Grantee Research Project Results
2005 Progress Report: Model-based Clustering for Classification of Aquatic Systems and Diagnosis of Ecological Stress
EPA Grant Number: R831368Title: Model-based Clustering for Classification of Aquatic Systems and Diagnosis of Ecological Stress
Investigators: Smith, Eric , Orth, Donald J. , Yagow, Gene , Berkson, Jim , Brannan, Kevin , Mostaghimi, Saied , Bates, Samantha
Institution: Virginia Tech
EPA Project Officer: Packard, Benjamin H
Project Period: November 10, 2003 through November 9, 2006
Project Period Covered by this Report: November 10, 2004 through November 9, 2005
Project Amount: $843,771
RFA: Development of Watershed Classification Systems for Diagnosis of Biological Impairment in Watersheds and Their Receiving Water Bodies (2003) RFA Text | Recipients Lists
Research Category: Watersheds , Water
Objective:
The objectives of this research are to develop methodologies for classifying watersheds and to evaluate the ability of this classification system to delineate areas of biological stress. Methods were developed to cluster regression relationships using mixture likelihoods, classification likelihoods, and a method based on Voronoi tessellations. Of the three approaches, we believe the program based on Voronoi tessellations will be most useful, as it allows for general spatial patterns to be used for clusters. In addition, the program also allows a user to search for interesting patterns associated with a particular stressor such as total phosphorus, giving an indication of the degree and extent of the stressor-response relationship. The methods were applied to data from Ohio, the Environmental Monitoring and Assessment Program (EMAP) Mid-Atlantic Integrated Assessment (MAIA) data using the Index of Biotic Integrity (IBI) as the biological response and several environmental measurements as stressors. The analysis of the data on MAIA has found several interesting clusters based on watershed area, pH and other stressors. An application to benthic indices resulted in clusters that were somewhat similar in spatial extent to those from IBI, although the stressors were different.
Progress Summary:
In addition to the focus on finding patterns of environmental stress, we also are continuing our research on improving Total Maximum Daily Loads (TMDLs). Because of the high costs associated with the implementation of TMDLs, it is essential that TMDLs be developed using sound scientific methods that are able to accurately reflect the pollutant loadings from the potential sources within a watershed. Currently, nonpoint source (NPS) pollution models are most frequently used to determine the maximum allowable loading rates of bacteria from the identified sources. Most existing NPS models simulate bacteria transport during overland flow events as a dissolved pollutant, neglecting the attachment to sediment particles. The few models that do account for attachment require the user to input unknown partitioning coefficients. The U.S. Environmental Protection Agency has identified bacteria, predominantly from agricultural sources, as the leading cause of impairments in rivers and streams in the United States. The goal of this study was to improve understanding of partitioning between planktonic and particulate attached Escherichia coli (E. coli) and Enterococcus in runoff from pasturelands. A comparative study was conducted to identify the best method of dispersing wild strains of indicator organisms from sediment and organic matter particles present in runoff from pasturelands. Chemical surfactants (Tween-80, Tween-85, sodium pyrophosphate), mechanical dispersion techniques (sonication, hand shaker), and combinations of the techniques are being compared. Following identification of the optimal dispersion technique, screen filtration will be used to identify the particle sizes to which E. coli and Enterococcus preferentially attach. Preliminary results found Tween 85 (1,000 mg L-1) to increase enumeration of E. coli in runoff samples by 26%. Preliminary results using the screen filtration device indicate 56% of E. coli was in the planktonic form or associated with sediments passing through a 25 im screen, 15.5% was associated with sediments retained on a 60 im screen and 28.5% was retained by a 120 im screen.
Future Activities:
The next year will involve the following activities:
- Apply the method to the data set from the Maryland biological survey and to a data set on brook trout presence-absence in the eastern United States.
- Extend the methods to allow for nonparametric regression analysis. The nonparametric approach may improve the strength of the relationships.
- Extend the methods to allow for multivariate regression approaches. The extension will allow for the analysis of sets of metrics rather than just a single metric. Also, methods such as canonical correspondence analysis will be implemented to allow for the inclusion of species counts as response variables as well as metrics.
- Develop methods to improve evaluation of standards for water quality using information from multiple sites or historical data. Use clustering to find sets of sites with similar characteristics to the target site. We expect this to improve the power of the decision about site quality.
- Complete TMDL studies on surfactants and E. coli.
Journal Articles:
No journal articles submitted with this report: View all 3 publications for this projectSupplemental Keywords:
RFA, Scientific Discipline, Water, ECOSYSTEMS, Ecosystem Protection/Environmental Exposure & Risk, Water & Watershed, Aquatic Ecosystems & Estuarine Research, Monitoring/Modeling, Aquatic Ecosystem, Terrestrial Ecosystems, Environmental Monitoring, Ecological Risk Assessment, Ecology and Ecosystems, Watersheds, risk assessment, ecosystem modeling, anthropogenic stress, watershed classification, watershed, ecosystem monitoring, decision making, water quality, model based cluster anaysis, ecological risk, aquatic ecosystems, environmental stress, stressor effect relationships, ecological indicators, ecology assessment models, ecosystem stress, watershed assessment, ecological models, water monitoring, adaptive implementation modeling, stress responseProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.