Natural Disasters

Follow Us:    |   Share:

Geocoded Disasters (GDIS) Dataset, v1 (1960 – 2018)

Downloads

Data:
View Recommended Citation(s)
X
Recommended Citation(s)*:

Rosvold, E., and H. Buhaug. 2021. Geocoded Disasters (GDIS) Dataset. Palisades, New York: NASA Socioeconomic Data and Applications Center (SEDAC). https://doi.org/10.7927/zz3b-8y61. Accessed DAY MONTH YEAR.

ENW (EndNote & RefWorks)†
RIS (Others)

Rosvold, E.L., and H. Buhaug. 2021. GDIS, a Global Dataset of Geocoded Disaster Locations. Scientific Data 8:61. https://doi.org/10.1038/s41597-021-00846-6.

ENW (EndNote & RefWorks)†
RIS (Others)

* When authors make use of data they should cite both the data set and the scientific publication, if available. Such a practice gives credit to data set producers and advances principles of transparency and reproducibility. Please visit the data citations page for details. Users who would like to choose to format the citation(s) for this dataset using a myriad of alternate styles can copy the DOI number and paste it into Crosscite's website.

† For EndNote users, please check the Research Note field for issues with importing authors that are organizations when using the ENW file format.

This data set consists of 39,953 locations for 9,924 disasters that occurred worldwide for the years 1960 to 2018 and is available in Geodatabase (.gdb), GeoPackage (.gpkg), CSV, and R-Workspace (.rdata) formats, along with the R-script source code (.r) and a Codebook (.pdf).

The Geodatabase, GeoPackage, and R-Workspace formats include the geography of each unit, while the CSV only contains the attributes with latitude and longitude coordinates for the centroids of each unit. To obtain information on the disasters, the geographic data can be merged with EM-DAT data on disaster impacts via the "disasterno" variable. Subsequent to the release of this data set, EM-DAT changed the code for disasterno to include a dash followed by the ISO3 code (e.g., 2016-0032-FRA for a disaster in France). Users wishing to link GDIS to EM-DAT may wish to use the R code provided below to remove the last four characters of the disasterno string (download the "R-script source code"). The geographic data can be used with the PRIO-GRID conversion CSV file to merge the GDIS data into the PRIO-GRID structure.