Using multi-resolution remote sensing to measure ecosystem sensitivity and monitor land degradation in response to land use and climate variability

Climate change and land degradation, which is defined as the decline in the productive capacity of the land, have profound implications for resource-based livelihoods and food security. In this dissertation, I use remote sensing to improve understanding of how climate variability affects the productivity of global pasturelands and to quantify the spatial and temporal patterns of land degradation in the Southern Cone region (SCR) of South America. In the first chapter, I characterize the sensitivity of global pastureland productivity to climate variability by analyzing the relationship between MODIS enhanced vegetation index and gridded precipitation data. Results show that pasturelands are least capable of withstanding precipitation deficits in Australia, while pasturelands in Latin America recover more slowly after drought compared to other regions. In the second chapter, I use Landsat observations to measure the magnitude, geography, and rate of change in the amount of bare ground, herbaceous and woody vegetation in the SCR since 1999. Paraguay experienced the highest proportional increase in herbaceous cover as a result of agricultural expansion and intensification, while Uruguay experienced the highest proportional increase in woody cover as a result of afforestation. Argentina, the largest and most heterogeneous country in the SCR, experienced widespread land cover changes from deforestation, reforestation, afforestation, and desertification, each of which varied in extent and magnitude by ecoregion. In the third chapter, I assess patterns of land degradation in the SCR using the United Nations Sustainable Development framework. My results show that 67.5% of the SCR experienced changes in land cover properties in the 21st century, with widespread improvement (i.e., increased productive capacity), along with substantial hotspots of degradation caused by expansion of agriculture and systematic decreases in precipitation. Monitoring degradation is necessary to assess ecosystem services, ensure food security, and develop land use policies designed to increase the resilience of land systems to the joint stresses imposed by climate change and a growing global population. The methods, datasets, and results from this dissertation provide an improved basis for creating such policies in some of the world’s most vulnerable and food insecure regions