Objective: Determine the extent to which remote-sensing techniques, primarily lineament analysis, can be used to identify geologic structures that may affect the location and migration of subsidence over abandoned underground coal mines in the Colorado Springs, CO, area. Compile data, using a geographic information system (GIS) to allow efficient and reproducible comparisons and analyses. Background: The basis for evaluating abandoned mine land (AML) sites with remote-sensing techniques comes from past and ongoing Bureau ground control research using remote-sensing data to identify potential hazards caused by geologic structures (faults, joints, etc.) above active underground coal mines. Research showed that lineament (linear feature) analysis of Landsat Multispectral Scanner (MSS) images could indicate locations of fractures and fracture trends that had varying, but significant, impacts on the stability of mine openings. Such geologic structures should be identifiable over abandoned mines as well. If fractures are present, even if they are not faults, they should represent areas of potential instability over these mines and, therefore, should tend to localize subsidence when it occurs, particularly for deep and moderately deep portions of mines. The general study area used in this research was the northern and northeastern Colorado Springs, El Paso County, CO, area, where coal mining was active from the 1870's until 1957. Mines range in depth from 0 ft (shaft-slope openings) to approximately 500 ft. This area has a lengthy history of subsidence events, is known to have scattered active subsidence, and presents subsidence hazards for existing and planned urban and suburban development. The locations and approximate extents of most abandoned mines are known. A subsite at the Colorado Springs Country Club (in the Cragmor-Country Club area of Colorado Springs) was of specific interest in this investigation because a previous AML project on passive seismic monitoring of subsidence activity had identified subsurface seismic events apparently related to active caving above one of the mines. Although the site of this previous project was quite small, the availability of known seismic events would allow some comparison with lineament locations. Therefore, the seismic events might provide some subsurface definition of the impact of geologic structures on active subsidence in the area. The study area is heavily developed by urban and suburban buildings and other surface facilities (golf courses, relocated or modified drainages, roads, etc.). However, the cultural overprinting of the study area made field checking difficult or impossible for most of the area. As a result, much of the data analysis involved in this study was done using data from previous subsidence studies. Approach: Geographic information system (GIS) technology affords the means to overlay, integrate, and analyze diverse data sets that are referenced to the same geographic area, as in the case of .the Colorado Springs study area. The Bureau uses the ARC/INFO software package for GIS work. In the GIS, map data can be used in any analysis or can be related to any other data set that geographically overlaps. Data stored in a GIS are not tied to a particular scaling factor (e.g., 1 in = 200 ft), as hard-copy maps are, but are referenced only to a standard geographic coordinate system. All data and maps to be compared or otherwise analyzed must be in the same coordinate system, or the data sets will not coincide with respect to their coordinates. Because these data sets all are tied to the same geographic coordinate system (Universal Transverse Mercator for this study), any comparison or map overlay will accurately represent the true locations of features within the limits of the accuracy of the original data. This allows for conclusions on whether modifications should be made to existing subsidence risk predictions or on whether further site-specific investigations should be pursued based on the data available. Data Available: The ARC/INFO GIS data base for the Colorado Springs area includes data from the following sources: 1. Maps of (1) workings of most abandoned underground coal mines, (2) subsidence features (sinkholes and troughs), (3) projected subsidence risk for the undermined areas, (4) overburden thickness. 2. Locations of (1) shallow drill holes from engineering studies of the area, (2) fracture orientation measurements (values) collected during this research, and (3) geophone arrays and locatable seismic events for the Colorado Springs Country Club AML study site. 3. A U.S. Geological Survey surficial geologic map of the area. 4. Digital elevation models (DEM data) for the four 75-min topographic quadrangles covering the study area. 5. Lineaments interpreted from Landsat MSS and Thematic Mapper (TM) images and 1947-vintage black-and-white aerial photographs of the study area, including control points used to rectify the lineaments to topographic maps. 6. The map index system of Dames & Moore, the company that did the original subsidence risk study. These data sets all are in the form of point, line, or polygon "coverages," which are linked locational files (i.e., maps) and attribute files (e.g., ID numbers and fracture orientations). The data base also contains ARC/INFO "lattice" and/or triangulated irregular network (TIN) files of XYZ values for topography, overburden thickness, and coal seam elevation data. "Map compositions" (plan, three-dimensional, and other line drawings) and plot files created for use in presentations and in plotting hard copies of the figures also have been included in the data base to allow replication of project report figures and their use in demonstrations of the GIS and remote-sensing technology. The original ARC/INFO GIS data base is approximately 75 Mbyte. The total size of the ARC/INFO EXPORT-format data base is approximately 100 Mbyte. The total size of the DXF-format data base (Drawing Interchange File) is approximately 110 Mbyte. Data Formats: Because the data base was created using the ARC/INFO GIS (version 5.0.1), all data sets and maps are available as export files from that format. The export files should be compatible with any ARC/INFO software of that revision or newer on any hardware system. The data sets and maps also have been converted to DXF-format files that can be read into most CAD systems. The plot files and map compositions cannot be converted to DXF format, so these files are not included in the DXF-format data base.