Analytical data in this report provided the basis for interpretation of the loading, partitioning, and transport of metals from source areas upstream from Silverton, Colo. to Aztec, New Mexico under low-flow conditions. The synoptic sampling reported in this section took place at low flow in October, when bed sediments were not being transported thus, the focus of transport in this report is on the dissolved and colloidal loads. It is important to realize that at high flow, during storms, or snowmelt runoff, there will be large amounts of metal transported with movement of the bed sediments. On an annual basis, the amount of metals transported with the bed sediments could greatly exceed the annual fluxes measured for the dissolved and colloidal loads.
Site identifications, field measurements, and major ion chemistry of samples collected for water and colloids are in appendix III. Results of chemical determinations for dissolved and colloidal metal concentrations are in appendix IV. In both appendices, data are sorted in downstream order in for the Animas River (mainstem) and tributaries (Mineral and Cement Creeks) to emphasize the changes downstream. These analyses were compared to water-quality standards for aquatic life for Colorado given in table 1 (CDPHE, written commun., 1997).
In the headwater stream reaches upstream from Silverton, aluminum coatings (chalk-white precipitates) seen on the rocks on the sides of the stream channel indicate that aluminum is partially removed from the streams before entering the Animas Canyon. Concentrations of dissolved aluminum ranged from a high of 0.42 mg/L at site A-72, downstream from Silverton, to an average of 0.09 mg/L downstream from the Animas Canyon. Thus, not all the aluminum was removed as coatings before entering the Animas River. The dissolved aluminum concentrations in the Silverton area are in excess of the water quality standard for aquatic life, but below this standard downstream from the Animas Canyon. However, much of the aluminum in the Animas River downstream from Mineral Creek is colloidal (appendix IV). Recent work has noted that colloidal aluminum may be toxic to fish whereas dissolved aluminum may not be (Witters and others, 1996). In fact, Witters and others (1996) found that freshly formed aluminum-hydroxide precipitates may be most toxic to trout. This would be the dominant form of aluminum in the waters within the Animas Canyon reach.
________________________________________________________________________________________ | ||||
|---|---|---|---|---|
| Metal | Aquatic Life Standard1 | Agricultural2 | Domestic2 | |
| Acute | Chronic | Standard | Standard | |
________________________________________________________________________________________ | ||||
Aluminum |
750 |
87 |
|
|
Arsenic |
360 |
150 |
100 |
50 |
Cadmium |
e(1.128[ln(hardness)] - 2.905) Trout = e(1.128[ln(hardness)] - 3.828) |
e(0.7852[ln(hardness)] -3.490) |
10 |
5 |
Chromium+3 |
e(0.818[ln(hardness)] - 3.688) |
e(0.818[ln(hardness)] - 1.561) |
100 |
50 |
Chromium+5 |
16 |
11 |
100 |
50 |
Copper |
1/2e(0.9422[ln(hardness)] - 0.7703) |
e(0.8545[ln(hardness)] - 1.465) |
200 |
1,000 |
| Iron | 1,000 (total recovery) |
300dis | ||
| Lead | 1/2e(1.6148[ln(hardness)] - 2.1805) |
e(1.417[ln(hardness)] - 5.167) |
100 |
50 |
Manganese |
1,000 |
200 |
50dis | |
Mercury |
2.4 |
0.1 |
2 | |
Nickel |
1/2e(0.76[ln(hardness)] + 4.02) |
1/2e(0.76[ln(hardness)] + 1.06) |
200 |
100 |
Selenium |
20 |
5 |
20 |
50 |
Silver |
1/2e(1.72[ln(hardness)] - 6.52) |
1/2e(1.72[ln(hardness)] - 9.06) Trout = e(1.72[ln(hardness)] - 10.51) |
|
100 |
Zinc |
e(0.8473[ln(hardness)] + 0.8604) |
e(0.8473[ln(hardness)] + 0.7614) |
2,000 |
5,000 |
_______________________________________________________________________________________
1 Metals for aquatic life standards are dissolved concentrations are given in mg/L. Water hardness to be usedConcentrations of cadmium were below the limit of detection (<0.001 mg/L) in most of the water samples. Values downstream from Cement Creek and Mineral Creek were measurable where the pH was lower. Cadmium concentrations were not elevated in water or colloids from sites downstream from Silverton during low flow and, at these low concentrations, cadmium did not exceed any water-quality standards.
In samples from the upper Animas River, dissolved copper concentrations averaged 0.003 mg/L and colloidal copper concentrations averaged 0.006 mg/L. These values are near the lower detection limit for copper (<0.001 mg/L). The highest concentrations of dissolved copper were in Cement Creek and Mineral Creek. Due to a pH of 6.35 and a substantial source of copper in its headwaters, Mineral Creek had the highest concentration of colloidal copper at 0.055 mg/L. Cement Creek, on the other hand, had the highest concentration of dissolved copper at 0.06 mg/L. Sites downstream from the Animas River-Cement Creek confluence and the Animas River-Mineral Creek confluence had the highest colloidal concentrations of copper at 0.016 mg/L and 0.022 mg/L respectively. Downstream from the Animas Canyon at the KOA Campground site (72.5 km), there was a relatively high colloidal copper concentration of 0.006 mg/L, but all the remaining sites in the basin had concentrations less than 0.004 mg/L. Concentrations at the majority of sites were less than the concentration in the field equipment blank (appendices III and IV), so it is not possible to make distinctions among them. One relatively high concentration of dissolved copper, 0.008 mg/L, was observed at the 32nd Street Bridge in Durango. This must have been from some unidentified source near Durango. The acute water-quality standard for dissolved copper was exceeded only in Cement Creek. However, colloidal copper concentrations exceeded the water-quality standard for chronic toxicity at most other sites. It is not clear what aquatic-life standards are applicable to the colloids, but these copper concentrations could cause chronic toxicity if copper in the colloids were available to aquatic life.
Iron is the main component of the colloidal material that forms in the streams upstream from Silverton. Concentrations of colloidal iron were measured in all samples collected all the way downstream to Aztec. The pattern was similar to that observed downstream from mining sources in the Arkansas Basin (Kimball and others, 1995) both rivers show a large increase in the colloidal iron concentration immediately downstream from the source and then a gradual decrease downstream. In the Animas River, the highest concentrations of dissolved iron were 0.64 mg/L downstream from the confluence of Cement Creek and 0.68 mg/L downstream from the confluence of Mineral Creek. Downstream from the Animas Canyon, only a few dissolved iron concentrations were above the limit of detection (<0.001 mg/L), with the greatest concentration at the KOA Campground site (0.012 mg/L) just downstream from the Animas Canyon.
There is an aquatic life standard for total recoverable iron of 1 mg/L. The colloidal concentrations reported here from the sites downstream from Cement Creek and Mineral Creek exceed this water-quality standard. Although iron does not cause acute toxicity, it does influence habitat and chronic toxicity because of its deposition on the substrate of the stream bed and incorporation into the food chain. Witters and others (1996) suggest that colloidal iron may be dissolved in fish guts releasing the sorbed metals.
As observed in many other streams (Kimball and others, 1995), manganese in the Animas River resided principally in the dissolved phase rather than the colloidal phase. The pattern of dissolved manganese was very similar to the pattern of iron the highest concentrations were downstream from the confluences of Cement and Mineral Creeks, and then the concentrations decreased downstream. The highest tributary concentration of dissolved manganese was from Cement Creek, with 1.3 mg/L. The dissolved concentrations in the Silverton area were almost at the chronic water-quality standard for manganese however, there is no manganese standard for acute toxicity.
Although strontium is not a toxic metal, it is useful to contrast its distribution pattern with other metals. Almost all of the measured strontium was in solution rather than in the colloidal phase. Strontium concentrations were high in the Animas River near Silverton, decreased downstream and then increased again due to the high inflow concentrations downstream from the Animas Canyon. The highest tributary concentration was in Cement Creek, at 2.6 mg/L, which may reflect the leaching of carbonate minerals (calcite, CaCO3, and rhodochrosite, MnCO3) present in the Eureka graben vein-type ore. Strontium will substitute for manganese in the rhodochrosite structure. The pattern of concentration downstream differed from that of the mining-related metals because there were substantial sources of strontium downstream from the Cretaceous Picture Cliffs, Kirtland, and Fruitland Formations. The next highest inflow concentrations were from Basin, Hermosa, and Lightner Creeks near Durango. Thus, the strontium data illustrate the pattern of an element that does not originate from the headwaters near Silverton.
Like manganese, zinc occurred principally in the dissolved phase. The highest dissolved concentrations were near Silverton. Cement Creek had a dissolved zinc concentration of 1.3 mg/L, the highest of any tributary, and the measured pH was 3.89. Unlike some other metals, the dissolved concentration of zinc at Eureka (pH 7.06), upstream from Silverton, was high at 0.46 mg/L. Downstream from Silverton, the concentrations of zinc consistently decreased as pH increased from 6.62 to 7.87 in the Animas Canyon reach. Downstream from the Animas Canyon, the dissolved zinc concentration in the Animas River decreased substantially from that upstream but did not decrease to levels that were comparable to the colloidal zinc concentration until near Bondad Hill where the pH increased to about 8.6.
The concentrations of dissolved zinc exceeded water-quality standards for both chronic and acute toxicity at each of the sites upstream from the Animas Canyon (fig. 10). Concentrations of colloidal zinc in these reaches were below aquatic-life standards, but if sorbed zinc were released in the guts of fish as suggested by the work of Witters and others (1996), colloidal zinc could contribute to chronic zinc toxicity.
Samples were analyzed for arsenic, mercury, and selenium to evaluate the dissolved and colloidal loads. None of the samples had concentrations of these metals above the detection limits. The detection limits were 1 microgram per liter (mg/L) for arsenic and selenium, and 0.1 mg/L for mercury. The lack of detection in water and colloids emphasizes the tendency of these elements to be associated with the bed sediments.