Document Type

Thesis

Degree Name

Master of Science (MSc)

Department

Geography & Environmental Studies

Program Name/Specialization

Human Geography

Faculty/School

Faculty of Arts

First Advisor

Robert McLeman

Advisor Role

thesis co-supervisor

Second Advisor

Colin Robertson

Advisor Role

thesis co-supervisor

Abstract

Through the second half of the 20th century, the North American Great Plains saw widespread rural out-migration, a continuation of trends that began with the Dust Bowl crisis during the Great Depression of the 1930s. As part of a wider academic focus on the roles climate and environmental changes have on migration, this research project sought to understand the relationship between drought conditions and rural population decline on the Great Plains. In this explorative research, census population data for Canada and the US from 1970-2010 were analyzed along with temperature, precipitation, and Palmer Drought Severity Index data for the same period using a variety of regression to seek out possible association between drought conditions and population loss at local scales. As part of this process, a novel index for identifying drought likelihood was also developed and tested. Results indicate that the significance and direction of the relationship between drought and rural population loss is spatially heterogenous. Geographically weighted regression models are demonstrated to have better predictive power than traditional regression methods, although that predictive power deteriorates through the decades in the study period. Small clusters of counties were detected where the drought-population loss is relatively strong in certain decades, but generally the results suggest that non-climatic factors were the primary drivers of population loss across the Great Plains. The modelling results are discussed in the context of a case study of Lincoln County, Colorado, a dryland county visited as part of field research for this project.

Convocation Year

2022

Convocation Season

Spring

Share

COinS