Freshwater Availability Anomalies and Outbreak of Internal War: Results from a Global Spatial Time Series Analysis. Thorkelson, C. L. 2005.
Freshwater Availability Anomalies and Outbreak of Internal War: Results from a Global Spatial Time Series Analysis [link]Paper  abstract   bibtex   
We investigated the relationship between water availability and internal war outbreak. This work constitutes the first test of climate-security connections utilizing global subnational time series data. We created harmonized spatial time series databases on a subnational global grid of internal war, renewable freshwater surface water resources (in the form of runoff), rainfall deviations and population for the period 1980-2002. We utilize national-level data on infant mortality, political institutions, and trade openness as controls. We find that at the global scale there is a highly significant relationship between rainfall deviations and the likelihood of outbreak of a high-intensity internal war. When rainfall is significantly below normal, the likelihood of conflict outbreak is significantly elevated in the subsequent year. We do not find a similar effect for the mean annual runoff at the global scale, but find some evidence at the continental scale. We also find no significant relationship between rainfall deviations and the onset of low-intensity internal wars. The capacity to geographically reference social science and biogeophysical data sets will create new opportunities for hypothesis testing with respect to the sources of internal conflict in the fast of climate change invariability.
@misc{thorkelson_freshwater_2005,
	title = {Freshwater {Availability} {Anomalies} and {Outbreak} of {Internal} {War}: {Results} from a {Global} {Spatial} {Time} {Series} {Analysis}},
	url = {https://scholar.princeton.edu/cthorkel/publications/freshwater-availability-anomalies-and-outbreak-internal-war-results-global-spa},
	abstract = {We investigated the relationship between water availability and internal war outbreak. This work constitutes the first test of climate-security connections utilizing global subnational time series data. We created harmonized spatial time series databases on a subnational global grid of internal war, renewable freshwater surface water resources (in the form of runoff), rainfall deviations and population for the period 1980-2002. We utilize national-level data on infant mortality, political institutions, and trade openness as controls. We find that at the global scale there is a highly significant relationship between rainfall deviations and the likelihood of outbreak of a high-intensity internal war. When rainfall is significantly below normal, the likelihood of conflict outbreak is significantly elevated in the subsequent year. We do not find a similar effect for the mean annual runoff at the global scale, but find some evidence at the continental scale. We also find no significant relationship between rainfall deviations and the onset of low-intensity internal wars. The capacity to geographically reference social science and biogeophysical data sets will create new opportunities for hypothesis testing with respect to the sources of internal conflict in the fast of climate change invariability.},
	publisher = {International Workshop on Human Security and Climate Change, Holmen, Norway, 21 to 23 June 2005},
	author = {Thorkelson, C. L.},
	year = {2005},
	keywords = {CK, Untagged},
}

Downloads: 0