Use of Community Data to Diagnose the Causes of Stream Impairment III
Presiding: M Paul, Howard University; S Norton, U.S. Environmental Protection Agency
NB44B-01 INVITED 15:30h
Pulling Pattern From Pandemonium: Using Stream Community Data to Diagnose Stressors
Proliferation of bioassessment methods has increased knowledge about biological condition of streams in the US. Identifying degradation is only the first step; diagnosing likely causes is a harder task and community data have not traditionally been used in this process. However, the potential for taxa or even assemblage based responses to specific stressors exists. In this paper, I discuss two approaches using community assessment data for diagnosing stressors. I built RIVPACS type predictive models using the Maryland Biological Stream Survey invertebrate dataset. These models were used to identify degraded sites. These were split into primary degradation types based on land use and water chemistry data. I then compared two methods to identify taxa related to the specific stressors. The first was to ordinate community data from reference and stressor sites to identify specific taxa responses to stress, and then weigh the responses to generate stressor probabilities. The second approach used predictive model output to identify tolerance characteristics of each taxon to specific stressors. Taxa scores for each stressor were summed to score likely stressors for a site. Both approaches had advantages and disadvantages, but the predictive modeling derived scores relied on less additional analyses. Both approaches hold promise for diagnosis.
NB44B-02 INVITED 15:45h
Multiple Lines and Levels of Evidence for Detecting Ecological Responses to Management Intervention
Every year, millions of dollars are spent on river management and rehabilitation activities across Australia. Most of these activities are based on assumptions about the ecology of river systems and potential causes of degradation. River management activities are widespread but assessment of the their effectiveness relative to anticipated ecological outcomes is less common. Rivers usually have multiple stressors and causal relationships between them, management interventions and environmental condition are of interest. Evaluating the ecological response of rivers to management interventions can be complex. Limited opportunities for replication and randomization often makes the more common impact assessment methods (e.g. BACI designs) difficult to apply, thereby reducing our ability to draw inferences on causality. A Multiple Levels and Lines of Evidence (MLLE) schema is presented from which it is possible to examine evidence for causality between environmental stressors, management interventions and ecological outcomes. MLLE was originally developed for epidemiological studies when it was difficult to assign causality. Here we apply the MLLE schema to the design of monitoring programs for assessing the ecological outcomes of environmental flow releases. The method complements the approaches adopted by various jurisdictions, such as the IMEF process in NSW and the Australian Water Quality Monitoring Guidelines.
http://freshwater.canberra.edu.au
NB44B-03 16:00h
Taxonomic Specificity or Indices? Aquatic Community Data or Water Chemistry? Applying Artificial Neural Network Models to a Variety of Aquatic Monitoring Approaches
This study evaluates the potential use of artificial neural network (ANN) modeling for freshwater conservation planning based on stream macroinvertebrate and water chemistry data from three years (2000-2002) at 60 sites in the New York City drinking water watershed. ANNs can learn from examples, generalize to novel situations, tolerate noisy data and approximate any continuous function, making ANNs ideal for modeling the complexities of aquatic community data. ANNs based on several levels of taxonomic resolution are compared to determine the gain in predictive capacity with increasing taxonomic specificity. In addition an ANN model using water chemistry data is compared to the effectiveness of using aquatic macroinvertebrate community data. The ANN models were trained to independently identify the impact of percent agriculture, impervious surface and forested cover. Set aside datasets validated the capacity of the ANN models to correctly predict the three landcover impacts. Models using species indices such as EPT richness, HBI, or water quality scores had relatively low predictive power. A combination of 73 baseline water chemistry variables demonstrated greater predictive power than family level data but by far the best model used species level taxonomy.
NB44B-04 16:15h
Separating the Effects of Fine Sediments From Those of Nutrients in Agricultural Streams
Increased inputs of both fine sediments and nutrients to streams are associated with alterations of catchment land cover. However, the mechanisms by which altered land cover causes increased inputs differs between sediments and nutrients. This results in variation in the relative roles of these stressors in reducing biotic integrity in different stream reaches. Therefore, research that differentiates the relative effects of these stressors on biotic communities can guide efforts to ameliorate these effects. Using data collected from 1999 to 2003 in 35 tributary reaches in the Little Miami River basin, Ohio, we assessed the relationships between nutrients, bedded fine sediments, and community metrics for periphyton, macroinvertebrates, and fish. Periphyton exhibit the most distinctive responses to nutrients and sediments and are at least partially responsible for the effects observed in macroinvertebrates and fish. The positive relationship between periphytic biomass and increased nutrients affects variation in dissolved oxygen, whereas the negative relationship between periphytic production and increased fine sediments affects trophic resources, particularly for macroinvertebrate grazers. Attention to such contrasting effects can guide the selection of community metrics that can better discriminate the relative severity of these two stressors in streams.
NB44B-05 16:30h
Relationships among Land-Use, In-Stream Stressors, and Biological Condition in Prince George's County, MD
As human disturbance increases in watersheds there is a resulting change in hydrologic stability that leads to alterations of in-stream habitat conditions. These in-stream alterations are called stressors because they represent sub-optimal to lethal conditions for aquatic organisms. The linkages and mechanisms that relate multiple stressors with complex landscape features (sources) are currently the focus of research across North America. The objective of this project was to illustrate the linkages and biological responses for coastal plain watersheds that contain a gradient of severity and types of human disturbance. Stepwise multiple-regression was performed on a routine biological monitoring database from Prince George's County, MD. Results demonstrated that urban land-use sources (medium-density residential, commercial, and industrial) and in-stream stressors of reduced physical habitat complexity (e.g., decreased gravel substrate, reduced channel sinuosity etc.) were most related to biological degradation. The biological response variables that had the strongest relationships with these sources and stressors were: the indexes of biotic integrity, EPT Index, Beck's Biotic Index, % Dominant Fish Species, and % generalists, omnivores, and invertivores. These types of analyses need to be conducted in a variety of ecoregions to understand the dynamics of the relationship between human disturbance and the biological condition of streams.
NB44B-06 16:45h
Developing an Environmental Decision Support System for Stream Management: the STREAMES Experience
Transferring research knowledge to stream managers is crucial for scientifically sound management. Environmental decision support systems are advocated as an effective means to accomplish this. STREAMES (STream REAach Management: an Expert System) is a decision tree based EDSS prototype developed within the context of an European project as a tool to assist water managers in the diagnosis of problems, detection of causes, and selection of management strategies for coping with stream degradation issues related mostly to excess nutrient availability. STREAMES was developed by a team of scientists, water managers, and experts in knowledge engineering. Although the tool focuses on management at the stream reach scale, it also incorporates a mass-balance catchment nutrient emission model and a simple GIS module. We will briefly present the prototype and share our experience in its development. Emphasis will be placed on the process of knowledge acquisition, the design process, the pitfalls and benefits of the communication between scientists and managers, and the potential for future development of STREAMES, particularly in the context of the EU Water Framework Directive.
http://www.streames.org