Recent studies have looked at co-occurring combinations of natural disasters/hazards, referred to as compound risks (IPCC, 2021; Kruczkiewicz et al., 2021; Zscheischler et al., 2018). In the last couple of years, many parts of the world experienced the compound risks of extreme heat and COVID-19. A significant number of people suffered from hot indoor environments during shelter-in-place (SIP) in the summer. To reduce the impact of indoor extreme heat exposure, “cooling shelters,” which provide a cool indoor environment to the public, were used to be provided. However, during the SIP, cooling centers are closed or operated with limited access. Therefore, some local governments decided to provide cooling units to vulnerable groups. However, since the 1980s was the most recent nationwide household air conditioning (AC) ownership survey, this has been a barrier to new heat prevention measures. Therefore, this study partnered with the real estate company “Estated” and sought to answer these questions. How can we identify areas with the most significant AC needs? What are the socioeconomic characteristics related to AC prevalence?

Our study is focused on Florida State, US. Statewide temperatures are higher than 30 degrees Celsius during the summer, and some residents lack AC units to manage indoor environments. To answer the questions, we gathered property detail information at the census tract level from a real estate company, Esated, and American Community Survey (ACS)(2019). Among many property characteristics, we included the US census definition of county, census tract, lot size acres, year built, market value, the average number of rooms per residence, % households with complete plumping, and air conditioning types. Socioeconomic variables such as % 65 or over lives alone, % renter-occupied, % Black or African American, % Hispanic or Latino, median household income have shown relationships with AC ownership in the previous research. Recent studies argue that “redlining,” or historic disinvestment in neighborhoods with foreign-born populations, is related to contemporary health inequalities (Nardone et al., 2021; Schell et al., 2020). Therefore, we considered historical % Black or African American from ACS 1970. The Urbanicity index from The United States Department of Agriculture (2013) was included to consider the geographical characteristics.

We conducted Moran’s I test with the count of households without AC to examine spatial autocorrelation in Florida. Then we applied Local Indicators of Spatial Association (LISA) to identify local clusters and local spatial outliers of AC availability in Florida. We used the Spatial Durbin Model (SDM) to examine the relationship between the percentage of households without AC and socioeconomic variables.

The results from Moran’s I and LISA analysis illustrate a significant spatial autocorrelation of AC ownerships in Florida. We applied SDM to investigate the association between socioeconomic, urbanicity, and percentage of households without AC. We found lower AC access disparities by sociodemographic. Black/ African American households, low median household income, and % 65 or older living alone had lower rates of AC ownership. Based on our findings, we suggested prioritizing these groups for providing resources such as AC units, energy subsidies, and green spaces. For example, to minimize overall health risks associated with heat, policies such as the Housing Energy Assistance Program and Low Income Home Energy Assistance Program (LIHEAP) should target neighborhoods with low AC prevalence and/or lower socioeconomic position (Fraser et al. 2017).

Yoonjung Ahn is a Ph.D. Candidate at the FSU Department of Geography.