Landscape ecotoxicology: linking catchment-scale geology to trace-metal bioavailability and benthic macroinvertebrate population and community responses

Abstract

The mobilization of trace-metals in surface water is a natural phenomenon resulting from the weathering of bedrock. However, mining and mineral extraction operations expose large volumes of newly disturbed, finely crushed rock which causes unnaturally high weathering rates and results in poor water quality, degraded stream habitat, and exposure of aquatic com m unities to toxic concentrations of trace-metals. Two missions of federal property managers are to manage lands for recreation (i.e., fishing) and the production of fresh water. As the population of the intermountain west grows, the demand for fishable and potable water will continue to increase. In anticipation of this need, and the knowledge that trace-metal contamination from historical land uses may impair water quality, the federal government is interested in determining to what extent mining has impaired water quality above and beyond natural background levels. In an effort to develop baselines for geochemical and biological conditions using catchment-scale geology, we conducted a regional-scale monitoring study to quantify the relationship between geology and water chemistry in streams throughout the mineralized belt of the Central Colorado Rocky Mountains. Because total-recoverable metals is a poor predictor of toxicity to aquatic organisms and because there is no universal way to quantify the bioavailability and toxicity of any single trace-metal, we developed a model capable of predicting toxicity due to trace-metal mixtures. This new model, the Chronic Criterion Accumulation Ratio (CCAR), is derived from the Biotic Ligand Model (BLM) and incorporates current theory about the interactions between aqueous constituents (e.g., hardness, DOC, pH) that affect trace-metal toxicity and accumulation. We first employed CCAR to determine which geologies produced water qualities that impair aquatic communities. We tested if the BLM-derived CCAR was predictive of trace-metal accumulation and population responses of three native taxa (i.e., ephemeropterans Rhithrogena spp. and Drunella spp. and the trichopteran Arctopsyche grandis) which are candidates as bio-indicators of trace-metal contamination. This evaluation also allowed us to compare information provided by presence/absence data vs. population density responses of Rhithrogena spp., Drunella spp., and A. grandis to trace-metals and to test the hypothesis that accumulation of trace-metals in individuals results in deleterious effects that propagate to a population-level response. We also tested the model by comparing the predictive capacities of CCAR to a traditional toxic unit model, Cumulative Chronic Units (CCU). Geologic alteration was found to be a strong driver of dissolved trace-metal concentrations and alkalinity in Rocky Mountain streams. Primary rock-type (i.e., sedimentary, metamorphic, igneous) and geologic age (i.e., Cenozoic, Mesozoic/Paleozoic, Pre-Cambrian) interact with hydrothermal alteration and pyrite mineralization to determine water quality and toxicity to aquatic organisms. Three levels of impairment of surface water due to geology were identified in this research: 1) geologies that produce such high concentrations of trace-metals that other water quality characteristics that determine trace-metal bioavailability were unimportant; 2) geologies that produce waters of moderate toxicity to aquatic organisms and where other water quality parameters may play a role in determining impairment and toxicity; and 3) those geologies which do not produce trace-metals at sufficient concentrations to impair water quality or aquatic organisms. We hypothesize that geologies that produce moderately toxic waters are the most susceptible to further impairment by mining activities. CCAR was found to be correlated with whole body metal concentrations of benthic macroinvertebrates Rhithrogena spp., Drunella spp. and A. grandis. The probability of detection for a given taxa, as determined from logistic regression of presence/absence data, was relatively insensitive to trace-metals as compared to population density. This disparity indicated that native populations are likely comprised of subpopulations which are differentially sensitivity to trace-metal exposure. These data support the conclusion that toxicity is not constant for all individuals of a population, thus violating an important assumption of the BLM. Also, water quality criteria based on the BLM were protective of some native taxa (e.g., Drunella spp.), but not others (e.g., A. grandis or Rhithrogena spp.). Results of the comparison between CCAR and CCU indicated that CCAR explained more variation in benthic macroinvertebrate com m unity metrics (e.g., richness, abundance and composition). More importantly, we found deleterious effects in benthic macroinvertebrate communities at levels of trace-metal contamination previously thought to be safe. Utilizing simple linear regression equations developed in this study, we calculated the average change in macroinvertebrate comm unity metrics between CCAR = 0.07 (i.e., background) and CCAR = 1 (i.e., the theoretical threshold of chronic toxicity). On average we observed a 27% reduction in richness, a 72% decrease in abundance, and a 15% change in community composition between background concentrations of trace-metals and the threshold of chronic toxicity. This research was the first to investigate use of the Biotic Ligand Model (BLM) for evaluating native taxa responses to trace-metals. Because CCAR is a closer approximation of the causal processes which determine toxicity of trace-metals, we found that aquatic communities were more sensitive to trace-metals exposure than previously thought. This research was also the first to link accumulation of trace-metals in individuals to population-level responses and to relate trace-metal bioavailability to catchment-scale rock type and geologic processes. Results of these studies will be used to develop protocols for setting biological and geochemical baselines so that federal land managers can properly target and prioritize abandoned mine lands for reclamation to improve potable water quality and fisheries habitat in Colorado.