Quantifying areal change of lacustrine environments through enhancement and mensuration of space images: Monitoring symptoms of regional climatic change.

摘要

This dissertation examines the suitability of space photographs for providing quantitative evidence of regional environmental change for the lacustrine sites of Great Salt Lake, Utah and Lake Chad, Africa. Because closed-basin lakes are sensitive to the balance between precipitation and evaporation, surface areas are directly linked to atmospheric circulation and, thus, climate. The dissertation problem is stated as: (1) Can lake-area estimates be effectively derived from digitized and registered space photographs? (2) Can surface area estimates of closed-basin lakes (as derived from space photographs) serve as direct or surrogates to inputs required by modelers for study of climatic change?;The methodology consists of selecting eight space photographs of Great Salt Lake using specific image selection criteria. These photographs are digitized, the pixels iteratively classified as land or water, and then the image is registered to a base map using tiepoints and warping routines. Nearest neighbor resampling is used during registration. Mensuration consists of scaling the images to the base map and by tabulating land-versus water pixels. Space-based measurements range from 3,657 km;Space-based measurements of Great Salt Lake are compared to USGS estimates. The mean difference between both sets of data is -2.51 percent and the standard deviation is 2.78 percent. There is no difference between the means at the 0.025 level of significance.;The dissertation methodology is extended to ten photographs of Lake Chad. Space-based lake-area measurements range from 22,772 km;Review of the literature and personal communication with climate change researchers indicates that these space-base lake-area measurements cannot serve as direct or surrogate inputs to models of climatic change. However, the dissertation technology is applicable to other areas of global change research, such as inventory and monitoring of lakes, and particularly, to the calibration of models for calculating proxy paleolake data. Paleolake levels and present-day lake levels can be calculated through use of the mass-balance equation. Space-base and model-based lake-area estimates can be compared to determine if calibration to the climate model is required.