Cite the code and data:
This repository contains data and code needed to recreate Figure 2 from the following publication:
Drake, John; Marty, Eric; Gandhi, Kamal; Welch-Devine, Meredeith; Bledsoe, Brian; Shepherd, Marshall; Seymour, Lynne; Fortuin, Christine; Montes, Cristian. “Disasters collide at the intersection of extreme weather and infectious diseases.” Ecology Letters, accepted for publication 05-Feb-2023.
Preprint DOI: https://doi.org/10.22541/au.166999211.18330817/v1
The preprint DOI will point to the published article when published.
Data and code prepared by Eric Marty.
The file data/master_dataset_hsa_2020.Rds contains COVID-19 outcomes
and social vulnerability scores for Health and Human Services (HHS)
regions 4 and 6, by Hospital Service Area (HSA).
Data from COVID-19 outcomes (cases and deaths) and social vulnerability (SoVI and BRIC) were originally at the county level. We aggregated the county-level data to the Hospital Service Area (HSA) level by population, using data from the 2020 US census.
File format: The file is an R data file containing an R dataframe of
class sf, or “simple features.” The columns of the dataset are
described in the file data/master_dataset_hsa_2020_metadata.csv.
COVID-19 cases and deaths were derived from county-level cases and deaths reported by Johns Hopkins University:
COVID-19 outcomes were aggregated to HSA by population.
The Social Vulnerability Index (SoVI) 2010-2014, developed by the Hazards & Vulnerability Research Institute (HVRI) at the University of South Carolina provides measures of social vulnerability to environmental hazards for US counties.
General information on the SoVI is available at https://sc.edu/study/colleges_schools/artsandsciences/centers_and_institutes/hvri/data_and_resources/
The SoVI data inputs consist of 29 census variables from the 2010 US Census, normalized as percentages, per capita values, or density functions.
Input variables were standardized using z‐scores, giving variables with mean 0 and standard deviation 1. The overall SoVI score is an additive model using PCA to determine weights of components in each of 8 factors, which are summed to obtain the overall score.
We aggregate from county-level to HSA-level scores using the population-weighted mean of county-level SoVI scores (mean 0, standard deviation 1), using 2010 Census numbers.
Original SoVI 2010-2014 scores by county are available here:
The Baseline Resilience Indicators for Communities (BRIC), developed by the Hazards & Vulnerability Research Institute (HVRI) at the University of South Carolina provides measures of resilience for US counties.
BRIC input variables are grouped into “capitals,” normalized 0-1 and averaged to obtain a BRIC capital score for each capital. (1 indicates high resilience.)
The overall resilience score is a sum of 6 capital scores, with theoretical maximum of 6.
We aggregate from county-level to HSA/HRR-level scores by taking the population-weighted mean of BRIC capital scores for each HSA, then summing HSA-level BRIC capital scores to obtain the BRIC overall score for each HSA.
Origical county-level data are available here:
References for SoVI and BRIC:
Cutter, S.L., Ash, K.D. & Emrich, C.T. (2014). “The geographies of community disaster resilience.” Glob. Environ. Change, 29, 65–77.
Cutter, S.L., Boruff, B.J. & Shirley, W.L. (2003). “Social vulnerability to environmental hazards.” Soc. Sci. Q., 84, 242–261.
The file data/hurr_tracks.sf.Rds contains Atlantic hurricane tracks
for the 2010 through 2021 seasons.
Data from COVID-19 outcomes (cases and deaths) and social vulnerability (SoVI and BRIC) were originally at the county level. We aggregated the county-level data to the Hospital Service Area (HSA) level by population, using data from the 2020 US census.
File format: The file is an R data file containing an R dataframe of
class sf, or “simple features.” The columns of the dataset are
described in the file data/hurr_tracks_metadata.csv.
The original hurricane track data were downloaded from the US National Oceanic and Atmospheric Administration’s “Atlantic hurricane database” (HURDAT2) 1851-2021, here:
https://www.nhc.noaa.gov/data/hurdat/hurdat2-1851-2021-100522.txt
See https://www.nhc.noaa.gov/data/hurdat/ for the latest data.
The file Map.Rmd contains the R code needed to reproduce figure 2. The
file is in R Markdown format.
A PDF containing the same code and code outputs, including the figure,
is also available: Map.pdf
The file R/visualization contains functions needed to create the map.
The code produces the figure in jpeg and pdf formats:
covid_sovi_hurr.jpg and covid_sovi_hurr.pdf.