Integrating Hydrology and Ecology for Watershed Research I
Presiding: T Mihuc, Plattsburgh State University; E Romanowicz, Plattsburgh State University
NB51C-01 08:30h
Influence of Natural and Anthropogenic Disturbance Regimes on the Fate and Transport of Metals in a Mine-Impacted River
A complex set of disturbances, including hydrology, contaminated floodplains, and remediation, impact temporal trends of metal contamination in the Clark Fork River, MT. A 15-year dataset showed declining Cu trends in resident biota and sediment, indicating a positive remediation response in the upper 30 Km. However, identifying effects of remediation from 85-190 Km were confounded by the significant positive correlation between bioaccumulation and stream discharge. Because remediation focuses primarily on the upper 45 Km, spatial correlations in temporal Cu trends were used to determine how a reduction in upstream contamination levels would influence metal concentrations at each downstream station. Site-specific influence on metal concentrations was inferred when temporal patterns in downstream stations were significantly correlated to patterns at an upstream station. Correlations indicated strong connections between adjacent stations; 60-75 % of adjacent stations were significantly correlated over time, suggesting that events upstream usually affect the next downstream station. Correlations also indicated two upstream sites that appear to be major sources of metal. For example, sediment Cu concentrations at 5 and 10 Km were significantly correlated with concentrations at five out of seven downstream stations, showing evidence of upstream-downstream contaminant linkages along a 190 Km reach.
NB51C-02 08:45h
The influence of catchment geology on water quality and benthic macroinvertebrate communities in streams from the Colorado Mineral Belt.
The influence of catchment geology on stream geochemistry and aquatic communities in streams located in the central part of the Colorado Mineral Belt was investigated. Catchments (n=44) were selected based on being underlain by a single (n=23) or a combination from among eight lithologies (e.g., pluton, mafic metavolcanic, tuff). Water quality and Hess samples of benthic macroinvertebrate communities (n=5) were collected from one representative riffle area along each stream. Statistical differences observed (ANOVA, Dunnet's C post hoc test, p<0.05) in water quality parameters (e.g., Cd2+, Cu2+, Zn2+, Ca2+, Mg2+, pH) and benthic community metrics (e.g., richness, EPT-richness) were attributable to catchment geology categorization. Principle components analysis described over 94% of the variation in water quality data and showed both negative [Cd2+, Cu2+, Zn2+] and positive [CaCO3, Cl-, Mg2+] gradients among streams sampled. Stepwise canonical discriminant function analysis of community composition data resulted in 87% correct classified of apriori catchment geology groupings. Intensity of hydrothermal alteration and/or historical mining activity within geologic units contributed to variability in stream chemistry and community ecology measures. Expansion of the study will include additional stream sampling and the analysis of stream and catchment-scale habitat data.
NB51C-03 INVITED 09:00h
Catchment Geology Affects Nitrogen Export in Streams Draining the Teton Range, Wyoming
The presence of bedrock minerals containing phosphorus (P) may influence catchment nutrient retention. For example, catchments with P-poor lithology may export atmospheric nitrogen (N) inputs via stream flow to a greater degree than P-replete catchments. The Teton Range in northwest Wyoming presents ideal conditions to test this hypothesis because Teton catchments have varying amounts of P-rich, calcareous bedrock and P-poor, crystalline bedrock, yet all receive roughly equal atmospheric N inputs. Over several days in summer 2004, we measured discharge and sampled water in nearly every stream draining the Teton Range (n=29). Following sample collection, we determined bedrock geology upstream of our sampling points using GIS. All streams we sampled across the P-poor to P-rich lithology gradient had orthophosphate concentrations below detection. Specific conductivity was tightly linked to measures of catchment lithology, increasing with calcium concentration and proportion of calcareous bedrock (multiple linear regression (MLR), r2=0.96, p<0.001). Dissolved inorganic N flux (mgN/s) decreased with increasing specific conductivity and calcareous bedrock area and increased with crystalline bedrock area (MLR, r2=0.79, p<0.001); particulate organic N flux followed the same trend (MLR, r2=0.85, p<0.001). These data support our hypothesis that catchments with P-poor lithology export more N than catchments with P-rich bedrock.
NB51C-04 09:15h
Inferring Sources of Nitrate and the Effect of In-stream Processes on Stream Nitrate Concentrations Through End-Member Mixing Analysis in a Mediterranean Stream.
The goals of this study were to infer catchment nitrate sources and to elucidate the effect of in-stream processes on streamwater nitrate through the application of end-member mixing analysis (EMMA). The study was performed in Fuirosos, an intermittent stream draining a forested Mediterranean catchment (10.5 km2). Streamwater was collected during 25 storms. Event water, riparian groundwater and water from headwater springs at the point of discharge were also monitored. Streamwater data encompassed the mixing space defined by the 3 end-members only during the 12 storms occurred during the wet period (from December to May). The estimated proportion of water coming from each source of runoff was plotted against stream nitrate concentrations. A positive relationship was understood such as the compartment was a source of nitrate. Nitrate was coming either from event water or groundwater. Nitrate arriving from the catchment to the stream was estimated by multiplying the flowpath nitrate concentrations by the flowpath fractions and summing the products. These predicted concentrations were compared with those measured in Fuirosos on each sampling data. Only at low flows (< 100 l/s) stream nitrate was lower than predicted from catchment sources, indicating that nitrate was retained in the stream.
NB51C-05 09:30h
Integrating Hydrology and Ecology for Watershed Research in Topographically Rugged Landscapes: The Landscape Continuum Model
Our Landscape Continuum Model (LCM) explicitly links terrestrial ecosystems to each other and to aquatic ecosystems in topographically rugged landscapes. The heart of the model is that strong linkages are generated among landscape components as a result of transport processes caused by extreme topography. These transport agents cause biogeochemical amplification and attenuation of processes not observed in most landscapes. We illustrate the LCM for alpine/subalpine aquatic systems with two decades of limnological research at the Niwot Ridge LTER site, located in the Colorado Front Range. Our measurements of streamwater quantity and quality show that the amount of inorganic nitrogen per unit catchment area exported downstream in stream water exceeds average wetfall inputs from the atmosphere near the mountaintops and then decreases downcanyon. Moreover, as elevation decreases, there is an almost linear pattern of increasing concentrations of dissolved organic carbon (DOC) and decreasing nitrate concentrations, suggesting retention and conversion of inorganic nitrogen to organic matter within aquatic communities. Consistent with the down-gradient increase in DOM is a switch from labile hydrophilic fractions of DOC in alpine systems to recalcitrant hydrophobic fractions of DOC in subalpine aquatic systems. Further, dual-isotope analysis of nitrate combined with end-member mixing analysis (EMMA) indicates that the snowmelt period begins with a dominance of atmospheric nitrate in stream water and then a gradual decline in atmospheric dominance and replacement by nitrate originating in talus fields. Additionally the flowpath results indicate a dominance of groundwater and return flow in stream water and also thus a dominance in biogeochemical control on nitrogen export in the groundwater of the basin after snowmelt.
NB51C-06 INVITED 09:45h
Effects of Fragmentation on Macroinvertebrate Community Composition in Vernal Pools
The dynamic nature of vernal pools attracts aquatic communities highly adapted to hydrologic disturbance, but these systems remain susceptible to landscape alterations. We examined the effects of forest fragmentation on invertebrate community structure in temporary pools in northern Minnesota. D-frame nets were used to sample invertebrates in 18 fragmented and 20 unfragmented pools in the early spring and summer during high and low water levels. Twenty metrics based on invertebrate characteristics were analyzed using a mANOVA to examine treatment and seasonal interactions. A majority of the variance (50 percent) was explained by factors related to treatment, and Canonical Correspondence Analysis (CCA) results showed weak relationships between community structure and habitat variables. Total taxa increased slightly between early and late sampling for both treatments, but the fragmented pools contained significantly more taxa and a higher overall abundance than in the unfragmented pools. These differences may be attributed to hydroperiod, as fragmented pools dried up with less frequency, and experienced less water volume loss than the unfragmented pools. Although seasonal changes in vernal pool communities occurred, aquatic insects showed a greater response to local habitat conditions. Landscape alterations such as forest fragmentation can significantly alter temporary pool invertebrate community composition.