climate Projection Information Relevant to Philadelphia Public Health Extreme Heat Interests

The heat index metric (a function of temperature and humidity) is used often in the United States as a quantitative measure associated with human heat exposure.  Heat index values are part of warm season forecasts issued by the US National Weather Service, incorporated into heat warning systems used by governmental entities, and appear in many scientific publications, especially interdisciplinary studies linking public health with weather and climate. Yet, there is more than one way to calculate heat index values from observational data and from model predictions and projections. Not surprisingly, the choice of input data sets, computational algorithms used, and other methodological choices can lead to different quantitative results and hence uncertainties. Our experiences suggest that some sources of uncertainties are usually considered (e.g., scenario and model uncertainty, using the nomenclature of Hawkins and Sutton [2009]) by those without formal climate science training, whereas other uncertainties are not necessarily appreciated. Neglecting sources of uncertainty can lead to overconfidence in quantitative results produced in a particular study.  Here we present information developed during a research project investigating historical and projected daily maximum heat index values for the northeastern United States, with a focus on the city of Philadelphia. We illustrate how, in some cases, uncertainties associated with the choice of downscaling or bias correction methods, the observational data product used when bias correcting, the temporal resolution of weather and climate data, and common estimation methods used in some calculations, can lead to uncertainties whose magnitudes can rival those of scenario and model uncertainty. Comments will be offered regarding the practical challenges of seeking to promote improved practices when limited by data availability or other resources.
Hawkins, E., and R. Sutton, 2009: The Potential to Narrow Uncertainty in Regional Climate Predictions. Bulletin of the American Meteorological Society.