Land use land cover change and atmospheric feedback: Impact on regional water resources
Abstract
Land Use and Land Cover (LULC) change, such as conversion of natural vegetation into agricultural land and urbanization, is a major global change phenomenon. Between 1700 and 2000, the global extent of natural vegetation has decreased by 45%, and agricultural land area has increased five fold. LULC change impacts hydrology by changing regional climate (temperature and precipitation) and land surface hydrologic responses (evapotranspiration, runoff, and ground water recharge). Evapotranspiration (ET) is a major pathway through which the land surface interacts with the atmosphere, and it is a major component (60-65% of total precipitation) of the global hydrologic cycle. The three objectives of this study are to: (1) investigate impacts of regional scale LULC change on the regional hydroclimatology, and compare impacts of LULC change with impacts of climate change arising from elevated green house gas emissions, (2) evaluate uncertainties in reanalysis and climate model ET outputs using AmeriFlux observations and a basin scale water and energy balance studies, and (3) quantify contributions of major driving forces for LULC change in the United States. Large scale drainage of wetlands was carried out in the latter half of the 19th century and early half of the 20th century to bring swamp/marshy land of the Midwestern United States (presettlement landscape) into intensive agricultural production (Corn Belt of USA). Impacts of wetland drainage (LULC change) on hydroclimatology of the Midwestern United States are compared with impacts of climate change using a coupled land-atmosphere global climate model (CCSM3). The wetland drainage data are obtained from United States Census reports. Results from this study suggest that impacts of wetland drainage can be of comparable magnitude to impacts of climate change attributed to greenhouse gas emissions. The Community Land Model (CLM) and North American Regional Reanalysis (NARR) outputs are evaluated using AmeriFlux observations, PRISM precipitation and temperature data, and USGS streamflow observations in the Mississippi River Basin. Based on averages over 11 AmeriFlux sites, NARR shows higher biases (59%) in ET compared to CLM (11%). Issues related to point scale observations versus climate model grid cell outputs, and model parametrization differences between CLM and NARR are also investigated in this study. The land-cover change history of the United States is investigated to determine major driving/governing forces. County level cropland and population data from 1850 to 2000 (per decade), and high resolution topography, climate, and biophysical suitability data are used. Results from this study suggest that the spatial distribution of cropland was governed by population distribution during the 19th century, and biophysical suitability (for cropland) during the 20th century. The major influence of biophysical suitability is expected to continue in the near future landscape of the United States.
Degree
Ph.D.
Advisors
Merwade, Purdue University.
Subject Area
Hydrologic sciences|Climate Change|Water Resources Management
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.