Michigan Technological University

Department of Geological and Mining Engineering and Sciences

Research Project

Thesis:

A DIGITAL PROCESSING AND DATA COMPILATION APPROACH FOR USING REMOTE SENSING IMAGERY TO IDENTIFY GEOLOGICAL LINEAMENTS IN HARD-ROCK TERRAINS: AN APPLICATION FOR GROUNDWATER EXPLORATION IN NICARAGUA

Click here for full document (JNBruning_thesis.doc).

Click here for thesis defense presentation (defense.ppt).

ABSTRACT:

Lineament mapping using satellite remote sensing offers a low-cost, non-invasive approach for improving groundwater exploration in hard-rock terrains by identifying zones of secondary porosity in the form of fractures. A variety of lineament analysis techniques exist using remotely sensed data and have been developed in near ideal settings where influences of anthropology, vegetation, and climatic situations are minimal. Furthermore, the majority of lineament studies have been executed in regions with reputable knowledge of hydrology and geology. For these reasons there is no well-accepted or proven protocol for mapping lineaments nor have different approaches been compared in non-ideal regions. A new approach is presented here and establishes an appropriate method and identifies successful satellite imagery types for lineament analysis, resolving many deficiencies when employing traditional lineament analysis techniques in challenging settings. Four complementary satellite remote sensors, including ASTER, Landsat7 ETM+, QuickBird, and RADARSAT-1, are utilized for lineament analysis in Boaco, Nicaragua where conditions are non-ideal. A variety of image processing techniques are employed and lineament interpretations are performed on resulting image products. Lineament interpretations are synthesized using GIS, creating a coincidence raster which is used to filter lineament interpretations. Results show that this method to filter and evaluate lineament interpretations is an appropriate alternative to traditional orientation-based lineament filters for this setting. The coincidence raster correlates well with the few faults mapped in the study area and with ground-based lineament observations. Additionally, results from manual pumping tests suggest that as well productivity can be linked to proximity to lineament features as interpreted from the coincidence raster. Lineament interpretation results show that interpretations from RADARSAT-1 products are superior over interpretations from other sensor products. This suggests that quality lineament interpretation in this region requires anthropogenic features to be minimized and topographic expressions to be maximized. However, results also indicate that utilizing RADARSAT-1 imagery alone for lineament interpretations may miss lineaments. For this reason, it is suggested that image products derived from both radar and optical sensors along with a DEM are employed to generate a coincidence raster from which a final lineament map can be drawn with confidence. The presented approach has the potential to improve groundwater exploration and well sighting in non-ideal regions.

Committee Members:

Dr. John Gierke (adviser)

Dr. Ann Maclean

Dr. Debbie Huntzinger