MSU: Department of Zoology, Michigan State University, National Food Safety and Toxicology Center and Institute of Environmental Toxicology, East Lansing, MI 48824, USA.

MDEQ: Surface Water Quality Division, Michigan Dept of Environmental Quality, P.O. Box 30273, Lansing, MI 48909-7773, USA.

PREFACE

Hundreds of chemical contaminants have been identified in the Great Lakes System. Depending on the agency or organization, various subset lists of these contaminants have been identified as chemicals of concern. However, there is no agreement on the method which should be used to make this decision. Except for approximately 40 chemicals that most agencies agree are potential problems, no agreement has been reached on what priority these chemicals should receive for further action. That leaves hundreds of Great Lakes contaminants yet to be evaluated. Concern for adverse environmental impacts has been identified for chemicals with the characteristics of persistence, bioaccumulation, and toxicity. Except for the International Joint Commission’s definition of persistence (> 8 weeks in air, water, soil or sediment), there is little agreement as to what defines these characteristics. The State of Michigan currently has no established definition for persistent, bioaccumulative, toxic substances. Further, there is no established process to rank chemicals from high to low hazard based on the characteristics of persistence, bioaccumulation, and toxicity. A chemical hazard scoring and ranking model has been developed to provide Michigan with an objective process to rank order chemicals based on these characteristics. Use of the model may assist in pollution prevention activities and other future chemical control efforts, allowing decision making on a most hazardous first basis. The scope of this project was to develop a scoring system, to test the system by scoring a few example chemicals, and to determine which of the identified Great Lakes contaminants have sufficient data for scoring.

EXECUTIVE SUMMARY

It has been estimated that there are approximately 100,000 chemicals in commercial use. Of these, approximately 40,000 are used in the Great Lakes Basin. Approximately 2,300 new chemicals are submitted to the U. S. Environmental Protection Agency Office of Pollution Prevention and Toxics (OPPT) each year (Zeeman et al., 1995). [See Final Report for references] Simple scoring and ranking criteria have been used in selecting compounds to be regulated (MDNR, 1987; MOEE, 1990; Meek, 1996; Hertel, 1996; Hansen, 1994; Klöpffer, 1994a,b,c,d; Anon., 1995) or even sunsetted from use (Foran and Glenn, 1993).

There are a number of schemes that have been suggested to screen for potential of compounds to cause adverse effects once released into the environment. The University of Tennessee Center for Clean Products and Clean Technologies identified 147 different scoring and ranking systems and described and critically reviewed 51 of those systems (Davis et al., 1994). These have been reviewed, compared, contrasted, and evaluated in a workshop conducted by the Society of Environmental Toxicology and Chemistry (SETAC, 1997). This panel of experts gave suggestions on the preferred methods of estimating, aggregating, and weighting of parameters.

While these vary in the parameters used or in methods of estimation or aggregation of the parameters within the model, they are all basically the same in the manner in which they function. The parameters used, either measured or modeled, are estimators or surrogates for the potential for exposure and hazard. In general some estimates of persistence, bioaccumulation, and toxicity are the main parameters used (SETAC, 1997). In addition, the amount of compound manufactured or used or the types of uses can also be considered.

Even though little empirical information is available about the fate, exposure, or hazard profiles of a compound, estimates of key parameters that determine these properties can be determined by quantitative structure-activity relationships (QSARS) (Hermens et al., 1995). These statistical relationships are often quite powerful, but there is still a great deal of uncertainty in the exposure or hazard profiles based on these incomplete data sets (Fiedler et al., 1990; Tosato et al., 1991). None of the previously developed scoring and ranking systems have taken this uncertainty into account. The Scoring and Ranking Assessment Model (SCRAM) was developed to serve as an analytical tool in chemical scoring and ranking. This simple spreadsheet model allows risk assessors and managers to calculate a simple index that is based on the potential exposure and hazard profiles of chemicals and aids in determining the sufficiency of available data. SCRAM is unique in its treatment of chemicals for which data are lacking. If few data are available for a chemical, instead of ignoring the chemical, the system flags it as one for which there is a great deal of uncertainty. Thus, chemicals can become candidates for special attention based either on inherent chemical characteristics or on a lack of information. This strategy allows regulators to bin chemicals into three groups: those which clearly have the potential to cause problems in the environment, those which do not have this potential, and those for which there is little information.

While the scores can be aggregated to spur action to reduce uncertainty, they are also kept separate to make the reason for a chemical score transparent. The model is weighted to be biased toward giving greater scores to more persistent chemicals. This was done intentionally because even if a compound is deemed to be fairly nontoxic based on current knowledge, if it is persistent and bioaccumulative it may have the potential to cause adverse effects to endpoints or receptors that are not currently considered.

SCRAM is a chemical scoring and ranking model designed to evaluate a minimal amount of data to perform an initial screening of large numbers of new and existing chemicals to identify those which have the potential to cause problems in the environment. This model will allow resources to be concentrated on evaluating more fully those chemicals identified by the model as potential problems. SCRAM is not a replacement for a risk assessment. It is intended to lead to risk assessments of chemicals which are identified as potentially serious concerns.

This model is designed to screen individual chemicals. It is not equipped to deal with mixtures, with the exception of mixtures of defined composition, such as some pesticide mixtures or Aroclor mixtures. Mixtures of defined composition can be screened because environmental monitoring for these mixtures takes place, and the individual compounds in these mixtures occur together in a constant, identifiable ratio. Other mixtures do not occur in constant ratios in the environment. Limited toxicity data are available for mixtures, and as the mixture changes, toxic effects may change. Although no organism in the environment is exposed to a single chemical, and evaluation of mixtures is important, this model is not designed to deal with these issues.

ACKNOWLEDGMENTS

This project was completed through the combined efforts of the Surface Water Quality Division, Michigan Department of Environmental Quality and the Department of Fisheries and Wildlife, Michigan State University. Assistance and advice in developing the chemical ranking and scoring model was provided by an advisory committee of toxicologists, chemists, and biologists from several Michigan industrial organizations, environmental groups, and academicians, as well as technical representatives from USEPA Region 5, the Ontario Ministry of Environment and Energy, Wisconsin, and Minnesota. Funding for this project was provided through a grant from the Michigan Great Lakes Protection Fund, administered by the Office of the Great Lakes, Michigan Department of Environmental Quality.

PROJECT REPORT

The report on the project, entitled, "SCRAM: A Scoring and Ranking System for Persistent, Bioaccumulative, and Toxic Substances for the North American Great Lakes," was published in four parts by Environmental Science & Pollution Research (ESPR), Ecomed Publishers, D-86899 Landsberg, Germany. Part I, Part II, Part III, and Part IV were published online by ESPR on 29 October 1999. Print publication began with Part I in the January 2000 issue, and the remaining parts followed in succeeding issues. The publisher has graciously provided electronic versions of all four parts and has granted permission for them to be made available on this site.

The files below are in Adobe Portable Document Format (PDF), as provided by the publisher. They can be viewed and printed using the free Adobe Acrobat Reader.

• Part I: Structure of the Scoring and Ranking System (ESPR No. 1, 2000).
  <espr199910.009.pdf, 2/2000, 393 Kb>
  
  
• Part II: Bioaccumulation Potential and Persistence (ESPR No. 2, 2000).
  <espr199910.010.pdf, 2/2000, 354 Kb>
  
  
• Part III: Acute and Subchronic or Chronic Toxicity (ESPR No. 3, 2000).
  <espr199910.011.pdf, 2/2000, 364 Kb>
  
  
• Part IV: Results from Representative Chemicals, Sensitivity Analysis, and Discriminatory Power (ESPR No. 4, 2000).
  <espr199910.012.pdf, 2/2000, 367 Kb>
  

References:
EM Snyder, SA Snyder, JP Giesy, SA Blonde, GK Hurlburt, CL Summer, RR Mitchell, and DM Bush. SCRAM: A scoring and ranking system for persistent, bioaccumulative, and toxic substances for the North American Great Lakes. Part I. Structure of the scoring and ranking system. ESPR -- Environ. Sci. & Pollut. Res. Online Publication (Online First): October 29th, 1999. Print Publication: ESPR 7 (1) 2000.

EM Snyder, SA Snyder, JP Giesy, SA Blonde, GK Hurlburt, CL Summer, RR Mitchell, and DM Bush. SCRAM: A scoring and ranking system for persistent, bioaccumulative, and toxic substances for the North American Great Lakes. Part II. Bioaccumulation potential and persistence. Acute and subchronic or chronic toxicity. ESPR -- Environ. Sci. & Pollut. Res. Online Publication (Online First): October 29th, 1999. Print Publication: ESPR 7 (3) 2000.

EM Snyder, SA Snyder, JP Giesy, SA Blonde, GK Hurlburt, CL Summer, RR Mitchell, and DM Bush. SCRAM: A scoring and ranking system for persistent, bioaccumulative, and toxic substances for the North American Great Lakes. Part III. Acute and subchronic or chronic toxicity. ESPR -- Environ. Sci. & Pollut. Res. Online Publication (Online First): October 29th, 1999. Print Publication: ESPR 7 (3) 2000.

EM Snyder, SA Snyder, JP Giesy, SA Blonde, GK Hurlburt, CL Summer, RR Mitchell, and DM Bush. SCRAM: A scoring and ranking system for persistent, bioaccumulative, and toxic substances for the North American Great Lakes. Part IV. Results from representative chemicals, sensitivity analysis, and discriminatory power. ESPR -- Environ. Sci. & Pollut. Res. Online Publication (Online First): October 29th, 1999. Print Publication: ESPR 7 (4) 2000.

See below for information on scoring and ranking of chemicals using the SCRAM.

CHEMICAL SCORES & RANKINGS

The Scoring and Ranking Assessment Model (SCRAM) was used to calculate scores and rank 142 chemicals that have been reported found in the Great Lakes. An article published in the April 2002 issue of the Journal of Human and Ecological Risk Assessment reports the scores and relative rankings that were generated. The publisher has graciously provided an electronic version of the article, and has granted permission for it to be made available on this site:

• SCRAM: A scoring and ranking system for persistent, bioaccumulative, and toxic substances for the North American Great Lakes -- resulting chemical scores and rankings.
  <200501.pdf, 4/2002, 561 Kb>       About PDFs
  

As a supplement to the article, the authors are making a portion of the data available here for viewing and download:

• Table 3: Chemical, uncertainty, and composite scores.
• Table 4: Relative rankings by chemical score and composite score.

Reference:
RR Mitchell, CL Summer, SA Blonde, DM Bush, GK Hurlburt, EM Snyder, and JP Giesy. SCRAM: A scoring and ranking system for persistent, bioaccumulative, and toxic substances for the North American Great Lakes -- resulting chemical scores and rankings. Human Ecol Risk Assess. 8(3):537-557. April 2002.

Last Updated: 17 April 2002

URL: http://www.epa.gov/toxteam/pbtrept/pbtreport.htm