About

Methodology

Defining Historical Heat Health Events

A Heat Health Event (HHE) is any event that reults in negative public health impacts, regardless of the absolute temperature. Each local area has a unique HHE specific to its climate and the historical sensitivity of people in that area to past heat events. We arrived at these historical HHE thresholds by using a distributed lag non-linear model (DLNM). Daily meteorological data (1984-2013) was paired with the most recently available emergency department visitation data (2005-2013) to identify the signatures of past heat events associated with high effect sizes, or rather, heat-related public health impacts. We obtained meteorological data from the PRISM Climate Group and extracted data for minimum temperature (tmin), maximum temperature (tmax), minimum vapor pressure (vpdmin), and maximum vapor pressure deficit (vpdmax) at a daily time-step and at a resolution of 4 kilometers. We used emergency department (ED) data from the California Office of Statewide Health and Planning (OSHPD) for the years 2005-2013, covering the summer months (MJJAS).

Projecting Changing Conditions

Projections are based on a subset of the Localized Constructed Analogs (LOCA) downscaled projections developed by the Scripps Institute (Pierce at al., 2014; Pierce et al., 2015) including the twenty-four models that provide daily minimum and maximum relative humidity in addition to daily minimum and maximum temperature. LOCA projections were developed for both RCP 4.5 and RCP 8.5 in order to represent the difference between business as usual emissions (RCP 8.5) and a moderate mitigation scenario (RCP 4.5), resulting in a total of forty-eight projections, two for each model. The spatial resolution of the LOCA projections is 1/16th degree or approximately six kilometers, and daily values of the variables are available from 2006 through 2099. We analyzed the frequency and average characteristics of HHEs for twenty-year periods centered around 2020 and every ten years following through 2099. For heat signatures associated with each population type, and the July 2006 Heat Wave, we sorted the projections in ascending order based on the number of qualifying events. In order to characterize the range of projected outcomes, we then captured the HHE frequencies and average characteristics from the projections nearest the 5th, 33rd, 50th, 67th and 95th percentile.

Intrepreting the HHE Projections

CHAT uses a dataset that was designed to investigate how the frequency of HHEs will change throughout the 21st century, using an ensemble of 24 downscaled climate models. Results are shown for each location and time period, representing the 5th, 33rd, 50th, 67th and 95th percentiles in the ranking of HHE frequency. In addition to HHE frequency, the dataset also includes average maximum and minimum daily temperature and maximum and minimum daily relative humidity for the HHEs. You may notice some things that may surprise you, such as average HHE temperatures going down over time. This is because, as warming continues, more of the daily temperature distribution exceeds the HHE thresholds with an increasing number of days with relatively lower temperatures included in the averages. Peak temperatures during the parts of the year that included in the dataset are continuing to rise, however. Perhaps most surprising, you may notice HHE frequency going down over time, especially in warmer parts of the state. This is the result of the climate regime shifting relatively rare, distinct HHEs to more persistent HHE conditions. Even though the frequency of HHEs is decreasing, the total number of days exceeding HHE thresholds continues to increase.