Date of Completion

6-26-2019

Embargo Period

6-24-2022

Keywords

Climate change; Connecticut River Basin; Land use/cover change; Quantitative assessment; Non-stationarity; Future scenarios

Major Advisor

Scott R. Stephenson

Associate Advisor

Chuanrong Zhang

Associate Advisor

William B. Ouimet

Associate Advisor

Xiusheng Yang

Field of Study

Geography

Degree

Doctor of Philosophy

Open Access

Open Access

Abstract

The connections between environmental change and human activities are complex. Scientists have been working on understanding the interactions between hydrological processes, land use/cover change (LUCC) and climate change in both qualitative and quantitative ways for several decades. Although previous studies show that interactions between these three aspects are typically multidimensional and occur in multiple spatial and temporal scales, a systematic investigation of their historical and future relationships is still lacking at a local scale, especially when considering the non-stationarity of LUCC. This doctoral dissertation applies quantitative research methods, such as hydrological and LUCC modeling methods, to cover two general study directions: (1) how human activities (e.g., LUCC), climate change and hydrological processes interact with each other, and (2) how to analyze these interactions when taking local spatial variance into consideration. To follow these directions, this research includes three main sections: First, by integrating a new elasticity of runoff method and a water balance model, I separate and quantify the impacts of climate change and LUCC on increasing surface runoff change in the lower Connecticut River Basin. Inverse variation trends of LUCC on opposite sides of the river is found in this section, giving us motivation to hypothesize that human activity could influence our landscape to varying degrees in different locations. Second, I identify spatially non-stationary relationships between driving factors and land use/cover categories at a local scale by applying geographically weighted logistic regression model. Sensitivity of simulated LUCC to spatial non-stationarity is then examined. Third, based on the previous conclusions, I simulate the streamflow change in a small basin under future LUCC and various climate change scenarios, and ultimately quantify the relationship of change rate between streamflow and climate variables in the future.

COinS