USING SATELLITE IMAGERY AS A TOOL TO ENHANCE GEOLOGICAL MAPPING

Geological mapping is generally conducted seasonally. Maps are completed by using ground surveys, along with aerial photography, satellite imagery, Lidar and other data sources. These are all used as references by geologists, to better understand the geology of the study area. Unfortunately, reference materials, such as aerial photography and previously published geological maps may be old and out of date. Many natural and manmade events, such as debris flows, environmental disasters, housing projects, watershed projects and road development can affect the landscape, invalidating many of the traditional reference locations. In addition, geological maps are often made with constraints of time and resources, so that geological features inferred from imagery cannot be adequately checked in the field. Mapping accuracy can also be restricted by the ability to physically survey the difficult to reach areas due to safety concerns.

The focus of this project is to research the viability of using satellite imagery, varying in spatial resolution, to help geologists complete their mapping, as well as adding to the accuracy of the mapping. The Jones Hill and Marmot Peak quadrangles in Colorado were used as a test case. The initial results of using LandSat 8 imagery at a 30 meter resolution and a level 1 terrain correction and running a classification of 14 categories on the quadrangles show promise for augmenting and improving mapping, even with minimally processed imagery (using either ArcGIS or ENVI) in the area of interest. The results were compared to recently published digital geological maps of the quadrangles. The maps were produced by doing ground surveys and recording geological features on aerial photos. Features were transferred from the aerial photos to topographic base maps with Stereo Analyst for ArcGIS. The results obtained from analysis of the LandSat 8 imagery show a strong, positive correlation with geological features shown on the published maps.