APPENDIX I--METHODS USED TO PREPARE REGIONAL SEDIMENT GEOCHEMICAL MAPS

         Data were retrieved from the National Uranium Resource Evaluation (NURE) data base (Hoffman and Buttleman, 1994) covering a rectangular area of southwestern Colorado between 37o 00' and 38o 30' north latitude and 107o 00' and 109o 00' west longitude. The evaluation boundary extended outside of the watershed area to eliminate "edge effects" in the grid and contour generation. The total number of sample sites in the bounding area is 4,057, 444 of which are in the Animas River watershed. The NURE samples generally were collected along tributary drainage systems (Shannon, 1980 Warren and others, 1981).

         The ASCII data files for copper, lead, zinc, and calcium concentrations were evaluated and formatted for grid and contour generation in EarthVision, a software package developed by Dynamic Graphics, Inc (ERDAS). Each chemical constituent was processed separately to evaluate the relative quality of the data and to compare with individual point coverages subsequently generated in ARC/INFO. The intervals generated for the contoured grid surface reflect multiples of the CAV (Fortescue, 1992). The multiples chosen for analysis and display are 1, 3, 5, 10, and 20 times crustal abundance values (CAV figs. 5-8). The procedure in EarthVision involved a few basic steps:

         1) defining raster input headers
         2) registering and transforming geographic coordinates
         3) creating a 2-D minimum-tension grid with 1,000 meter grid spacing
         4) visualizing and generating contour intervals based on crustal abundance values (CAV)
         5) processing grid files in ERDAS including: flipping, resampling (100m), and converting the data into a
                  vector format
         6) exporting grid files to ARC/INFO for vector processing.

The contoured surfaces were transferred to the vector-base ARC/INFO software package to allow overlay and integration with other spatial themes in the Animas River watershed. ARC/INFO processing steps included:
         1) generating NURE point locations and linking attribute information
         2) clipping point and contour coverages with the Animas River watershed boundary
         3) conversion and smoothing of contours vectorized in ERDAS
         4) editing coverages and creating a relational data base
         5) overlay analysis and map generation.

In addition to the tributary NURE information, sediment samples from major river drainages were collected by USGS personnel in the 1995-96 field season. These samples supplement the regional geochemical analysis of the NURE samples and provide current site-specific information along the Animas River. These data were processed in ARC/INFO using the procedures described above. The Animas River and its major tributaries were extracted from a hydrology layer and buffered (1000 meters) to create a polygon for shading based upon copper, lead, and zinc concentrations. This polygon or "ribbon" was segmented based on the concentration multiples (1,3,5,10, and 20) of CAV. The "ribbons" of concentration are shown in figures 19-21.