Hydrology and glacier-lake-outburst floods (1987-2004) and water quality (1998-2003) of the Taku River near Juneau, Alaska

Abstract

The Taku River Basin originates in British Columbia, Canada, and drains an area of 6,600 square miles at the U.S. Geological Survey's Taku River gaging station. Several mines operated within the basin prior to 1957, and mineral exploration has resumed signaling potential for future mining developments. The U.S. Geological Survey in cooperation with the Douglas Indian Association, Alaska Department of Environmental Conservation, and the U.S. Environmental Protection Agency conducted a water-quality and flood-hydrology study of the Taku River. Water-quality sampling of the Taku River from 1998 through 2003 established a baseline for assessing potential effects of future mining operations on water quality. The annual mean discharge of the Taku River is 13,700 cubic feet per second. The monthly mean discharge ranges from a minimum of 1,940 cubic feet per second in February to a maximum of 34,400 cubic feet per second in June. Nearly 90 percent of the annual discharge is from May through November. The highest spring discharges are sourced primarily from snowmelt and moderate discharges are sustained throughout the summer by glacial meltwaters. An ice cover usually forms over the Taku River in December persisting through the winter into March and occasionally into April. Glacier-lake-outburst floods originating from two glacier-dammed lakes along the margin of the Tulsequah Glacier in British Columbia, Canada, are the source of the greatest peak discharges on the Taku River. The largest flood during the period of record was 128,000 cubic feet per second on June 25, 2004, resulting from an outburst of Lake No Lake. Lake No Lake is the larger of the two lakes. The outburst-flood contribution to peak discharge was 80,000 cubic feet per second. The volume discharged from Lake No Lake is relatively consistent indicating drainage may be triggered when the lake reaches a critical stage. This suggests prediction of the timing of these outburst floods might be possible if lake-stage data were available. Further increases in the volume of Lake No Lake are unlikely as all tributary glaciers have retreated out of the lake basin. Decreasing outburst-flood volumes from Tulsequah Lake suggests a continued decline in the volume of this lake. Physical and chemical parameters and concentrations of basic water-quality constituents indicate good water quality. Samples collected at the Taku River gaging station contained low concentrations of trace elements in the dissolved phase. Trace elements sampled were within acceptable limits when compared with the Alaska Department of Environmental Conservation aquatic-life criteria for fresh waters. The highest concentrations of total trace elements sampled were collected during glacial-outburst floods and likely are associated with suspended sediments. Total trace-element concentrations generally increase with increasing water discharge, although a high correlation for all constituents sampled does not always exist.