ASTER digital image processing for geological mapping: Examples from Neoproterozoic Allaqi-Heiani Suture, Egypt

This dissertation constitutes three manuscripts summarizing efforts in developing effective algorithms for the analysis of the Advanced Space-borne Thermal Emission and Reflection Radiometer (ASTER) data for geological mapping using the Neoproterozoic Allagi-Heiani Suture (AHS) in southern Egypt as a test site.; The first manuscript presents statistical approach for the selection of ASTER bands to be used for Red-Green-Blue (RGB) color combination images. The Optimum Index Factor (OIF) and determinants techniques are proposed to rank all possible ASTER RGB color combinations for the western part of AHS. ASTER has three subsystems including Visible and Near Infra-Red (VNIR), Short Wave Infra-Red (SWIR), and Thermal Infra-Red (TIR). Hence, ASTER bands can be used in seven Electromagnetic Spectrum (EMS) regions combinations. These are VNIR only, SWIR only, TIR only, VNIR+SWIR, VNIR+TIR, SWIR+TIR, and VNIR+SWIR+TIR. The OIF and determinants results agree in four of the seven EMS regions combinations (VNIR only, SWIR only, TIR only, and VNIR+SWIR), but differ in three (VNIR+TIR, SWIR+TIR, and VNIR+SWIR+TIR).; The second manuscript outlines an effective algorithm to spectrally separate ophiolite components in AHS and use these as markers to trace the continuity of geological structures. Principal component analysis (PCA), Fast Fourier Transform (FFT), and Redundant Wavelet Transform (RWT) are used to identify ophiolite components as well as Neoproterozoic ductile structures to trace along-strike continuation in AHS. This work helps trace structures along AHS and reveals structures that have not been identified before. Nappes identified in the west are traced into the central part of the suture allowing for subsequent modification by younger structures.; The third manuscript outlines a methodology to quantitatively evaluate ASTER band-ratios that can be effectively used for thematic classification aimed at separating distinctive lithologies in AHS. The Within-Class Homogeneity Criteria (WHC) is proposed for the selection of specific band-ratios. The effectiveness of WHC is tested by using the classification correction evaluation curve to select fewer band-ratios. The accuracy of WHC is tested by comparing previously published geological map covering AHS with the WHC-based classification results and those based on all 14 ASTER bands and all 182 ASTER band-ratios. Commission and emission errors frequently occurred during traditional thematic classification are conquered by using WHC-based classification.