Skip to contents

Retrieves FEMA Major Disaster Declarations at the county level, aggregated by year and month. Tribal declarations are stored separately as an attribute.

Usage

get_fema_disaster_declarations(file_path = NULL, api = FALSE)

Arguments

file_path

The path to the raw data. If NULL (default), reads the most recently cached file for this dataset from get_openfema_cache_path().

api

If TRUE, access data from the API. Else (default), read from file_path (or the local OpenFEMA cache, if file_path is NULL).

Value

A dataframe comprising Major Disaster Declarations by month by year by county. Tribal declarations are stored as an attribute (tribal_declarations). Columns include:

unique_id

Unique identifier for each observation.

GEOID

Five-digit county FIPS code.

year_declared

Year the disaster was declared.

month_declared

Month the disaster was declared (1-12).

declaration_title

Title(s) of the disaster declaration(s).

incidents_all

Total count of disaster declarations in the county-month.

incidents_natural_hazard

Count of natural hazard declarations.

incidents_*

Additional columns for other incident types, each of which reflects the count of the given incident type.

Details

Data are from FEMA's OpenFEMA API. See https://www.fema.gov/openfema-data-page/disaster-declarations-summaries-v2. Statewide declarations are expanded to all counties in the state.

Connecticut's pre-2022 counties (FIPS 09001-09015) are crosswalked onto the 2022-vintage planning regions (09110-09190) using a binary any-overlap assumption: a planning region is treated as having received a declaration if ANY overlapping pre-2022 county received it, even if only part of that county's area falls within the region. This is not a proportional/population-weighted allocation (unlike the crosswalks this package uses for dollar-valued FEMA datasets), since a declaration is a binary event, not a divisible amount.

Examples

if (FALSE) { # \dontrun{
get_fema_disaster_declarations(api = TRUE)
} # }