WSP10284

• File Number: 8170
• Description: Arkansas Basin Water Quality Issues
• State: CO
• Basin: Arkansas
• Water Division: 2
• Date: 1/1/1998
• Author: USGS
• Title: Water-Quality Assessment of the Arkansas River Basin - Southeastern Colorado - 1990-93
• Water Supply Pro - Doc Type: Report/Study

cadmium concentrations. Trace-element load esti-
mates provide a general characterization of the
temporal trace-element load sources and are reported
for the seven main-stem river reaches downstream
from Leadville. Within each of the seven river reaches,
instantaneous trace-element loads were computed for
the major tributaries and for the main-stem site at the
upstream end of its respective reach. These loads were
compared to the trace-element load at the main-stem
site located at the downstream end of its respective
reach in order 10 estimate the percentage of trace-
element load contributed by the tributaries and by the
main-stem river in each of the seven reaches. Stonn
runoff was sampled only once in the upper basin.
Because storm runoff can transport substantial
amounts of trace-element-enriched sediment, stonn
loads are underrepresented in the load estimates.
Loads are not reported for the reach upstream from
Leadville because the sampling site on the East Fork
of the Arkansas River was upstream from the
Leadville Mine Drainage Tunnel, which is the single
largest trace-element source in the reach. Therefore,
trace-element loading from the East Fork of the
Arkansas River could not be accurately estimated.
For the purposes of this report, the downstream
temporal variability in trace-element concentrations in
the upper basin is described in terms of four distinct
streamflow regimes. The tour regimes include low
flow (October-March), early-snowmelt runoff (April),
snowmelt runoff (May-June), and post-snowmelt
runoff (July-September). These four regimes differ
from the three streamflow regimes described earlier in
this report. The distinction of an early-snowmelt
runoff regime is important because lower-elevation
snowmelt generally occurs during April and begins to
flush abandoned mines, mine dumps, and tailing piles
of trace-element-enriched water. Although the volume
of water that actually flows into the river from these
sources during the early-snowmelt regime is relatively
small, the effect on trace-element concentrations can
be substantial. Low-flow, snowmelt-runoff, and post-
snowmelt-runoff regimes are similar to those
described previously in the "Streamflow" section of
this report. This definition of streamflow regimes for
trace-element concentrations in the upper basin serves
as a general guideline for the interpretation of
temporal trace-element concentrations and is subject
to some degree of variability due to year-to-year vari-
ability in weather conditions and reservoir operations.

The primary source of trace elements in the
upper Arkansas River is metal-laden drainage from
abandoned mines and mine tailings (Moran and
Wentz, 1974; Wentz, 1974). Nonpoint sources of
trace-element-enriched drainage in the upper basin
probably are substantial because of the extensive
distribution of mine tailings and mine-waste piles
throughout the upper basin. However, the most
substantial sources of mine drainage are located in the
Leadville area and include the Leadville Mine
Drainage Tunnel (LMDT) and the Yak Tunnel (Moran
and Wentz, 1974; Wentz, 1974; Clements, 1991). The
LMDT discharges to the East Fork of the Arkansas
River upstream from Leadville, and the Yak Tunnel
discharges to California Gulch near Malta (pI. I).
Water-treatment plants were operating at both sites in
the spring of 1992. Both plants use a chemical-
precipitation process to remove metals from the mine
drainage before the mine-drainage water is discharged
to the receiving streams. The pre-treatment (April
1990-March 1992) and post-treatment (April 1992-
March 1993) dissolved- and total-recoverable trace-
element concentration data were compared for statis-
tical differences using the Wilcoxon rank-sum test.
Statistical testing on flow-adjusted concentrations was
not done because of the absence of any statistically
significant explanatory variables for metal concentra-
tions. The instantaneous streamflow associated with
the water samples collected during the pre-treatment
and post-treatment periods also was tested for signifi-
cant differences using the Wilcoxon rank-sum test. It
was assumed that streamflow had no statistically
significant effect on the distribution of trace-element
concentrations during the two periods because no
significant difference was detected between pre-
treatment and post-treatment instantaneous streamflow
(T.A. Cohn, U.S. Geological Survey, oral commun.,
1995). In cases where more than 50 percent of the
trace-element data at a given site were less than the
reporting level, the rank-sum test was not done.

Cadmium

The median dissolved-cadmium concentration
typically was largest at Malta and decreased substan-
tially downstream (fig. 6). Moore and R~mamoorthy
(1984) reported that dissolved-cadmium concentra-
tions in unpolluted freshwater typically range from 10
to 100 ngfL. The substantial decrease in dissolved-
cadmium concentrations between Malta and Granite

16 Wata,-Quallty A..a..ment oltha Arkan.a. Rive' Basin, Southea.tern Colorado, 199~3