GIS AND STATISTICAL ANALYSIS OF WATERSHED PHYSICAL CHARACTERISTICS, SILVERTON MINING AREA, SAN JUAN COUNTY, COLORADO

One objective of the U.S. Geological Survey (USGS) Mineral Environmental Assessment project is to develop tools for predicting the potential impacts of future mining on water quality. We are developing an empirical modeling approach, employing GIS and statistical analyses to better define the relative importance of physical watershed variables that might control acid-rock drainage. Data for analyses were acquired as part of a USGS, abandoned mine lands study conducted in the Silverton, Colo. area between 1996 and 2002. A GIS software system (the USGS, GIS Weasel), which interfaces with commands available in ArcGrid™ was used to analyze a digital terrain model and delineate a hydrologic network and corresponding catchments. Grids of alteration type, geology, and vegetation were analyzed to determine geographic intersections between 48 catchments, and corresponding hydrologic networks. These analyses permit area calculations for alteration type, geologic unit, vegetation, and alteration types adjacent to streams that were defined using a GIS buffer. Catchment characteristics (slope, aspect, stream length, drainage density) were also calculated. Statistically-based comparisons between catchments reveal potentially important information about processes controlling acid-rock drainage. For example, less-acidic to neutral catchments (pH 5.5 and higher) have median propylitic areas that are 1.5 times greater than acidic catchments (pH 5.5 and lower). This is probably due to the acid neutralizing capacity of the propylitic mineral assemblage (calcite, chlorite, and epidote). Correlations between (a) surface water geochemistry acquired for samples collected at the mouth of catchments at base flow and (b) catchment physical properties, such as areas of geologic units and alteration types were evaluated statistically. Simple linear regression, comparing pH (response) and multiple (predictor) variables indicate that the total area of quartz-sericite-pyrite (QSP) has the highest correlation when compared to all other study area alteration types and to buffered area of QSP alteration along streams. These analyses suggest that total alteration area and not just area of alteration intersecting primary streams is an important factor influencing pH and thus, metal influx to streams.