Heat Index & WBGT at Census-Tract Scale

Extreme heat is an increasing public health threat. While most epidemiological research relies on dry-bulb temperature, human thermoregulation is influenced by additional meteorological factors including humidity, solar radiation, and wind.

We present a high-resolution, open-source dataset of two physiologically relevant heat indices— Heat Index (HI) and Wet Bulb Globe Temperature (WBGT)—across census tract geographies in the contiguous United States from 1980 to 2019, for summer months (May–September). These indices are widely used to predict heat-related illness and occupational stress.

Using validated Python packages and physics-based models, we derive daily minimum, mean, and maximum values from satellite-informed meteorological inputs: NASA Daymet (1 km) and ERA5-Land (~9 km). Our calculations preserve physical coherence by aligning diurnal extremes of component variables.

To our knowledge, this is the first dataset providing daily, historic HI and WBGT estimates at 1 km resolution aggregated to census tract boundaries. Moreover, we provide an API enabling coordinate-based queries supporting integration into public-health and climate-adaptation tools. All workflows are reproducible via open-source notebooks.