A GIS-based integrated approach to explore land-use/cover change dynamics in south-central Indiana.

摘要

Changes in land use and land cover especially forest cover contribute to broad-scale ecological changes and many critical environmental alterations. The general research questions that this dissertation addresses are: (1) What Land-Use/Cover Change (rate, direction, and spatial pattern) occurred in Monroe County Indiana 1984--2004? (2) What are the drivers of land-cover change and do the drivers differ by scale of analysis? (3) How do spatially heterogeneous local decision-makers and their spatial interactions contribute to macro-scale landscape outcomes? In this dissertation, three complementary methods are employed to investigate LUCC from different perspectives in the main Chapters (3--5). Chapter 6 links these approaches to form an integrated analytical framework. Specifically, Chapter 3 focuses on an empirically-based spatial analysis of major LUCC dynamics at multiple scales in Monroe County, Indiana between 1984 and 2004. Chapter 4 builds a Cellular Automata (CA) model which is compared with two Agent-Based models for the same study area of Indian Creek Township within Monroe County. Chapter 5 evaluates key dynamics of decision-making related to land use in a spatially explicit context through a set of spatial decision-making lab experiments. This dissertation demonstrates how various spatially-explicit data can be linked to examine individual land management decisions and the resulting land-cover outcomes at local to regional levels. It also explores the trade-offs as well as the predictive accuracy of three separate models. This research contributes to the emerging land change science by examining LUCC dynamics from the perspective of different methodological approaches. The findings contribute to the understanding of the process of reforestation in south-central Indiana, a study area that has undergone changes in land cover similar to much of the Eastern United States. Identified comparative advantages and disadvantages of different modeling options facilitate the selection of proper models in different applications. And the innovative spatially explicit lab experiments offer a cost-effective way to investigate real human decision-making related to land use under well-controlled settings. Improved understanding of critical LUCC dynamics facilitated by these complementary approaches will lead to more informed landscape management policies that are essential for long-term sustainable development.