THE APPLICATION OF GEOSTATISTICS TO EVALUATE BASELINE WATER-QUALITY DATA, UPPER ANIMAS RIVER WATERSHED, SILVERTON, COLORADO

Describing premining sources of metals is an important aspect needed to establish realistic remediation goals for mine-land reclamation. In the Upper Animas River watershed, southwestern Colorado, determination of premining conditions is complicated by the temporal and spatial overlap of mineralization, hydrothermal alteration, and fracturing within the San Juan and Silverton calderas, where mining activities were centered. To evaluate premining metal sources, two geostatistical methods (probability plots and principal component analysis-PCA) were used to analyze a large geochemical data set collected as part of the USGS Abandoned Mine Lands Initiative. These statistical techniques are useful for simplifying large data sets to a few key groups or variables. The data set included water samples collected from 241 spring and stream sites and 75 mine sites that were analyzed for 26 chemical constituents. For the spring and stream sites, a ranking system, primarily based on field observations, was devised to evaluate the potential for mining activity effects. The ranking system consisted of four categories ranging from category I (no evidence of mining activity) to category IV (direct discharges from mine sites). Ranges and medians for sites unaffected by mining (category I and II) were: pH 2.58-8.49, 4.89; SO₄ 1-1300, 90 mg/L; and Zn <20-14,300, 28 mg/L; and for mine waters: pH 2.35-7.77, 5.72; SO₄ 45-2720, 310 mg/L; and Zn <20-228,000, 620 mg/L. Sulfate and zinc probability plots, using the complete data set, contain 8 to 10 nodes, consistent with multiple sources of these constituents. Results from PCA demonstrate that category I-II scores plot in fields defined by the intensity of hydrothermal alteration. Mine sites tend to overlap fields defined by more intense alteration or proximity to mineralized veins. Overall, the statistical methods demonstrate that unaffected sites exhibit a wide range in water quality, primarily controlled by bedrock alteration. The mine waters also have a wide range in water quality, likely resulting from three factors: chemistry of water entering mines varies because of weathering of different alteration assemblages, weathering of different ore and gangue minerals depending on mineral deposit type, and some of the sites classified as mining-affected were metal-poor prospects.