Technical Service Center
Environmental Applications and Research Group — Publications
Monitoring of Heavy Metal Concentrations in the Arkansas River Using Transplanted Aquatic Bryophytes
Technical Memorandum No. 8220-96-18
S. Mark Nelson
The upper Arkansas River has long been impacted by metals contamination from mining areas. The effects of poor water quality related to the ecology and distribution of aquatic macroinvertebrates in the Arkansas River has been studied for a number of years (LaBounty et al., 1975; Roline, 1988; Kiffney and Clements, 1994). Many of the pollution sources, such as the Leadville Mine Drainage Tunnel (LMDT) and California Gulch, have been treated recently (since 1992) and water chemistry along with biota (Nelson and Roline, in press) has been substantially improved. As a supplement to the more standard biomonitoring programs which use aquatic macroinvertebrates and, in some cases, fishes, we used metal uptake by aquatic bryophytes (moss) to examine relative concentrations of metals in the system. Bryophyte tissue may be a valuable tool for demonstrating differences in water quality because of the ability of bryophytes to accumulate and concentrate metals. Transplanting a known species of bryophyte for a given period of time may also allow for comparisons between sites and provide some diagnostic capability.
The purpose of this study was to identify distribution of several trace metal contaminants using transplanted aquatic bryophytes. We also compared metal concentrations found at sites on the Arkansas River to others in the southern Rocky Mountains to relate the recovery of the Arkansas River to a larger data base.
Five study sites were selected, two of which were on the East Fork of the Arkansas River at sites above (EF-01) and below (EF-03) the inflow of the LMDT. Other sites were below the confluence of the East Fork with Tennessee Creek (AR-01) and further down-river, below the town of Buena Vista, sites above (ACC) and below (BCC) the confluence of Chalk Creek with the Arkansas River. Nate Creek Ditch (NCD) was the donor site for bryophytes and is part of the Uncompahgre River drainage in western Colorado.
Bryophytes (Hygrohypnum ochraceum) were collected from Nate Creek Ditch during July of 1996. Bryophytes were vigorously washed by hand using tap water to remove sediment and attached invertebrates. Deionized water was used as a final rinse. Bryophyte samples of about 1`0-g spin-dried weight were placed into nylon mesh bags (mesh size 4 mm) and transplanted to the various monitoring sites. Bags were tethered to bricks that were kept in place with river rock. Three randomly-selected replicate bags were placed at different locations within each of the sites. This experimental design was similar to the comparative mensurative experiment described by Hurlbert (1984) used to determine whether there were significant differences among metals taken up at different sites. Three replicate samples from NCD were analyzed for baseline metal concentrations. Transplanted samples were left in place for 19 days from 25 July 1996 to 13 August 1996, then transported to the laboratory in polyethylene bags, washed again as described above, and analyzed for metals.
Bryophyte samples for metals analyses were oven-dried at 60'C for 48-hr and then acid digested. Inductively coupled plasma spectroscopy (ICP) was used to analyze bryophyte tissue for cadmium (Cd), copper (Cu), lead (Pb), and zinc (Zn). Analytical accuracy of metal analysis was determined by using certified reference bryophyte material (National Institute of Standards and Technology, SRM 8031).
The sequential Bonferroni technique (Rice, 1989) was used with ANOVA to test for significance at the table-wide level (P<0.0125). Metal concentrations in bryophyte tissue at the different stations which differed significantly at the table-wide level were then analyzed using the multiple comparison Tukey's test to compare all stations to each other for the given parameter.
Water samples for total metals analyses at the five stations were collected at the beginning and end of the bryophyte exposure period. Samples were collected in polyethylene bottles and preserved with concentrated HNO3. Metal determinations were made by ICP.
Comparison of Arkansas River Sites With Others
Cluster analysis of site mean normalized (log base 10) data was performed on bryophyte material from several different drainages in the southern Rocky Mountains of Colorado and New Mexico. Drainages included in the data were the Uncompahgre (Nelson and Campbell, 1995), Rio Grande (Porter and Carter, unpublished data), Clear Creek (Porter, unpublished data), and tributaries to the Arkansas River (Nelson and Campbell, unpublished data). Cluster analysis was used to separate sites according to concentrations of Cd, Cu, Pb, and Zn. Methods used for bryophyte preparation for these data were similar to those used in the present study: bryophytes were transplanted in mesh bags, Hygrohypnum ochraceum from NCD was used, and cleaning and drying techniques were similar. Exposure times, however, did vary. Although most samples were exposed for 20 days, some were only exposed for 10 days while others went as long as 84 days. Comparison of sites with different exposure times may be justified because metal uptake is very rapid and exposure times past the first few days do not show a large increase in concentration (Porter and Nelson, 1995). Of course this is assuming a somewhat constant water concentration and does not take into consideration intermittent pollution events that may increase in probability as exposure time increases.
Water chemistry results (Table 1) indicated that, with the exception of a single Cu value at ACC, only Zn was detected in water using ICP. Zinc concentrations generally increased downstream to the confluence of Chalk Creek and the Arkansas River.
Arkansas River Bryophytes
Laboratory accuracy determined from comparison with certified values of Cd, Cu, Pb, and Zn ranged from 84.0 to 101.6% suggesting suitable data quality. Average metal concentrations from Arkansas River locations and Tukey's test results are summarized in Table 2. Metals uptake at the different sites were statistically different (P<0.0005) at the table-wide level. The site furthest upstream on the Arkansas River (EF-01) did not differ statistically in the amount of metals relative to baseline (NCD) concentrations with the exception of Cd which was higher at EF-01. EF-03 which was downstream of the LMDT, a point-source treated since 1992, was similar to EF-01, except for significantly higher concentrations of Zn. Metal concentrations in the LMDT concentrations in the LMDT concentrations in the LMDT effluent meet NPDES permit requirements, but result in higher Zn concentrations in the East Fork of the Arkansas River at EF- 03. It is also possible that metal sources are present in the river or wetlands between the LMDT and EF-03.
Cd continued to increase significantly at stations downstream of the East Fork and Zn was especially high at stations above and below the confluence of Chalk Creek with the Arkansas River. Highest concentrations of all metals were found at these sites. These data suggests that other metal sources besides the LMDT continue to find their way into the Arkansas River. Other potential sources include St. Kevins gulch above AR-01 and California Gulch at some distance above the Chalk Creek sites. It is also possible that metals remain sequestered but occasionally available within the river channel. Chalk Creek did not appear to have any statistically significant impact on metal uptake in bryophytes located above and below the confluence in the Arkansas River (Table 2). Bryophyte tissue data suggests that there are additional Zn sources above EF-03 and ACC; Cd sources above EF-01, AR-01, and ACC; and a Cu source above ACC.
Comparison of Arkansas River Sites With Others in the Southern Rockies
Cluster analysis of Southern Rocky Mountain bryophyte data resulted in Figure 1. The clusters suggested that there were at least three and perhaps four main separations of data ranging from baseline or non-contaminated to severely contaminated (Figure 1). Groupings contained sites that differed in geology, altitude, stream order, and water quality parameters such as alkalinity and major ions. This suggests that differences in bryophyte metal concentrations are largely related to levels of metal contamination, rather than being caused by a natural longitudinal change affecting the bryophyte's ability to take up metals.
Concentrations of metals (mean ± standard error) determined from each of these classification groups are shown in Table 3. Metal concentrations from baseline sites were similar to those reported in reference site mosses from studies in other countries (Goncalves et al., 1992). Comparison of Arkansas River metal concentrations (Table 2) with other sites (Table 3) indicates that all of the Arkansas River sites have high Cd and Pb concentrations. Pb concentrations may be an anomaly in the Arkansas River analysis because concentrations at NCD were higher than had been observed in the past. If Pb concentrations are discounted, then EF-01, with the exception of Cd, would be placed in the non-contaminated or baseline category. Other Arkansas River sites would fall into the slightly to moderately contaminated category. None of the Arkansas River sites appeared to be severely contaminated (Table 3).
Despite cleanup efforts, however, concentrations of metals taken up by transplanted bryophytes in the Arkansas River suggest continuing metal impacts relative to other baseline Rocky Mountain sites and a continued need for monitoring and restoration efforts.
Transplanted bryophytes in the Arkansas River at various locations differed significantly in amount of metals up-take with concentrations in bryophyte tissue generally increasing in a downstream direction. Comparison of metal concentrations in transplanted Arkansas River bryophytes to groupings formed from other studies in the Rocky Mountains suggests that the Arkansas River is slightly to moderately contaminated with metals. Even the station furthest upstream, which is presumably above most important metal sources, contained Cd at concentrations higher than background. Although several metals point sources have been attenuated during clean-up efforts, other non-point and intermittent sources may remain. While this study has identified some metal contaminant areas of interest, an increased number of sampling locations would be required to discriminate between point, intermittent, and groundwater metal sources in the Arkansas River. Other investigators have found bryophytes to be superior to water sampling in locating intermittent (Say and Whitton, 1983; Mouvet, 1984) and ground water (Tremolieres et al., 1994) metal sources because they integrate metal concentrations over time and concentrate metals allowing for easier detection. As point-sources continue to be cleaned-up in the Arkansas River, the use of bryophytes to accurately locate other less easily detected sources (e.g., intermittent and/or groundwater) may increase in importance.
It is uncertain how metal concentrations in bryophytes relate to other biotic responses. The slight bryophyte contamination as demonstrated at EF-03 may not be important to aquatic macroinvertebrates. In a study on the Uncompahgre River where bryophyte metal concentrations and a variety of measures of aquatic macroinvertebrate communities were compared, the macroinvertebrate community did not appear to be impacted (Nelson and Campbell, 1995) at a station containing bryophyte metal concentrations similar to those at EF-03. Macroinvertebrate studies on the Arkansas River above and below the LMDT also suggest that there are no large differences in the macroinvertebrate community since water treatment has taken place (Nelson and Roline, in press) despite the differences between EF-01 and EF-03 that were detected with bryophytes. This suggests that differences detected in metals uptake by bryophytes that range from slight to non-contaminated may not be detectable in aquatic macroinvertebrate communities. Whether this is because the concentrations indicate a no impact level or because - community data analysis for macroinvertebrates is unable to discriminate impacts at a fine enough level is unknown at this time. Limited macroinvertebrate community data from locations with moderate and severe-contamination bryophyte categories suggests that aquatic macroinvertebrate communities are impacted at locations where bryophytes contain these higher amounts of metals.
The use of transplanted bryophytes in the Arkansas River to detect metals that are in low concentrations in water but potentially important to biota may be useful in restoration efforts. Metal concentrations in bryophytes were orders of magnitude greater than those measured in water, and in some cases only detected in tissue. Cadmium is an example of a metal that is sometimes difficult to detect in water samples but easily found in bryophyte tissue. This metal may be of special interest to brown trout (Salmo trutta) recovery efforts because of the belief that Cd is playing a role in the ability of brown trout to reach large sizes in the Arkansas River (U.S. Fish and Wildlife Service, 1993). Bryophyte tissue analysis could play a role in locating sources of this metal contamination.
Thanks to S. Porter, L. Carter, and S. Campbell for sharing bryophyte data from different areas and to T. LaCasse for assistance in field work. R. DeWeese was a partner in bryophyte gathering at Nate Creek Ditch and provided a source for moss bag preparation. Funding for this study was provided by the Bureau of Reclarnation's Eastern Colorado Area Office project 6C135 and WATER project EE010.
Goncalves, E.P.R., R.A.R. Boaventura, and C. Mouvet. 1992. Sediments and aquatic mosses as pollution indicators for heavy metals in the Ave river basin (Portugal). The Science of the Total Enviromnent 114:7-24.
Hurlbert, S.H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54(2):187-211.
Kiffhey, P.M. and W.H. Clements. 1994. Effects of heavy metals on a macroinvertebrate assemblage from a Rocky Mountain stream in experimental microcosms. J.N. AM. Benthological Soc. 13(4):511-523.
LaBounty, J.F., J.J. Sartoris, L.D. Klein, E.F. Monk, and H.A. Salman. 1975. Assessment of heavy metals pollution in the upper Arkansas River of Colorado. Bureau of Reclamation REC-ERC-75-5.
Mouvet, C. 1984. Accumulation of chromium and copper by the aquatic moss Fontinalis antipyretica L. ex Hedw transplanted in a metal-contaminated river. Environmental Technology Letters 5:541-548.
Nelson, S.M. and S.G. Campbell. 1995. Integrated assessment of metals contamination in a lotic system using water chemistry, transplanted bryophytes, and macroinvertebrates. Journal of Freshwater Ecology 10(4):409-420.
Nelson, S.M. and R.A. Roline. 1996. Recovery of a stream macroinvertebrate community from mine drainage disturbance. Hydrobiologia in press.
Roline, R.A. 1988. The effects of heavy metals pollution of the upper Arkansas River on the distribution of aquatic macroinvertebrates. Hydrobiologia 160:3-8
Porter, S.D. and S.M. Nelson. 1996. Uptake and elimination of metals by transplanted aquatic mosses. Paper presented at NABS-1996 44th Annual Meeting, Kalispell, Montana.
Rice, W.R. 1989. Analyzing statistical tests. Evolution 43(l):223-225.
Say, P.J. and B.A. Whitton. 1983. Accumulation of heavy metals by aquatic mosses. 1: Fontinalis antipyretica Hedw. Hydrobiologia 100:245-260.
Tremolieres, M. U. Roeck, J.P. Klein, and R. Carbiener. 1994. The exchange process between river and groundwater on the central Alsace floodplain (Eastern France): II. The case of a river with functional floodplain. Hydrobiologia 273:19-36.
U.S. Fish and Wildlife Services. 1993. Assessment of the trout population in the upper Arkansas River basin of central Colorado. U.S.F.W.S., Colorado State Office, Ecological Services.