Document Type
Report
Publication Date
11-2003
Pages
21 pp.
Abstract
To meet the requirements of the Clean Water Act, the States of Maryland and Virginia are in the process of developing biological criteria for evaluating estuarine waters. The Chesapeake Bay Benthic Index of Biotic Integrity (B-IBI) is the basis for these biological criteria. As an initial step, a Workgroup developed an interim framework for the application of the B-IBI to the States’ water quality inventory report (305b report) and list of impaired waters (303d list). The Workgroup included representatives of the Maryland Department of the Environment (MDE), Maryland Department of Natural Resources (MDNR), Virginia Department of Environmental Quality (DEQ), USEPA Region III Office of Watersheds, Old Dominion University Department of Biological Sciences, and Versar, Inc.
Two alternative methods were initially proposed for 303(d) impairment decisions based on the BIBI: (1) a weighted (stratified) mean approach, and (2) comparisons of cumulative frequency distributions and proportions. In the first method, the mean benthic condition of a Chesapeake Bay segment (as determined by the stratified mean B-IBI value) would be compared against the mean of the reference condition (based on the original index development reference sites) using a t-test. Because a segment may be represented by more than one habitat for which reference conditions may differ, weights are applied to combine means for each habitat type (strata) (Cochran 1977). In the second method, the cumulative frequency distribution of B-IBI values for a segment would be compared to the cumulative distribution of B-IBI values for the reference condition, and a test conducted to determine whether the percentage of sites with low B-IBI scores (below a threshold) significantly exceeds the percentage of sites with low scores in the reference distribution, and whether the percent exceedance is above a specified threshold (e.g., 25%).
Each of the above methods presents advantages and disadvantages that can only be evaluated statistically. However, the weighted mean approach is only applicable if each habitat has two or more samples, while the cumulative frequency distribution method can be applied only when the reference distributions for the seven Chesapeake Bay habitats are similar and the data can be pooled into one reference distribution. Examination of the B-IBI reference distributions for the seven habitats indicated significant differences among habitats and small sample sizes. Based on this result, an alternative non-parametric frequency distribution approach that does not require the pooling of the reference data was applied. The method of choice, a stratified Wilcoxon rank sum test, treats the habitats within a segment as strata, and tests for significant shifts in the distribution of B-IBI data between the assessment and the reference B-IBI values. We describe below the method and present each of the analytical steps leading to the assessment and identification of Chesapeake Bay segments with estuarine benthic impairments.
Rights
Included with kind permission from the author(s).
Repository Citation
Llansó, Roberto J.; Volstad, Jon H.; and Dauer, Daniel M., "Decision Process for Identification of Estuarine Benthic Impairments" (2003). Benthic Applications. 5.
https://digitalcommons.odu.edu/chesapeakebay-archive_benthic/5
Included in
Environmental Indicators and Impact Assessment Commons, Environmental Monitoring Commons, Terrestrial and Aquatic Ecology Commons
Comments
This report, Decision Process for Identification of Estuarine Benthic Impairments, was prepared by Versar, Inc., at the request of the Maryland Department of Natural Resources, Maryland Department of the Environment, Virginia Department of Environmental Quality, and USEPA Region III under Cooperative Agreement CA-03-24-07-4-30831-3734 between Versar Inc., and the University of Maryland Center for Environmental and Estuarine Studies. Dr. Daniel M. Dauer, of Old Dominion University, contributed to project conceptualization and evaluation. Versar wishes to thank members of the Workgroup, including representatives of the above-mentioned agencies, for their support and for contributing guidance to the project. We are grateful to Ms. Jody Dew, Versar’s SAS Programmer, for skillfully managing data and conducting statistical analyses within the very short time frame of the study.