Phase III GIS-based Sediment Quality Database for the St. Louis … · 2012. 10. 4. · Phase III GIS-based Sediment Quality Database for the St. Louis River Area of Concern (AOC)—

Phase III GIS-based Sediment Quality Database for the St. Louis River Area of Concern (AOC)—

Minnesota Focus

Overview of Sediment Quality Conditions in the St. Louis River AOC

Final Report

Submitted to:

Karla Sundberg Minnesota Department of Natural Resources Minnesota’s Lake Superior Coastal Program

1568 Highway 2 Two Harbors, MN 55616

Submitted by:

Judy L. Crane

Minnesota Pollution Control Agency Environmental Analysis and Outcomes Division

520 Lafayette Road North St. Paul, MN 55155-4194

Email: [email protected]

December 2005

MPCA Document Number: tdr-fg05-04

mailto:[email protected]

Phase III Sediment Quality Report

DISCLAIMER

Minnesota’s Lake Superior Coastal Program

Name of Project: Phase III GIS-based Sediment Quality Database for the St. Louis River Area of Concern (AOC)—Minnesota Focus

Date of Publication: December 2005

Project No. 306-28-06______

Contract No. A68008_______

This project was funded in part under the Coastal Zone Management Act, by the National Oceanic and Atmospheric Administration’s (NOAA’s) Office of Ocean and Coastal Resource Management, in cooperation with Minnesota’s Lake Superior Coastal Program. In addition, the Minnesota Pollution Control Agency (MPCA) contributed over 50% of state matching funds to this project. This report has been reviewed and approved by Minnesota’s Lake Superior Coastal Program. Mention of trade names or commercial products does not constitute endorsement or recommendation for use by the MPCA, Minnesota’s Lake Superior Coastal Program, or by NOAA.

MPCA Document Number: tdr-fg05-04 ii


TABLE OF CONTENTS

Page Disclaimer ...................................................................................................................................ii Table of Contents..........................................................................................................................iii List of Figures ...............................................................................................................................iv List of Tables ................................................................................................................................v List of Acronyms and Abbreviations............................................................................................vi Acknowledgments.........................................................................................................................viii 1.0 Introduction............................................................................................................................1 2.0 Methods..................................................................................................................................4 2.1 Fall 2005 Stakeholder Meetings .................................................................................4 2.2 Data Analysis ..............................................................................................................4 2.2.1 Data Queries from Query Manager-Compatible Database Files ......................4 2.2.2 Data Queries from Phase III MS™ Access 2000 Sediment Quality Database.5 2.2.3 Processing of Data ............................................................................................5 3.0 Results....................................................................................................................................8 3.1 Stakeholder Meetings..................................................................................................8 3.2 Sediment Chemistry Results .......................................................................................9 3.3 Sediment Toxicity Test Results ..................................................................................11 4.0 Discussion ..............................................................................................................................12 5.0 Recommendations for Additional Data Analysis ..................................................................16 References .....................................................................................................................................18 Appendix A: Copy of Level I and Level II Sediment Quality Targets for the Minnesota Side of

the St. Louis River AOC Appendix B: Descriptive Statistics of Sediment Chemistry Parameters with Corresponding

Sediment Quality Targets, as well as Total Organic Carbon Appendix C: Results of Mann-Whitney Rank Sum Tests Between Surface and Subsurface

Sediments for the List of Chemicals with SQTs, as well as TOC Appendix D: Statistical Analysis of PAH Source Ratios

MPCA Document Number: tdr-fg05-04 iii


LIST OF FIGURES

Figure Page 1 Location of sediment stations in the St. Louis River AOC that were included in the Phase III GIS-based sediment quality database. ...............................24

2 Location of sediment stations in the Duluth-Superior Harbor that were included in the Phase III GIS-based sediment quality database. ...............................25 3 Example of GIS data for contaminated areas in the lower St. Louis River AOC..................................................................................................26 4 Example of GIS data for geographic features in the lower St. Louis River AOC..................................................................................................27 5 Example of including too much GIS data on a map of the St. Louis River AOC..................................................................................................28 6 Example of a more reasonable selection of GIS data to view of the St. Louis River watershed ..........................................................................................29 7 Example of GIS data for land uses in the Duluth-Superior Harbor...........................30 8 Example of GIS data for recreation uses in the lower St. Louis River AOC ............31 9 Example of GIS data included in the U.S. EPA Sensitivity Atlas for the lower St. Louis River AOC ............................................................................32 10 Example of GIS data for water quality parameters in the lower St. Louis River AOC..................................................................................................33 11 Example of GIS data for water uses in the lower St. Louis River AOC....................34 12 Example of a black and white map option for displaying GIS data in the lower St. Louis River AOC..................................................................................35 13 Example linkage of the Phase III MS™ Access 2000 database query results for mean PEC-Q ranges with the Phase II ArcMap 9 map document for Ecological Features ..............................................................................................36 14 Display of mean PEC-Q ranges in the lower St. Louis River from Cloquet, MN to the St. Louis River Interlake/Duluth Tar Superfund site.................37

MPCA Document Number: tdr-fg05-04 iv


LIST OF TABLES

Table Page 1 Summary of Data Sets Incorporated into the Phase III GIS-based Sediment Quality Database for the St. Louis River AOC..........................................................38 2 Statistical Summary of Sediment Chemistry Values in the Surficial Sediments of the St. Louis River AOC......................................................................42 3 Statistical Summary of Sediment Chemistry Values in Subsurface Sediments of the St. Louis River AOC......................................................................46 4 Determination of Statistical Significance Between Median Chemical Values in Surface and Subsurface Sediments from the St. Louis River AOC ..........49 5 Results of Regression Analyses of Chemical Parameters Included in the Phase III Sediment Quality Database for the St. Louis River AOC ....................52 6 Comparison of PAH Source Ratios for Various Sediment Depth Intervals in the St. Louis River AOC........................................................................................53 7 Comparison of Surficial Sediment Chemistry Data to Four Types of SQTs Used by the MPCA......................................................................................54 8 Percentage of Samples in Three Surficial Depth Intervals with Low, Moderate, or High Mean PEC-Q Ranges ..................................................................54 9 Compilation of Sediment Toxicity Test Results from the Phase III GIS-based Sediment Quality Database for the St. Louis River AOC........................55

MPCA Document Number: tdr-fg05-04 v


LIST OF ACRONYMS AND ABBREVIATIONS AOC Area of Concern As Arsenic AVS Acid Volatile Sulfides BC British Columbia BHC Hexachlorocyclohexane CAC Citizens Action Committee CCME Canadian Council of Ministers of the Environment CN Cyanide DDD Metabolite of DDT DDE Metabolite of DDT DDT Dichloro-diphenyl-trichloroethane DRO Diesel Range Organics DW Dry Weight F/P Fluoranthene to Pyrene Ratio GIS Geographic Information System HCB Hexachlorobenzene Hg Mercury ID Identification IJC International Joint Commission IL Illinois IT International Technology MB Manitoba MLE Maximum Likelihood Estimation MN Minnesota MPCA Minnesota Pollution Control Agency MS™ Microsoft™ N Number ND Nondetect NJ New Jersey NOAA National Oceanic and Atmospheric Administration NYSDEC New York State Department of Environmental Conservation OC Organic Carbon OCS Octachlorostyrene PA Pennsylvania P/A Phenanthrene to Anthracene Ratio PAH Polycyclic Aromatic Hydrocarbon Pb Lead PCB Polychlorinated Biphenyl PEC Probable Effect Concentration PEC-Q Probable Effect Concentration Quotient QAPP Quality Assurance Project Plan RAP Remedial Action Plan R-EMAP Regional Environmental Monitoring and Assessment Program

MPCA Document Number: tdr-fg05-04 vi


LIST OF ACRONYMS AND ABBREVIATIONS (cont.)

SEM Simultaneously Extractable Metals SLRIDT St. Louis River Interlake/Duluth Tar SQG Sediment Quality Guideline SQT Sediment Quality Target TCDD Tetrachlorodibenzo-p-dioxins TCDF Tetrachlorodibenzofurans TEC Threshold Effect Concentration TMDL Total Maximum Daily Load TOC Total Organic Carbon U.S. EPA United States Environmental Protection Agency USFWS United States Fish and Wildlife Service USS United States Steel UV Ultraviolet VOCs Volatile Organic Compounds WA Washington WDNR Wisconsin Department of Natural Resources WI Wisconsin WLSSD Western Lake Superior Sanitary District

MPCA Document Number: tdr-fg05-04 vii


ACKNOWLEDGMENTS

The Phase III GIS-based sediment quality database would not have been possible without the assistance and input of those people who helped to bring the Phase I and Phase II sediment quality databases and associated GIS products to fruition. These individuals were previously acknowledged in the Phase II report documentation (Smorong and Crane 2004; Smorong et al. 2004a,b). For the Phase III project, individuals who provided sediment quality data sets for this project were acknowledged in the Addendums to the Phase II Help Section for Database Users and the Phase II Technical Documentation (Crane and Myre 2005a,b). For this report, the input of Minnesota Pollution Control Agency (MPCA) staff and stakeholders who participated in stakeholder meetings held in St. Paul, MN on October 31, 2005 and in Duluth, MN on November 1, 2005 was useful for prioritizing information to include in this report. The Phase III database products used in preparation of this report were completed by Exa Data & Mapping Services, Inc. and their subcontractors. The technical team members for the database component of this project were as follows: MPCA: Judy Crane Exa Data & Mapping Services, Inc.: Peggy Myre, Stacy Larsen Premier Environmental Services, Inc.: Corinne Severn, Laurel Menoche Searay Environmental: Raymond Valente TerraStat Consulting Group: Lorraine Read The draft report was reviewed by Karla Sundberg (Minnesota’s Lake Superior Coastal Program), the Grants Specialist for this project. Tricia Ryan and Pat Collins were the successive program managers to Minnesota’s Lake Superior Coastal Program. The funding acknowledgments for this project are provided in the Disclaimer of this report. Jennifer Holstad, Jan Eckart, and Julie Lendway (MPCA) provided word processing and report production support. Todd Biewen (MPCA) provided helpful supervisory assistance for this project, particularly with securing state match money for use in this project. This report should be cited as: Crane, J.L. 2005. Phase III GIS-based sediment quality database for the St. Louis River Area of

Concern (AOC)—Minnesota focus. Overview of sediment quality conditions in the St. Louis River AOC. Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN. MPCA Document Number tdr-fg05-04.

MPCA Document Number: tdr-fg05-04 viii


CHAPTER 1

INTRODUCTION The lower St. Louis River provides an important coastal resource to western Lake Superior. In particular, this transboundary waterway provides critical habitat to invertebrate, fish, and waterfowl species, and it also provides an economic venue for Great Lakes shipping and business in the Duluth-Superior Harbor. The lower estuary and harbor are bisected by the Minnesota-Wisconsin state line (Figure 1). Contaminated sediments have contributed to several use impairments in the lower St. Louis River and were a factor in the International Joint Commission’s (IJC) decision to designate the lower St. Louis River as one of 43 Areas of Concern (AOCs) in the Great Lakes basin (IJC 1989). As such, contaminated sediments are an important nonpoint source of pollution to the St. Louis River AOC, especially in the Duluth-Superior Harbor. Consequently, several sediment quality and fish tissue studies have been conducted in the lower St. Louis River AOC, particularly since 1990, to delineate the extent and magnitude of contaminants of potential concern and to assess the potential for ecological effects (Figure 1). As part of the remedial action plan (RAP) process for the St. Louis River AOC, the Minnesota Pollution Control Agency (MPCA) and the Wisconsin Department of Natural Resources (WDNR), along with citizen and technical stakeholders, identified the need to compile sediment quality data collected from the AOC in a database format for mapping and evaluation purposes. This need was the basis for securing funding to conduct earlier phases of the Geographic Information System (GIS)-based sediment quality database, as well as this Phase III project. A subsequent Phase IV database project will be completed by June 30, 2006. These updates of the database will support the assessment, preservation, and restoration of the St. Louis River AOC. The primary applications of the Phase III GIS-based sediment quality databases are to: allow users to access sediment quality data for the AOC; allow users to view the data spatially, along with GIS-watershed data (from the Phase II project); to assist agencies and stakeholders with implementing the three-phase sediment strategy for the Stage II St. Louis River RAP (MPCA/WDNR 1995), along with newer sediment management efforts; to further the goals of the Lake Superior Lakewide Management Plan by tracking the occurrence of critical pollutants; and to allow analysis of sediment quality data in conjunction with the detailed habitat evaluation and mapping accomplished through the St. Louis River Habitat Plan (St. Louis River CAC 2002). Much of the GIS data from the Habitat Plan have been incorporated into themed GIS map documents prepared for earlier phases of the database project (Smorong et al. 2004a). Sediment quality issues in the Lower St. Louis River AOC are of interest to local and state agencies in Minnesota and Wisconsin, as well as to federal agencies, tribal groups, responsible parties, and other concerned stakeholders. The St. Louis River RAP (MPCA/WDNR 1992, 1995), as well as a number of sediment investigations, have identified several areas of elevated sediment contamination (i.e., hot spot sites) within the Minnesota side of the lower estuary, including the:

MPCA Document Number: tdr-fg05-04 1


• St. Louis River Interlake/Duluth Tar (SLRIDT) Superfund site; • USS Superfund site; • Minnesota Slip; • Slip C; • Embayment encompassing the old Duluth sewage treatment plant near the Western Lake

Superior Sanitary District (WLSSD) and the confluence with Coffee and Miller Creeks; • Embayment encompassing the M.L. Hibbard plant/DSD No. 2, north of Grassy Point; • Bay near the DM&IR taconite storage facility; and • Dakota Pier.

Contaminated hot spot sites within the Wisconsin side of the St. Louis River AOC include the:

• Kopper’s Industries Inc. site; • Howard’s Bay; and • Area near the City of Superior wastewater treatment plant.

Final sediment remediation of the Hog Island Inlet and Newton Creek site in Superior, WI was completed during November 2005 through the use of Great Lakes Legacy Act funding (http://www.epa.gov/glnpo/sediment/legacy/hogisland/index.html). Sediment remediation of the SLRIDT Superfund site will begin in 2006 by the responsible parties. Great Lakes Legacy Act funding is currently being sought by the MPCA to remediate Minnesota Slip. The MPCA, with the contractual assistance of Exa Data & Mapping Services, Inc. and their subcontractors, have completed Phase III of a GIS-based sediment quality database for the St. Louis River AOC (Crane 2005; Crane and Myre 2005a,b). This phase of the database focused on the Minnesota side of the AOC (Table 1). The database includes data on sediment chemistry, sediment toxicity, benthic invertebrate community structure, physical variables (e.g., particle size), and sediment bioaccumulation and tissue residue parameters. A history of the development of earlier phases of this sediment quality database are provided in the Phase II project reports (Smorong and Crane 2004; Smorong et al. 2004a,b). The following deliverables were produced for the Phase III project:

• Phase III Quality Assurance Project Plan (QAPP; Crane 2005): http://www.pca.state.mn.us/publications/tdr-fg04-05.pdf;

• Phase III Microsoft™ (MS™) Access 2000 Screening Database (for internal use only

since it contains the screening reviews of sediment quality data sets considered for inclusion in each phase of the database);

• Phase III MS™ Access 2000 GIS-based Sediment Quality Database for the St. Louis

River AOC: ftp://files.pca.state.mn.us/pub/sedimentDB/STLR_SED_DB_PH3_FinEdited.zip (this file will be available until Spring 2006, when it will be replaced by the corresponding Phase IV database file);


http://www.pca.state.mn.us/publications/tdr-fg04-05.pdf

ftp://files.pca.state.mn.us/pub/sedimentDB/STLR_SED_DB_PH3_FinEdited.zip


• Phase III MS™ Access ‘97 GIS-based Sediment Quality Database for the St. Louis River AOC (for those users lacking MS™ Access 2000 software): ftp://files.pca.state.mn.us/pub/sedimentDB/STLR_SED_DB_PH3_97.zip (this file will be available until Spring 2006, when it will be replaced by the corresponding Phase IV database file);

• Phase III Query Manager-compatible Database Files for the St. Louis River AOC (to be

used with the National Oceanic and Atmospheric Administration’s [NOAA’s] free Query Manager software which is available at: http://response.restoration.noaa.gov/type_subtopic_entry.php?RECORD_KEY(entry_subtopic_type)=entry_id,subtopic_id,type_id&entry_id(entry_subtopic_type)=378&subtopic_id(entry_subtopic_type)=5&type_id(entry_subtopic_type)=3): ftp://files.pca.state.mn.us/pub/sedimentDB/STLR_SEDB_QM_Final.zip (this file will be available until Spring 2006, when it will be replaced by the corresponding Phase IV database file);

• Phase III Addendum to Phase II Help Section for Database Users (Crane and Myre

2005a): http://www.pca.state.mn.us/publications/tdr-fg05-01.pdf;

• Phase III Addendum to Phase II Technical Documentation (Crane and Myre 2005b): to be posted at http://www.pca.state.mn.us/publications/tdr-fg05-02.pdf by the end of December 2005;

• Phase III Fall 2005 Stakeholder Presentation by Judy Crane (MPCA):

http://www.pca.state.mn.us/water/sediments/slr-p3stakeholder-f05.pdf; and

• Phase III Overview of Sediment Quality Conditions in the St. Louis River AOC (i.e., this report): to be posted at http://www.pca.state.mn.us/publications/tdr-fg05-04.pdf by the end of December 2005.

The Phase III QAPP was an extra task not originally included in the MPCA’s interagency grant agreement with Minnesota’s Lake Superior Coastal Program. The MPCA was required to provide a 50% state match for this project, which amounted to 900 hours of Dr. Crane’s work time through her salary, fringe, and indirect expenses. The 50% state match requirement was met by early August 2005, and the MPCA has contributed over 320 hours of additional state match time to this project through Dr. Crane’s time. As such, preparation of this report was reduced in scope in order to complete the data analysis and report preparation within a 3 week period of time. This report focused on sediment chemistry and toxicity testing results in the Phase III GIS-based sediment quality database for the St. Louis River AOC, with additional emphasis on Minnesota sites and issues. Interested Agency staff and stakeholders are encouraged to explore the easy-to-use query options in the Phase III Query Manager-compatible database files, as well as the results of a small number of queries available in the MS™ Access version of the database (as well as queries of their own design) to learn more about sediment quality conditions in the St. Louis River AOC.


ftp://files.pca.state.mn.us/pub/sedimentDB/STLR_SED_DB_PH3_97.zip

ftp://files.pca.state.mn.us/pub/sedimentDB/STLR_SEDB_QM_Final.zip



http://www.pca.state.mn.us/water/sediments/slr-p3stakeholder-f05.pdf



CHAPTER 2

METHODS 2.1 FALL 2005 STAKEHOLDER MEETINGS MPCA staff and stakeholders were invited to attend two stakeholders meetings during the fall of 2005 in which an overview presentation of the Phase III database project was given and input was sought on the development of this sediment quality report. Five people participated in the meeting held on October 31, 2005 in St. Paul, MN, and approximately 20 people participated in a similar meeting held on November 1, 2005 in Duluth, MN. A copy of the presentation given at these meetings is available at: http://www.pca.state.mn.us/water/sediments/slr-p3stakeholder-f05.pdf. Several GIS maps of watershed data for the St. Louis River AOC were prepared based on the Phase II ArcMap map documents. Due to time constraints, the map screen images were copied from ArcMap 9 and pasted into the PowerPoint presentation, rather than preparing formalized maps with a legend, scale, etc. In addition, the results of a MS™ Access 2000 query of concentration ranges of mean probable effect concentration quotients (PEC-Qs) were plotted on the Phase II ArcMap map document for ecological areas based on the instructions in Crane and Myre (2005b). 2.2 DATA ANALYSIS 2.2.1 Data Queries from Query Manager-Compatible Database Files Sediment quality data were queried from the Phase III Query Manager-compatible database files using the standard drop down menu of queries developed by NOAA as described in Appendix C of Crane and Myre (2005b). NOAA’s operational definition of surface sediments is that the upper depth segment must be 0 cm, and the lower depth segment must be <30.5 cm. Query Manager users are provided two options for selecting subsurface sediment chemistry data. The first subsurface option, which was used in this study, includes all sediment chemistry data located below the surface (i.e., >0 cm) or within a core length greater than 30.5 cm; under this scenario, a 2 – 4 cm or 0 – 40 cm core depth interval would be included in the subsurface category. Query Manager also provides the option of including all surface and subsurface results in the subsurface query. The following queries were used for this study:

• Sediment chemistry (surface): o Concentration ranges, o Number of chemicals in a sample exceeding selected guidelines (chemicals with

below detection limit values were treated as one-half the detection limit), and o Select multiple chemicals (chemicals with below detection limit values were

treated as one-half the detection limit). Chemicals were selected from the list of chemicals with corresponding Level I and Level II sediment quality targets (SQTs; Appendix A) adopted for use in Minnesota (Crane et al. 2000, 2002a).





• Sediment chemistry (subsurface):

o Concentration ranges, and o Select multiple chemicals (chemicals with below detection limit values were

treated as one-half the detection limit). Chemicals were selected from the list of chemicals with corresponding Level I and Level II SQTs.

• Sediment bioassay:

o Test endpoints, toxic/nontoxic samples. 2.2.2 Data Queries from Phase III MS™ Access 2000 Sediment Quality Database Two queries that were included in a list of queries in the Phase III MS™ Access 2000 sediment quality database (Crane and Myre 2005a) were utilized in this study. The queries were:

• Polycyclic aromatic hydrocarbon (PAH) source ratios of phenanthrene/anthracene (P/A) and fluoranthene/pyrene (F/P) were calculated for the following depth intervals (chemicals with below detection limit values were treated as one-half the detection limit):

o ≥0 to ≤5 cm, o ≥0 to ≤15 cm, o ≥0 to ≤30 cm, o ≥15 to ≤30 cm, o ≥30 to ≤45 cm, and o ≥30 cm.

• Mean PEC-Q ranges of <0.1 (low), ≥0.1 to ≤0.6 (moderate), and >0.6 (high) were sorted for the following depth intervals (chemicals with below detection limit values were treated as one-half the detection limit, unless the value exceeded the corresponding Level II SQT in which case the value was excluded):

o ≥0 to ≤5 cm, o ≥0 to ≤15 cm, and o ≥0 to ≤30 cm.

2.2.3 Processing of Data The results of most of the queries were exported to MS™ Excel for further data processing. However, the results of the concentration ranges of surface and subsurface chemicals were displayed for single chemicals, and the summary data of the number of samples, arithmetic mean, minimum, and maximum values were recorded by hand on a summary sheet. The values on the summary sheet were hand-entered into an Excel worksheet, and the values in Excel were compared against the hard copy to determine if any transcription errors had occurred. Since the Query Manager option for querying concentration ranges does not provide an option for treating nondetect data, the results of this query were not used further. Instead, the sediment chemistry (surface and subsurface) data from the multiple chemical query options were used for further data analysis.



The results from the surface and subsurface multiple chemical queries were processed further in Excel by:

• Copying the multiple chemical query results to multiple worksheets within Excel so that the worksheets that would be used for regression analyses were trimmed to the two chemicals of interest. In addition, other worksheets were created for each chemical. Each worksheet was then sorted to remove missing values. A small number of duplicate stations and a few discrepancies in total organic carbon (TOC) values were discovered during this process.

o For the surface chemistry results, all of the stations for Study ID 03 were removed as being duplicate stations of Study ID 34 from the SLRIDT Superfund site. In addition, stations C-011 (1A) and C-002 (1A) from Study ID 37 were deleted as suspected duplicate stations of Study ID 34. The TOC value for Station ID SM-004 (1A) from Study ID 34 was changed from 132% to -8. These corrections will be made to the Phase IV GIS-based sediment quality database by February 2006.

o For the subsurface chemistry results, duplicate stations from Study ID 03 were removed, and the TOC value for Station ID DC-245 (AA) from Study ID 42 was changed from 130% to -8.

• For the single chemicals from the surface and subsurface queries, summary statistics were calculated in Excel for the following parameters: count, average, standard deviation, minimum, median, maximum, and geometric mean.

• The summary statistics for single chemicals were transcribed by hand to a hard copy table, one for surface sediments and the other for subsurface sediments. The resulting tables were entered into an Excel file, and the results of the electronic copies were proofed against the hard copies to check for any transcription errors. The median values were next compared to the corresponding Level I and Level II SQT values (Appendix A).

• The surface and subsurface data for each chemical were copied to a SigmaStat® 3.1 worksheet so that a new worksheet was created for each chemical (e.g., lead, zinc, anthracene). The descriptive statistics feature of SigmaStat® 3.1 was used, in particular, to obtain 25% and 75% percentiles, as well as measures of skewness, kurtosis, and normality. Skewness is a measure of how symmetrically the observed values are distributed about the mean. Kurtosis is a measure of how peaked or flat the distributions of observed values are, compared to a normal distribution. A nonparametric Mann-Whitney Rank Sum Test of the median values for each pair of surface and subsurface chemicals was run using SigmaStat® 3.1 to determine if the chemical concentrations were significantly different (p = 0.05).

• Paired surface chemistry data were copied from Excel to SigmaPlot® 9.0 where linear regressions of chemical parameters (e.g., lead versus zinc) were run. Ninety-five percent confidence intervals and prediction intervals were also plotted on the resulting graphs. The linear regression results of slope, intercept, and r2 value were obtained from SigmaPlot® 9.0 for each graph. The regression results were compiled in a table for all of the chemical pairs examined in this study.

For the Query Manager 2.56 query for determining the number of chemicals in a sample exceeding selected guidelines, queries were run for the Level I and Level II SQTs for the protection of benthic organisms, the SQT for the protection of wildlife, and the SQT for the protection of human health. All of the SQTs were adopted from other jurisdictions for use in



Minnesota (Crane et al. 2000, 2002a). The summary results for each query were recorded on a hard copy table that was entered into an Excel table. The electronic table was then proofed against the hard copy table to check for any transcription errors. Sediment bioassay test endpoints for toxic/nontoxic samples were determined from separate queries in Query Manager 2.56 for toxicity tests that had at least 20 test results. The results of each query were exported to Excel. Duplicate station results, as described previously, were deleted from the results for10-d survival and growth in the amphipod, Hyalella azteca, as well as for 10-d survival in the midge, Chironomus tentans. The number of samples with toxic/effect or nontoxic/no effect results were determined, and the percentage of samples in each category was calculated. The results were recorded on a hard copy table, which were subsequently entered into Excel and proofed for any transcription errors. The MS™ Access 2000 query for PAH source ratios was exported to Excel. Duplicate values, as previously described, were removed from the Excel file. The PAH source ratios for each depth interval were copied into separate worksheets in Excel. Summary statistics for each P/A and F/P column of data were calculated for the following parameters: count, average, standard deviation, minimum, median, maximum, and geometric mean. The number of samples and median values for each P/A and F/P source ratio for each depth interval were compiled, entered into Excel, and proofed against the hard copy table. Next, a Mann-Whitney Rank Sum test was run in SigmaStat® 3.1 between the P/A groups of 0 - 30 cm and >30 cm, as well as between the F/P groups of the same depth intervals. The MS™ Access 2000 query for mean PEC-Q ranges for three different surface depth intervals was exported to Excel. Duplicate values, as previously described, were removed from the Excel file. The mean PEC-Q results were sorted for each depth interval in separate worksheets. Samples with no results were deleted, and the data were summarized for the total number of samples and percentage of low, medium, and high mean PEC-Q ranges in each depth interval. The results were transcribed to a hard copy table, which was then entered into Excel and proofed for transcription errors.



CHAPTER 3

RESULTS

3.1 STAKEHOLDER MEETINGS A strong consensus did not emerge from the stakeholder group regarding the type of sediment quality data they would like to have queried from the database to include in this report, or the type of GIS data they would like to have presented on maps with sediment quality data. Some suggestions from the attendees of the St. Paul and Duluth meetings included:

• Matching sediment chemistry data with ecological functions; • Linking future land uses in the harbor with current contaminant data; • Mapping physical parameters like particle size (e.g., to determine sites with suitable

capping material for remediation options); • Providing guidance on how to handle nondetect data; • Mapping the distribution of mercury throughout the AOC, particularly for the benefit of

the total maximum daily load (TMDL) workgroup for the St. Louis River; and • Providing a holistic report on sediment quality conditions for the entire AOC, not just

pertaining to the Minnesota side of the AOC. Due to the time constraints of this project, it was not possible to address each of these suggestions. However, since the Phase III database includes all of the sediment quality data previously included in the Phase I and Phase II databases, the entire data set was used in the analyses presented in the following sections. This data set provides good spatial coverage of the Minnesota and Wisconsin sides of the St. Louis River AOC (Figures 1 and 2). In addition, sediment chemistry data were compared to corresponding SQTs adopted by the MPCA for use in the St. Louis River AOC (Crane et al. 2000, 2002a). The Wisconsin Department of Natural Resources (WDNR) has adopted some similar consensus-based sediment quality guidelines (SQGs), as well as other SQGs for additional chemicals (WDNR 2003). The Wisconsin SQGs are available in the MS™ Access 2000 and Query Manager-compatible databases for those stakeholders interested in comparing sediment chemistry data to those values. The GIS maps that were included in the fall stakeholder presentations are included in Figures 1 through 14 to demonstrate the type of GIS data that are available for users to use in the following themed map documents.

• Ecological areas (Figures 1 and 2); • Contaminated areas (Figure 3); • Geographic features (Figure 4) • Hydrology features (Figures 5 and 6); • Land uses (Figure 7); • Recreation uses (Figure 8); • U.S. Environmental Protection Agency (EPA) Inland Sensitivity Atlas (Figure 9);



• Water quality features (Figure 10); • Water uses (Figure 11); and • Black and white map option (Figure 12).

In addition, Figures 13 and 14 demonstrate how an integrative technique for compiling data for a mixture of contaminants (i.e., mean PEC-Qs) can be plotted on maps containing ecological GIS data. 3.2 SEDIMENT CHEMISTRY RESULTS The results of the sediment chemistry analyses are provided in the following tables:

• Table 2: Statistical Summary of Sediment Chemistry Values in the Surficial Sediments of the St. Louis River AOC;

• Table 3: Statistical Summary of Sediment Chemistry Values in Subsurface Sediments of the St. Louis River AOC;

• Table 4: Determination of Statistical Significance Between Median Chemical Values in Surface and Subsurface Sediments from the St. Louis River AOC;

• Table 5: Results of Regression Analyses of Chemical Parameters Included in the Phase III Sediment Quality Database for the St. Louis River AOC;

• Table 6: Comparison of PAH Source Ratios for Various Sediment Depth Intervals in the St. Louis River AOC;

• Table 7: Comparison of Surficial Sediment Chemistry Data to Four Types of SQTs Used by the MPCA; and

• Table 8: Percentage of Samples in Three Surficial Depth Intervals with Low, Moderate, or High Mean PEC-Q Ranges.

Due to the non-normal distribution and skewness of the sediment chemistry data (Appendix A), comparisons should be made with the median (i.e., “middle) values instead of the arithmetic average values. The geometric mean is also appropriate to use when there is a wide range of concentrations in the data set (e.g., for PAH compounds in this data set). However, the amount of data in the Excel spreadsheets exceeded the capabilities of this software to calculate geometric means in most cases. When geometric means could be calculated, the values were usually similar to the median values. Some of the major results of the sediment chemistry results include:

• Descriptive Statistics o Zinc was the only metal in this analysis for which the median value exceeded the

Level I SQT in the surface sediments (Table 2). Both lead and zinc exceeded the Level I SQT in the subsurface sediments (Table 3).

o All of the median concentrations of PAHs in Tables 2 and 3 exceeded the corresponding Level I SQTs in the surface and subsurface sediments. In addition, 2-methylnaphthalene exceeded the Level II SQT in both the surface and



subsurface sediments (Tables 2 and 3). The median concentrations of acenaphthene, naphthalene, phenanthrene, and pyrene also exceeded the corresponding Level II SQT values (Table 3).

o Polychlorinated biphenyl (PCB) results were available in the database for four different calculation methods. Three of these calculation methods resulted in exceedances of the median PCB concentrations compared to the Level I SQT value for total PCBs for both the surface and subsurface sediments (Tables 2 and 3).

o Median concentrations of seven pesticides exceeded the corresponding Level I SQT values in the surface sediments (Table 2), whereas five pesticides exceeded the corresponding Level I SQT values in the subsurface sediments (Table 3).

o The median of the mean PEC-Q values exceeded the Level I SQT value in both the surface and subsurface sediments (Tables 2 and 3).

o The median concentrations in the surface sediments were statistically less (p<0.05) than in the subsurface sediments for the following chemicals: arsenic, acenaphthene, acenaphthylene, anthracene, fluorine, naphthalene, phenanthrene, total PAHs13 (i.e., sum of the 13 low molecular weight and high molecular weight PAHs), and heptachlor epoxide.

o The median concentrations in the surface sediments were statistically greater (p<0.05) than in the subsurface sediments for the following chemicals: total PCBs - reported (can be recalculated), total PCBs – reported only, total congener PCBs, total chlordane (excluding high nondetect values), total chlordane – reported only, dieldrin, sum DDD (excluding high nondetect values), sum DDE (excluding high nondetect values), sum DDT (excluding high nondetect values), total DDT (excluding high nondetect values), endrin, and lindane.

• Regression Analyses o On an AOC-wide basis for the surface sediments, there was no correlation

(r2<0.05) between total PAHs13 and TOC (either all TOC data, the 0.2 - 12% TOC fraction, or the 0.5 - 10% TOC fraction).

o On an AOC-wide basis, there was no correlation (r2<0.006) between total PAHs13 and either mercury, lead, and zinc in the surface sediments. For example, some samples had high mercury and low total PAH13 concentrations and vice versa.

o On an AOC-wide basis, there were weak correlations between zinc, mercury, and lead with TOC in the surface sediments.

o On an AOC-wide basis for the surface sediments, the best correlation in the data analysis was between mercury and zinc (r2 = 0.570). However, there were some samples that had high mercury and low zinc concentrations and vice versa.

• PAH Source Ratios o The median P/A source ratios ranged from 2.1 to 4.6 in the six depth intervals

selected for this analysis, whereas the median F/P ratios ranged from 1.2 to 1.3. o The median P/A source ratio in the 0 – 30 cm depth interval was statistically less

(p = 0.002) than in the >30 cm depth interval. There was no statistical difference (p = 0.134) between the median F/P source ratios for these two depth intervals.



• Comparisons to Sediment Quality Targets o For those sediment samples in which sediment chemistry was analyzed, the

samples exceeded the Level I and Level II SQTs an average 9 and 4 times, respectively (Table 7).

o A lower number of average exceedances were observed based on the SQTs for wildlife and human health (Table 7).

• Mean PEC-Q Risk Categories

o Sediment samples from the 0 – 5 cm depth interval had a greater percentage of low risk samples (based on the mean PEC-Q values) than the 0 – 15 cm and 0 – 30 cm depth intervals (Table 8). Risk was in relation to the potential for harmful effects to occur in benthic invertebrates.

o Sediment samples from the 0 – 30 cm depth interval had the highest percentage of high risk samples compared to the other two depth intervals (Table 8).

3.3 SEDIMENT TOXICITY TEST RESULTS Sufficient data (i.e., at least 20 samples) were available to query the sediment toxicity test results for amphipods (H. azteca), midges (C. tentans), oligochaetes (Lumbriculs variegatus), and cattails (Typha latifolia) as presented in Table 9. The greatest number of results was available for the 10-d survival tests of both H. azteca and C. tentans (Table 9). Although these tests had a similar percentage of toxic and nontoxic samples, the effects did not necessarily correspond to the same samples. Thus, sediments that were toxic to H. azteca within a 10-d exposure period, were not always toxic to C. tentans. The chronic toxicity tests (i.e., 10-d growth or longer term tests) generally resulted in a higher percentage of toxicity or significant effects to a particular endpoint (e.g., growth, emergence) than the acute, 10-d survival tests (Table 9).



CHAPTER 4

DISCUSSION

The sediment quality data in the Phase III GIS-based sediment quality data includes data from 55 studies that had varying data quality objectives. Many of these studies took place at either known or suspected contaminated areas. Only one large randomly designed study, a Regional Environmental Monitoring and Assessment Program (R-EMAP) project (Breneman et al. 2000; Crane et al. 2005), has been conducted within the entire St. Louis River AOC. As such, the sediment quality data are skewed toward contaminated areas within the AOC. In particular, the SLRIDT Superfund site data set comprises the largest sediment quality data set for a hot spot area in the database. As some hot spot sites are remediated within the lower St. Louis River AOC and post-remediation data are collected, it would be useful to evaluate the pre- and post-remediation data to obtain a measure of improvement in sediment quality conditions within this AOC. PAHs are widespread chemicals of potential concern in the lower St. Louis River AOC, particularly starting at the USS Superfund site and progressing downstream to the rest of the estuary. The mean P/A and F/P ratios are indicative of pyrogenic (combustion) sources of PAHs to the AOC (i.e., P/A ratio <10 and F/P ratio >1.0; Budzinski et al. 1997). Similar results were found in the random sampling done in the lower estuary during 1995 for the R-EMAP project (Crane et al. 2005). The lack of correlations between total PAHs13 and TOC, lead, mercury, and zinc may be due to multiple sources of PAHs and metals to the lower St. Louis River watershed. In contrast, r2

values exceeding 0.76 were observed in Slip C, Duluth Harbor, for total PAHs and TOC (<10% TOC), lead, and mercury (Crane 1999). Slip C is a relatively small contaminated site that appears to have limited sources of contaminants. The findings of this study, though, were supportive of similar regression analyses done for the R-EMAP project (Crane et al. 2005), except that lead, mercury, and zinc accounted for additional variance in total PAHs12 from the R-EMAP study (Crane et al. 2005). The R-EMAP study also displayed stronger correlations between the metals (mercury, lead, zinc; Crane et al. 2005) than in the analysis of the entire surface sediment chemistry data set from the Phase III database. The R-EMAP study only included simultaneously extractable metals for cadmium, copper, lead, nickel, and zinc that were assumed to be equivalent to conventional metal concentrations when used with other metals data in the database; this assumption was based on studies done elsewhere (Crane et al. 2000). On an AOC-wide basis, mixing the metals data for the R-EMAP study with conventional metals data from other sites may introduce more uncertainty into the regression relationships. The distribution of sediment contaminants can be very heterogeneous within a short distance in the St. Louis River AOC, such as between field replicates in Minnesota Slip (Crane et al. 2002b). In addition, sediment cores from the USS Superfund site that were split and analyzed by different labs have recently yielded widely different sediment chemistry results that were not attributable to any quality assurance/quality control issues; these split samples were added to the Phase III database as separate studies (Study ID’s 54 and 55; Crane and Myre 2005b). The heterogeneity



of sediments at some of the contaminated sites can result in increased difficulty with analytical laboratories being able to adequately extract the chemicals of interest from the sediment matrix. Database users are encouraged to view the Qualify look-up table in the Phase III MS™ Access 2000 database to become familiar with the descriptions of qualifiers for sediment chemistry parameters. The St. Louis River AOC includes a gradient of sediment chemistry concentrations in the surface sediments, including fairly clean sites that present a low risk to benthic invertebrates, moderately contaminated sites that present some risk of impairment to benthic organisms, and highly contaminated sites that are likely to result in impairments to benthic organisms (Table 8). The Phase III Addendum to the Phase II Help Section for Database Users (Crane and Myre 2005a) provided a section on interpreting mean PEC-Q data contained within the database, as well as on recommended applications for using numerical SQTs in the St. Louis River AOC and on the limitations on the use of the Level I and Level II SQTs and mean PEC-Qs in this AOC (as referenced in Crane et al. 2002a and Crane and MacDonald 2003). As shown in Table 8, the mean PEC-Qs provide a screening-level indication of the risks benthic organisms may be exposed to from a mixture of contaminants in the surface sediments of the St. Louis River AOC. The 0-5 cm depth interval was comprised of the 1995/1996 R-EMAP data set, as well as a smaller number of samples collected elsewhere. The 0 – 15 cm depth intervals were generally collected from hot spot sites. The 0 – 30 cm depth interval included all of the data from 0 – 5 cm and 0 – 15 cm depth intervals, as well as data from 15 – 30 cm intervals and other intervals that fell within 0 – 30 cm (e.g., 2 – 10 cm). Since most of the data that comprised the 0 - 5 cm depth interval is about 10 years old, cleaner sediments may have been deposited over these sites since that time (or been eroded, too). The MPCA does not have much data on sediment deposition rates in the Duluth-Superior Harbor, partly because radioisotope dating of sediment cores has not always been successful. Some sections of the harbor were shown to have a sediment deposition rate of 0.5 cm/year (Schubauer-Berigan and Crane 1997), which would result in the deposition of 5 cm of new sediments within a 10-year period. If the newly deposited sediments are cleaner (due to source control actions in the St. Louis River watershed), then this may improve the habitat for benthic organisms in the AOC. For the Minnesota side of the St. Louis River AOC, the areas with mean PEC-Qs exceeding 0.6 (Figures 13 and 14) included the:

• SLRIDT Superfund site; • USS Superfund site; • Minnesota Slip; • Slip C; • Embayment encompassing the old Duluth sewage treatment plant near WLSSD and the

confluence with Coffee and Miller Creeks; • Dakota Pier; and • One site near Grassy Point.



Assessments of sediment quality in the St. Louis River AOC are best done using a weight-of-evidence approach since sediment chemistry alone does not address bioavailability and bioaccumulation issues. Thus, sediment toxicity data and tissue residue data are useful to collect on surface sediments with matching sediment chemistry data, especially for areas suspected of contamination. The treatment of nondetected data was discussed in Section 5.1 of the Phase II Technical Documentation (Smorong et al. 2004b). The GIS-based sediment quality database was designed with four data treatment options for censoring nondetected data:

• Substitute nondetected values with one-half the detection limit; • Delete nondetected values; • Substitute nondetected values with the detection limit; and • Exclude nondetected values with high detection limits.

Nondetects are also labeled as left-censored data since their values lie somewhere to the left of the detection limit threshold. Users should be aware that bias may be introduced into the results when the above data treatment options are used. For example, the presence of several nondetect values in the pesticide and PCB data should be examined further to assess whether the higher concentrations in the surface sediments compared to subsurface sediments are due to varying detection limits or real effects. Users interested in better approaches for analyzing censored data should consider using maximum likelihood estimation (MLE), imputation, or the Kaplan-Meier method (Helsel 2005a). MLE solves a “likelihood equation” to find the values for mean and standard deviation that are most likely to have produced both nondetect and detected data (Helsel 2005a). Imputation methods fill in values for censored or missing observations without assigning them all the same value (Helsel 2005a). Kaplan-Meier is a nonparametric method designed to incorporate data with multiple censoring levels and does not require specification of an assumed distribution (Helsel 2005a). Additional information about these methods is available in Helsel (2005a,b). Survival was the principal metric for the 10-d amphipod and midge toxicity tests that comprised most of the toxicity test data in the database (Table 9). The chronic 28- to 42-d H. azteca toxicity test is more sensitive than either the 10-d amphipod or midge tests (Ingersoll et al. 2001), and its use would reduce the potential for false negatives at low mean PEC-Qs. Because longer-term toxicity tests provide the most effective mean of discriminating among moderately contaminated sediment samples and because in situ benthic invertebrates are exposed to contaminated sediments for an extended period, it would be prudent in future investigations to evaluate sediment toxicity using the 28- to 42-d H. azteca test (endpoints: survival and growth) in sediments with mean PEC-Qs <5.0 (Crane et al. 2005). However, it would be more cost-effective to utilize acute toxicity tests to characterize the toxicity of highly contaminated sediments (i.e., mean PEC-Qs ≥5.0; Crane et al. 2005). Although benthic invertebrate community data were not examined for this sediment quality report, the R-EMAP project demonstrated that physical habitat characteristics had the greatest influence on the benthic invertebrate community in the lower estuary (Breneman et al. 2000;



Crane et al. 2005). The harbor area, in particular, is dominated by oligochaetes (aquatic worms) and chironomids (midges) that are more pollutant-tolerant organisms common to slower moving, less oxygenated waters. Agency staff and stakeholders are encouraged to use the Phase III GIS-based sediment quality database to obtain data of interest to them and to utilize the companion Phase II ArcMap map documents of GIS-watershed data for viewing the data. Questions regarding the Phase III database should be directed to: Judy L. Crane, Ph.D., Research Scientist III

Environmental Analysis and Outcomes Division MPCA 520 Lafayette Road North St. Paul, MN 55155-4194 Ph: 651-297-4068 Fax: 651-297-7709 Email: [email protected]


mailto:[email protected]


CHAPTER 5

RECOMMENDATIONS FOR ADDITIONAL DATA ANALYSIS

The Phase III GIS-based sediment quality database provides a valuable compilation of sediment quality data for the St. Louis River AOC, including information on sediment chemistry (including mean PEC-Q values), physical parameters (e.g., particle size), sediment toxicity, benthic invertebrate community structure, and tissue residue data. In addition, the database provides information on where most sediment samples have been collected in the AOC since 1990 and provides references and contacts for those data. This database is currently in the process of being expanded with additional sediment quality data sets from the Wisconsin side of the AOC (i.e., Phase IV database). A similar sediment quality report will be prepared for the Phase IV database which is being funded by the Wisconsin Coastal Management Program and NOAA, Office of Ocean and Coastal Resource Management under the Coastal Zone Management Act, Grant #NA05NOS4191067. The Phase IV grant was awarded to the St. Louis River CAC, who are using Exa Data & Mapping Services, Inc. for the contractual database work; the MPCA is providing the 60% state match for this grant through Dr. Crane’s time, and the WDNR are active collaborators on this project. If sufficient resources are available, the following analyses will be considered of the Phase IV data set:

• Update the mean PEC-Q analyses that were done for the Phase II database project that provided a comparison of surficial sediment quality in the St. Louis River AOC with other Great Lakes and U.S. sites (Crane et al. 2004).

• Update the sediment chemistry analyses presented in this report to include the Phase IV sediment chemistry data sets. Depending on the completion of user friendly queries in the Phase IV MS™ Access 2000 database, the compilation of descriptive statistics for sediment chemistry parameters will preferably be based on sediment depth intervals defined in the Access queries rather than using Query Manager’s queries. In addition, nondetect values may be treated a different way (Helsel 2005a) for at least one chemical to assess for bias between using one-half the detection limit for nondetect values and using other statistical procedures for analyzing censored data (e.g., Kaplan-Meier method).

o The sediment chemistry data sets can be further analyzed to remove all of the pre-remediation data for the Hog Island Inlet/Newton Creek stations since this site has been remediated. If post-remediation data are received in time for inclusion in the Phase IV database, those data would be included with the rest of the data set. In addition, the database includes sediment chemistry data for some lakes studied by the Fond du Lac Reservation that are within the St. Louis River watershed; these data could be analyzed separately from the St. Louis River AOC data set. In addition, the Koppers Industries Inc. site is located inland in Superior, WI, and it is hydrologically connected to Superior Bay through a ditch and creek; this site could also be removed from the data set of sediment stations physically located in the river and harbor.



o The sediment chemistry data can also be analyzed for different areas within the AOC, such as Allouez Bay, Superior Bay, and Duluth Harbor to examine the distribution of contaminants in distinct areas of the AOC.

• If new sediment toxicity data are added to the Phase IV database, the queries for sediment

toxicity may be updated. The data set can be evaluated for different areas and data sets as described above for sediment chemistry.

• Integrate more sediment quality data with the updated Phase IV ArcMap 9 map

documents in order to produce maps of interest to stakeholders. The Phase IV sediment quality report will be completed by June 30, 2006. This report will be accessible in pdf format on the MPCA’s Contaminated Sediment Web page at: http://www.pca.state.mn.us/water/sediments/studies-stlouis.html#assessment.



REFERENCES

AScI Corporation. 1996. Results of chronic toxicity testing with whole sediment samples from

IT Corporation – Interlake project. Prepared by AScI Corporation, AScI-Duluth Environmental Testing Division, Duluth, MN for International Technology (IT) Corporation, St. Paul, MN.

AScI Corporation. 2000. Results of ten-day Hyalella azteca and Chironomus tentans toxicity

tests with sediment samples for Fond du Lac Office of Water Protection received October 23, 2000. Prepared by AScI Corporation, Duluth, MN for Fond du Lac Reservation, Cloquet, MN.

ASci Corporation. 2003. Results of ten-day Hyalella azteca and Chironomus tentans toxicity

tests with sediment samples for Fond du Lac Office of Water Protection sediments received December 6, 2002. Prepared by ASci Corporation, Duluth, MN for Fond du Lac Reservation, Cloquet, MN.

Bay West. 2001a. (from NOAA’s St. Louis River Watershed database; reference citation

unavailable) Bay West. 2001b. (from NOAA’s St. Louis River Watershed database; reference citation

unavailable) Breneman, D., C. Richards, and S. Lozano. 2000. Environmental influences on benthic

community structure in a Great Lakes embayment. J. Great Lakes Res. 26:287-304. Budzinski, H., I. Jones, J. Bellocq, C. Pierard, and P. Garrigues. 1997. Evaluation of sediment

contamination by polycyclic aromatic hydrocarbons in the Gironde estuary. Marine Chem. 58:85-97.

CCME (Canadian Council of Ministers of the Environment). 1999. Canadian environmental

quality guidelines. Environment Canada, Winnipeg, MB. Costa, N. 2000. Quality assurance project plan for sediment quality assessment of reservation

lakes. Assistance ID No. GL-97504801-0. Fond du Lac Reservation, Office of Water Protection, Cloquet, MN.

Costa, N. 2001. Final report. Sediment quality assessment of reservation lakes. GLNPO # GL-

97504801-0. Fond du Lac Reservation, Cloquet, MN. Costa, N. 2002. Quality assurance project plan for Phase II sediment quality assessment.

Assistance ID No. GL2001-047. Fond du Lac Reservation, Office of Water Protection, Cloquet, MN.



Costa, N. 2004. Final project report. Phase II sediment quality assessment. GLNPO # GL-2001-047. Fond du Lac Reservation, Cloquet, MN.

Costa, N., M. Watkins, E. Zabel, and P. McCann. 2001. Fond du Lac/Grand Portage fish

consumption advisory project. Minnesota Department of Health, St. Paul, MN in collaboration with Fond du Lac Band of Lake Superior Chippewa and Grand Portage Band of Lake Superior Chippewa.

Crane, J.L. 1997. Toxicity test results of sediment samples collected in 1993 from the

Interlake/Duluth Tar and USX Superfund sites. MPCA technical memorandum sent March 25, 1997 to Steve Hennes, Brenda Winkler, John Betcher, Frank Wallner, and Eric Dott. Minnesota Pollution Control Agency, Water Quality Division, St. Paul, MN.

Crane, J.L. 1999. Assessment of contaminated sediments in Slip C, Duluth Harbor, Minnesota.

U.S. EPA Great Lakes National Program Office, Chicago, IL. EPA-905-R-99-007. (http://www.pca.state.mn.us/water/sediments/slipc.pdf)

Crane, J.L. 2005. Quality assurance project plan (QAPP): Phase III GIS-based sediment quality

database for the St. Louis River Area of Concern. Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN. MPCA Document Number tdr-fg04-05. (http://www.pca.state.mn.us/publications/tdr-fg04-05.pdf)

Crane, J.L. and D.D. MacDonald. 2003. Applications of numerical sediment quality targets for

assessing sediment quality conditions in a U.S. Great Lakes Area of Concern. Environ. Manage. 32:128-140.

Crane, J.L. and P.L. Myre. 2005a. Phase III GIS-based sediment quality database for the St.

Louis River Area of Concern (AOC). Addendum to the Phase II help section for database users. Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN and Exa Data & Mapping Services, Inc., Port Townsend, WA. (http://www.pca.state.mn.us/publications/tdr-fg05-01.pdf)

Crane, J.L. and P.L. Myre. 2005b. Phase III GIS-based sediment quality database for the St.

Louis River Area of Concern (AOC). Addendum to the Phase II technical documentation. Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN and Exa Data & Mapping Services, Inc., Port Townsend, WA. (http://www.pca.state.mn.us/publications/tdr-fg05-02.pdf)

Crane, J.L., M. Schubauer-Berigan, and K. Schmude. 1997. Sediment assessment of hotspot

areas in the Duluth/Superior Harbor. U.S. Environmental Protection Agency, Great Lakes National Program Office, Chicago, IL. EPA-905-R97-020. (http://www.pca.state.mn.us/water/sediments/94mudpuppy.pdf)

Crane, J.L., D.E. Smorong, D.A. Pillard, and D.D. MacDonald. 2002b. Sediment remediation

scoping project in Minnesota Slip, Duluth Harbor. U.S. Environmental Protection





http://www.pca.state.mn.us/water/sediments/94mudpuppy.pdf


Agency, Great Lakes National Program Office, Chicago, IL. EPA-905-R-02-002. (http://www.pca.state.mn.us/water/sediments/mnslipreport.pdf)

Crane, J.L., C. Richards, D. Breneman, S. Lozano, and J.A Schuldt. 2005. Evaluating methods

for assessing sediment quality in a Great Lakes embayment. Aquatic Ecosystem Health Manage. 8:1-27.

Crane, J.L., D.D. MacDonald, D.E. Smorong, L.J. Fisher, S.L. Hamilton, and C.A. Huntington.

2004. Phase II GIS-based sediment quality database for the St. Louis River Area of Concern (AOC): Comparison of surficial sediment contamination in the St. Louis River AOC with other areas of concern in the Great Lakes basin and other areas located elsewhere in the United States. Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN and MacDonald Environmental Sciences Ltd., Nanaimo, BC. (http://www.pca.state.mn.us/publications/tdr-fg04-04.pdf).

Crane, J.L., D.D. MacDonald, C.G. Ingersoll, D.E. Smorong, R.A. Lindskoog, C.G. Severn, T.A.

Berger, and L.J. Field. 2000. Development of a framework for evaluating numerical sediment quality targets and sediment contamination in the St. Louis River Area of Concern. U.S. Environmental Protection Agency, Great Lakes National Program Office, Chicago, IL. EPA 905-R-00-008. (http://www.pca.state.mn.us/water/sediments/sqt-slraoc.pdf and http://www.pca.state.mn.us/water/sediments/sqt-tables.pdf)

Crane, J.L., D.D. MacDonald, C.G. Ingersoll, D.E. Smorong, R.A. Lindskoog, C.G. Severn, T.A.

Berger, and L.J. Field. 2002a. Evaluation of numerical sediment quality targets for the St. Louis River Area of Concern. Arch. Environ. Contam. Toxicol. 43:1-10.

Helsel, D.R. 2005a. More than obvious: Better methods for interpreting nondetect data.

Environ. Sci. Technol. 39:419A-423A. Helsel, D.R. 2005b. Nondetects and data analysis. Statistics for censored environmental data.

Wiley-Interscience, Hoboken, NJ. 250 pp. IJC (International Joint Commission). 1989. Great Lakes water quality agreement of 1978 (as

amended by Protocol signed November 18, 1987). International Joint Commission, Windsor, ON. (http://www.ijc.org/rel/agree/quality.html)

Ingersoll, C.G., D.D. MacDonald, N. Wang, J.L. Crane, L.J. Field, P.S. Haverland, N.E. Kemble,

R.A. Lindskoog, C. Severn, and D.E. Smorong. 2001. Predictions of sediment toxicity using consensus-based freshwater sediment quality guidelines. Arch. Environ. Contam. Toxicol. 41:8-21.

IT (International Technology). 1993. (from NOAA’s St. Louis River Watershed database;

reference citation unavailable) IT. 1994. (from NOAA’s St. Louis River Watershed database; reference citation unavailable)


http://www.pca.state.mn.us/water/sediments/sqt-slraoc.pdf

http://www.pca.state.mn.us/water/sediments/sqt-slraoc.pdf

http://www.pca.state.mn.us/water/sediments/sqt-tables.pdf

http://www.ijc.org/rel/agree/quality.html


IT. 1996. (from NOAA’s St. Louis River Watershed database; reference citation unavailable) IT. 1997. (from NOAA’s St. Louis River Watershed database; reference citation unavailable) IT Corporation. 1997. Remedial investigation data report, sediment operable unit, St. Louis

River/Interlake/Duluth Tar site. Volume 1 of 5. Prepared by IT Corp., St. Paul, MN for the Interlake Corporation, Lisle, IL.

IT. 1998. (from NOAA’s St. Louis River Watershed database; reference citation unavailable) IT. 1999. (from NOAA’s St. Louis River Watershed database; reference citation unavailable) Krueger, G. 1998. Superfund site assessment/brownfield activities in Minnesota. Quality

assurance project plan. Minnesota Pollution Control Agency, Site Assessment Program, St. Paul, MN.

MacDonald, D.D., C.G. Ingersoll, and T.A. Berger. 2000. Development and evaluation of

consensus-based sediment quality guidelines for freshwater ecosystems. Arch. Environ. Contam. Toxicol. 39:20-31.

McCann, P. 2001. Quality assurance project plan. Grand Portage/Fond du Lac fish

consumption advisory project. Grant #X00562101-1. Minnesota Department of Health, St. Paul, MN.

MPCA (Minnesota Pollution Control Agency). Unpublished data. Sediment quality data

collected from the Interlake/Duluth Tar and USX Superfund sites during September 1993. Minnesota Pollution Control Agency, St. Paul, MN.

MPCA. Unpublished data. Sediment quality data collected from reference sites at the St. Louis

River Interlake/Duluth Tar Superfund site during 2001. Minnesota Pollution Control Agency, Environmental Outcomes Division, St. Paul, MN.

MPCA/WDNR (MPCA and Wisconsin Department of Natural Resources). 1992. The St. Louis

River system remedial action plan. Stage One. Minnesota Pollution Control Agency, St. Paul, MN and Wisconsin Department of Natural Resources, Madison, WI. (http://www.stlouisriver.org/rap.html)

MPCA/WDNR. 1995. The St. Louis River system remedial action plan. Progress report.

Minnesota Pollution Control Agency, St. Paul, MN and Wisconsin Department of Natural Resources, Madison, WI. (http://www.stlouisriver.org/stage2/stage2_report.pdf)

NYSDEC (New York State Department of Environmental Conservation). 1999. Technical

guidance for screening contaminated sediments. New York State Department of Environmental Conservation, Division of Fish, Wildlife and Marine Resources, Albany, NY. (http://www.dec.state.ny.us/website/dfwmr/habitat/seddoc.pdf)


http://www.stlouisriver.org/rap.html

http://www.dec.state.ny.us/website/dfwmr/habitat/seddoc.pdf


Schubauer-Berigan, M. and J.L. Crane. 1997. Survey of sediment quality in the Duluth/Superior Harbor: 1993 sampling results. U.S. Environmental Protection Agency, Great Lakes National Program Office, Chicago, IL. EPA 905-R97-005. (http://www.pca.state.mn.us/water/sediments/93mud.pdf)

Service Engineering Group. 2000. (from NOAA’s St. Louis River Watershed database;

reference citation unavailable) Service Engineering Group. 2001a. (from NOAA’s St. Louis River Watershed database;

reference citation unavailable) Service Engineering Group. 2001b. (from NOAA’s St. Louis River Watershed database;

reference citation unavailable) Service Engineering Group. 2004. Remedial design/response action plan. St. Louis

River/Interlake/Duluth Tar site remediation sediment operable unit. Prepared by Service Engineering Group, St. Paul, MN for XIK Corporation, Honeywell International, Inc., and Domtar Inc. (ftp://ftp.visi.com/users/servenv/RD%20RAP%20112404/RD%20RA%20Plan%20Report%20112404.pdf)

Service Engineering Group. Unpublished data. Sediment chemistry data files for sediment

samples collected from Stryker Bay and Slip 7 during 2003 and 2004. Service Engineering Group, St. Paul, MN.

Service Engineering Group. Unpublished data. Sediment chemistry data files for sediment

samples collected from the Cross Channel, Erie Pier, Kingsbury Bay, and Tallas Island areas of the lower St. Louis River Area of Concern (AOC) during 2004. Service Engineering Group, St. Paul, MN.

Smorong, D.E. and J.L. Crane. 2004. Phase II GIS-based sediment quality database for the St.

Louis River Area of Concern (AOC). Help section for database users. MacDonald Environmental Sciences Ltd., Nanaimo, BC and Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN. MPCA Document Number tdr-fg04-01. (http://www.pca.state.mn.us/publications/tdr-fg04-01.pdf)

Smorong, D.E., C.L. Mackenzie, and J.L. Crane. 2004a. Phase II GIS-based sediment quality

database for the St. Louis River Area of Concern (AOC). Help section for ArcView users. MacDonald Environmental Sciences Ltd., Nanaimo, BC and Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN. MPCA Document Number tdr-fg04-02. (http://www.pca.state.mn.us/publications/tdr-fg04-02.pdf)

Smorong, D.E., C.L. Mackenzie, J.L. Crane, D.D. MacDonald, L.J. Fisher, and C.A. Huntington.

2004b. Phase II GIS-based sediment quality database for the St. Louis River Area of Concern (AOC). Technical documentation. MacDonald Environmental Sciences Ltd.,






Nanaimo, BC and Minnesota Pollution Control Agency, Environmental Analysis and Outcomes Division, St. Paul, MN. MPCA Document Number tdr-fg04-03. (http://www.pca.state.mn.us/publications/tdr-fg04-03.pdf)

St. Louis River CAC (Citizens Action Committee). 2002. Lower St. Louis River habitat plan.

St. Louis River Citizens Action Committee, Duluth, MN. (http://www.stlouisriver.org/IAhabitatplan/habitatplan.html)

St. Louis River Natural Resource Trustees. 2002. Fish exposure and injury study workplan for

the St. Louis River/Interlake/Duluth Tar site. St. Louis River Natural Resource Trustees (composed of several state/federal government and tribal entities located in the Great Lakes basin). (http://www.pca.state.mn.us/publications/reports/stlouisriver-interlake-fishstudy.pdf)

Streitz, A. and S. Johnson. 2005. Detailed investigation of the Minnesota Slip featuring laser

induced fluorescence. Minnesota Pollution Control Agency, Remediation Division, Duluth, MN. MPCA Document Number tdr-g1-01. (http://www.pca.state.mn.us/publications/tdr-g1-01.pdf)

URS Corporation. 2003. Former Duluth Works tier II risk assessment work plan. Prepared by

URS Corporation, Franklin, TN for U.S. Steel, Pittsburgh, PA. U.S. EPA (United States Environmental Protection Agency). 2000. Prediction of sediment

toxicity using consensus-based freshwater sediment quality guidelines. U.S. EPA Great Lakes National Program Office, Chicago, IL. EPA 905/R-00/007.

U.S. Fish and Wildlife Service (USFWS). Unpublished data. Tissue residue data for mercury

and PAHs analyzed from fish collected from the St. Louis River/Interlake/Duluth Tar Superfund site (Stryker Bay, Keene Creek/Slip 7), Kingsbury Bay, and North Bay during 2001 and 2002. U.S. Fish and Wildlife Service, Bloomington, MN.

WDNR (Wisconsin Department of Natural Resources). 2003. Consensus-based sediment

quality guidelines. Recommendations for use and application. Interim guidance. Wisconsin Department of Natural Resources, Contaminated Sediment Standing Team, Madison, WI. WT-732 2003. (http://dnr.wi.gov/org/aw/rr/technical/cbsqg_interim_final.pdf)

West Central Environmental Consultants. 2004. Minnesota Slip site sediment sampling and

analysis plan. Prepared by West Central Environmental Consultants, Morris, MN for Minnesota Pollution Control Agency, Duluth, MN. (http://www.pca.state.mn.us/water/sediments/slipsite-samplingplan.pdf)



http://www.pca.state.mn.us/publications/reports/stlouisriver-interlake-fishstudy.pdf

http://www.pca.state.mn.us/publications/reports/stlouisriver-interlake-fishstudy.pdf

http://www.pca.state.mn.us/publications/tdr-g1-01.pdf

http://dnr.wi.gov/org/aw/rr/technical/cbsqg_interim_final.pdf


Figure 1. Location of sediment stations in the St. Louis River AOC that were included in the Phase III GIS-based sediment quality

database.



Figure 2. Location of sediment stations in the Duluth-Superior Harbor that were included in the Phase III GIS-based sediment quality

database.



Figure 3. Example of GIS data for contaminated areas in the lower St. Louis River AOC.



Figure 4. Example of GIS data for geographic features in the lower St. Louis River AOC.



Figure 5. Example of including too much GIS data on a map of the St. Louis River AOC.



Figure 6. Example of a more reasonable selection of GIS data to view of the St. Louis River watershed.



Figure 7. Example of GIS data for land uses in the Duluth-Superior Harbor.



Figure 8. Example of GIS data for recreation uses in the lower St. Louis River AOC.



Figure 9. Example of GIS data included in the U.S. EPA Sensitivity Atlas for the lower St. Louis River AOC.



Figure 10. Example of GIS data for water quality parameters in the lower St. Louis River AOC.



Figure 11. Example of GIS data for water uses in the lower St. Louis River AOC.



Figure 12. Example of a black and white map option for displaying GIS data in the lower St. Louis River AOC.



Figure 13. Example linkage of the Phase III MS™ Access 2000 database query results for mean PEC-Q ranges with the Phase II

ArcMap 9 map document for Ecological Features.



Figure 14. Display of mean PEC-Q ranges in the lower St. Louis River from Cloquet, MN to the St. Louis River Interlake/Duluth Tar

Superfund site.



Table 1. Summary of Data Sets Incorporated into the Phase III GIS-based Sediment Quality Database for the St. Louis River AOC

DESCRIPTION OF DATA AND DATA TYPES Reference Location Sampling

Year(s) Sediment Chemistry Toxicity

Test DataBenthic

Data Bioaccumu-lation Data

Total Number of Samples

AScI Corporation (1996) St. Louis River Interlake/Duluth Tar

(SLRIDT) Superfund site

1996 Grain size, ammonia, SEM metals, AVS, metals, PAHs

Yes No No 16

Bay West (2001a) SLRIDT Superfund site 2001 TOC, solids, grain size, metals, PAHs and other semi-volatiles,

volatiles

No No No 34

Bay West (2001b) SLRIDT Superfund site 2001 TOC, grain size, solids, cyanide, metals, PAHs and other semi-

volatiles, volatiles, dioxins/furans, organo-metals, PCBs

Yes No Yes 47

Breneman et al. (2000); Crane et al. (2005)*

Minnesota side of the St. Louis River AOC

1995; 1996

SEM metals, AVS, Hg, PAHs, TOC, particle size

Yes Yes No 306

Costa (2000, 2001); AScI Corporation (2000)

Reservation lakes near St. Louis River: Big Lake,

Deadfish Lake, Lost Lake, Joe Martin Lake, Pat

Martin Lake, Perch Lake, Rice Portage Lake, Simian

Lake, Sofie Lake, Third Lake, and West Twin Lake

2000 PCBs, Pb, Hg, total solids, total volatile solids, particle size

Yes No No 104

Costa et al. (2001) and McCann (2001)

St. Louis Bay and upper St. Louis River

2000 No No No Yes 99



Table 1. Continued



Test DataBenthic



Costa (2002, 2004); ASci Corporation (2003)

St. Louis River and reservation lakes listed for

Costa (2000, 2001)

2002 Pb, Hg, methyl Hg, PCBs, particle size, total solids, total volatile solids (methyl Hg was

also analyzed on archived reservation lake samples collected in 2000 as part of Phase I study;

Costa 2001)

Yes No No 21

Crane et al. (1997)* Minnesota side of the lower St. Louis River AOC

1994 SEM metals, AVS, As, Pb, Hg, TCDDs, TCDFs, PAHs, PCBs, Ammonia, TOC, particle size

Yes Yes No 225

IT (1993) SLRIDT Superfund site 1993 TOC, ammonia, cyanide, moisture, solids, metals, PAHs,

volatiles

No No No 22

IT (1994) SLRIDT Superfund site 1994 PAHs

No No No 1

IT (1996) SLRIDT Superfund site 1996 Grain size, ammonia, cyanide, solids, sulfate, TOC, metals,

PAHs, volatiles

No No No 56

IT (1997) SLRIDT Superfund site 1997 Solids, ammonia, cyanide, moisture, TOC, metals, PAHs,

volatiles

No No No 6

IT (1998) SLRIDT Superfund site 1998 Grain size, solids, specific gravity, TOC, metals, PAHs

No No No 26

IT (1999) SLRIDT Superfund site 1999 Grain size, pH, solids, specific gravity, TOC, metals, SEM

metals, PAHs

No No No 656



Table 1. Continued



Test DataBenthic



MPCA (unpublished data files - 1993); Crane

(1997)

USX Superfund site 1993 PAHs, phenol, metals, Hg, ammonia, CN, oil & grease, TOC

(matching sediment chemistry/toxicity data were

previously added to the database)

Yes No No 41

MPCA (unpublished data - 2001)

SLRIDT Superfund site: Reference sites

2001 TOC, moisture, metals, PAHs and other semi-volatiles, volatiles

No No No 20

Service Engineering Group (2000)

SLRIDT Superfund site 2000 TOC, volatiles No No No 17

Service Engineering Group (2001a)

SLRIDT Superfund site 2001 Grain size, moisture No No No 256

Service Engineering Group (2001b)

SLRIDT Superfund site 2001 Grain size, pH, solids, TOC, conductivity, alkalinity, sulfate, metals, PAHs and other semi-

volatiles, volatiles, organo-metals

No No No 100

Service Engineering Group (2004) and unpublished data

SLRIDT Superfund site: Stryker Bay and Slip 7

2003; 2004

PAHs, PCBs, metals, Hg, cation exchange capacity, organic matter, pH, total Kjeldahl nitrogen, total nitrogen,

phosphorus, TOC, total solids

No No No 52

Service Engineering Group (2004) and unpublished data

Tallas Island, Kingsbury Bay, Erie Pier, Cross

Channel

2004 PAHs, PCBs, pesticides and herbicides, metals, Hg,

nitrate/nitrite, phosphorus, pH, TOC, organic matter, % solids

No No No 31



Table 1. Continued



Test DataBenthic



St. Louis River Natural Resource Trustees

(2002); USFWS (in preparation)

SLRIDT Superfund site (Stryker Bay, Keene

Creek/Slip 7); North Bay, Kingsbury Bay

2001; 2002

No No No Yes 180

Streitz and Johnson (2005); West Central

Environmental Consultants (2004);

Krueger (1998)

Minnesota Slip 2004 PAHs, PCBs, DROs, metals, Hg, boron, particle size, TOC, bulk

density

No No No 14

URS Corp. (2003); report in preparation

USS Superfund site 2003 PAHs, PCBs, DROs, dioxins/furans, carbazole,

dibenzofuran, OCS, HCB, VOCs, metals, Hg, AVS, SEM, CN, TOC, ammonia, particle size

No No No 41

URS Corp. (2003); MPCA split samples

USS Superfund site 2003 PAHs, PCBs, DROs, dioxins/furans, carbazole,

dibenzofuran, OCS, HCB, VOCs, metals, Hg, AVS, SEM, CN, TOC, ammonia, particle size

No No No 16

As = Arsenic; AVS = acid volatile sulfides; CN = cyanide; DRO = diesel range organics; HCB = hexachlorobenzene; Hg = mercury; OCS = octachlorostyrene; PAH = polycyclic aromatic hydrocarbons; Pb = lead, PCBs = polychlorinated biphenyls; SEM = simultaneously extractable metals; TCDD = tetrachlorodibenzo-p-dioxins; TCDF = tetrachlorodibenzofurans; TOC = total organic carbon; VOCs = volatile organic compounds. * Sediment chemistry and sediment toxicity data were added to previous phases of the sediment quality database.



Table 2. Statistical Summary of Sediment Chemistry Values in the Surficial Sediments of the St. Louis River AOC

Standard 25th 75th Chemical N Average Deviation Minimum Percentile Median* Percentile Maximum

Metals (in mg/kg dry wt.)

Arsenic 228 6.7 11.6 0.05 2 3.3 6.3 142

Cadmium 230 1.0 1.2 0.010 0.25 0.57 1.2 7.2

Chromium, total 309 34.0 29.9 1.9 17.3 30 42 247

Copper 206 37.4 42.8 1.2 15.3 25 42.7 291

Lead 424 62.5 82.2 0.3 14 34.0 81.0 720

Mercury 768 0.22 0.26 0.00095 0.06 0.15 0.28 2.7

Nickel 200 22.2 19.0 2.3 12.9 20 26 166

Zinc 227 171 183 4.5 63.2 120 229 1498

PAHs (in µg/kg dry wt.)

2-Methylnaphthalene 335 21550 152057 0.25 55 267 1289 2063492

Acenaphthene 549 3974 26780 0.05 16.5 70 600 400000

Acenaphthylene 552 6732 67251 0.5 16.2 67 408 1100000

Anthracene 648 27270 249949 1.4 58 305 1800 5300000

Fluorene 632 23020 232363 0.5 42 166 932 5200000



Table 2. Continued


Naphthalene 552 148697 927665 0.5 82.5 286 3521 10793650

Phenanthrene 657 66495 691149 3.7 165 670 4749 16000000

Benz(a)anthracene 652 12134 66711 3.5 164 690 3915 1100000

Benzo(a)pyrene 651 7771 50108 3.5 131 588 2920 892857

Chrysene 653 11237 78834 5 173 740 3700 1700000

Dibenz(a,h)anthracene 546 776 2888 0.2 22.9 97.5 550 30645

Fluoranthene 660 45710 406538 6.2 280 1265 7060 9200000

Pyrene 660 29210 232745 5.6 274 1100 5976 5100000

Total PAHs(13) (exclude high ND) 661 365028 2645123 35.0 1542 6943 40971 53500000

PCBs (in µg/kg dry wt.)

Total PCBs - reported (can be recalculated) 71 199 146 4.2 108 150 267 777

Total PCBs - reported only 53 65.3 79.6 0.72 20.9 50 68.6 460

Total Aroclor PCBs (exclude high ND) 82 207 93.2 42 155 200 287 403



Table 2. Continued

Standard 25th 75th Chemical N Average Deviation Minimum Percentile Median* Percentile Maximum Total congener PCBs (exclude

high ND) 80 188 145 10.5 91.7 147 261 777

Pesticides (in µg/kg dry wt.) Total Chlordane (exclude high

ND) 18 4.0 1.6 2.2 3.2 3.9 4.3 8.1

Total Chlordane - reported only 18 4.3 2.7 1.8 2.1 2.5 6.5 9

Dieldrin 63 11.4 10.6 0.14 3.2 6.5 20.9 31

Sum DDD (exclude high ND) 39 4.7 3.8 0.13 2.1 3.4 8.6 12.5

Sum DDE (exclude high ND) 40 5.0 4.0 0.34 2.1 3.4 10 15

Sum DDT (exclude high ND) 65 11.9 11.1 0.005 2.7 7.5 25 31.5

Total DDT (exclude high ND) 67 17.6 10.5 0.54 8 12.8 29.5 31.5

Endrin 72 15.3 12.6 0.1 3.4 11.3 28.0 45.5

Heptachlor epoxide 27 2.1 0.68 1.1 1.8 2 2.2 4.1

Lindane (gamma-BHC) 28 1.7 0.65 0.01 1.3 2.0 2.1 2.5

Toxaphene 14 29.0 20.8 4.9 10 24.5 42 69



Table 2. Continued


Mean PEC-Q 891 11.0 98.4 0.00245 0.13 0.25 0.76 2330

TOC (%) 580 4.3 5.1 0.01 1.6 3.2 4.8 50.1 BHC = hexachlorocyclohexane; N = number of samples; ND = nondetect; PAHs = polycyclic aromatic hydrocarbons; PAH (13) = sum of 13 low molecular weight and high molecular weight PAHs; PCBs = polychlorinated biphenyls; PEC-Q = probable effect concentration quotient; and TOC = total organic carbon. * Values in italics and yellow shading exceed the corresponding Level I SQT; values in bold italics and orange shading exceed the corresponding Level II SQT.



Table 3. Statistical Summary of Sediment Chemistry Values in Subsurface Sediments of the St. Louis River AOC



Arsenic 294 8.8 11.1 0.05 2.2 4 11.1 66.4

Cadmium 271 1.1 1.2 0.013 0.25 0.52 1.5 7.4

Chromium, total 360 31.8 17.2 2.1 19.6 31 41.4 110

Copper 235 37.5 60.8 3.3 13.9 24 40 660

Lead 510 93.3 146 0.05 13 42 110 1500

Mercury 916 0.29 0.41 0.0005 0.052 0.16 0.37 3.9

Nickel 220 22.0 13.6 3.8 14.4 20.7 26.2 118

Zinc 325 199 288 0.4 59.8 124 241 3780


2-Methylnaphthalene 491 44251 467858 0.13 53.8 320 1999 9411765

Acenaphthene 600 16630 97842 0.19 35.0 221 1394 1269231

Acenaphthylene 613 18237 338035 0.045 19.0 85.0 622 8235294

Anthracene 781 31475 198707 0.19 68.0 556 4525 4117647

Fluorene 776 30553 210192 0.19 57.4 415 3925 4313726



Table 3. Continued


Naphthalene 671 583026 10636216 0.22 105 736 14582 274509813

Phenanthrene 786 95187 655887 0.4 165 1387 14677 14313726

Benz(a)anthracene 779 20765 145297 0.19 135 850 5838 3725490

Benzo(a)pyrene 757 15203 147722 0.11 90 549 3712 3921569

Chrysene 779 20691 126549 0.19 136 827 5688 2745098

Dibenz(a,h)anthracene 610 2375 17742 0.07 22.5 71.5 855 411765

Fluoranthene 787 65917 368369 0.4 177 1774 14000 7254902

Pyrene 788 44967 220256 0.4 165 1640 11402 3921569 Total PAHs (13) (exclude high

ND) 789 874233 12198991 4.89 1484 11710 90737 337274518

PCBs (in µg/kg dry wt.) Total PCBs - reported (can be

recalculated) 113 194 260 4.6 39.7 77.2 252 1273

Total PCBs - reported only 180 49.6 84.9 0.0033 9.9 17.0 48.5 612 Total Aroclor PCBs (exclude

high ND) 48 191 87.6 28.5 123 193 278 328

Total congener PCBs (exclude high ND) 120 187 254 4.6 40.2 78.0 204 1273

Pesticides (in µg/kg dry wt.)

Total Chlordane (exclude high ND) 28 2.3 3.4 0.023 0.13 0.35 7.1 8.4



Table 3. Continued

Standard 25th 75th Chemical N Average Deviation Minimum Percentile Median* Percentile Maximum Total Chlordane - reported

only 36 2.7 1.6 0.65 1.6 2.5 3.7 7.5

Dieldrin 59 3.0 2.4 0.01 0.45 3.5 5.4 9

Sum DDD (exclude high ND) 44 1.8 1.9 0.005 0.34 1 2.5 7.3

Sum DDE (exclude high ND) 44 3.4 4.5 0.012 0.82 2.4 4 26.6

Sum DDT (exclude high ND) 43 1.7 3.2 0.005 0.17 0.78 2.5 20

Total DDT (exclude high ND) 70 10.5 8.7 0.13 3.0 9.0 16.5 36.1

Endrin 60 2.8 2.5 0.01 0.22 2.5 5.2 9

Heptachlor epoxide 34 4.5 1.3 2.5 3.6 4.6 5.5 6.5

Lindane (gamma-BHC) 39 1.7 3.6 0.01 0.41 1.1 2.0 23

Toxaphene 32 34.8 47.9 1.5 7.2 10.0 62.0 204

Mean PEC-Q 1156 24.6 439 0.00038 0.086 0.29 1.6 14700

TOC (%) 770 4.4 5.6 0.01 1.2 2.9 4.9 48 BHC = hexachlorocyclohexane; N = number of samples; ND = nondetect; PAHs = polycyclic aromatic hydrocarbons; PAH (13) = sum of 13 low molecular weight and high molecular weight PAHs; PCBs = polychlorinated biphenyls; PEC-Q = probable effect concentration quotient; and TOC = total organic carbon. * Values in italics and yellow shading exceed the corresponding Level I SQT; values in bold italics and orange shading exceed the corresponding Level II SQT.



Table 4. Determination of Statistical Significance Between Median Chemical Values in Surface and Subsurface Sediments from the St. Louis River AOC

Chemical Surface: N

Subsurface: N

Surface: Median*

Subsurface: Median*

Statistical Significance**


Arsenic 228 294 3.3 4 Yes (p = 0.044)

Cadmium 230 271 0.57 0.52 No (p = 0.551)

Chromium, total 309 360 30 31 No (p = 0.429)

Copper 206 235 25 24 No (p = 0.500)

Lead 424 510 34.0 42 No (p = 0.112)

Mercury 768 916 0.15 0.16 No ( p = 0.061)

Nickel 200 220 20 20.7 No (p = 0.259)

Zinc 227 325 120 124 No (p = 0.661)


2-Methylnaphthalene 335 491 267 320 No (p = 0.301)

Acenaphthene 549 600 70 221 Yes (p = <0.001)

Acenaphthylene 552 613 67 85 Yes (p = 0.026)

Anthracene 648 781 305 556 Yes (p = 0.002)

Fluorene 632 776 166 415 Yes (p = <0.001)

Naphthalene 552 671 286 736 Yes (p = <0.001)

Phenanthrene 657 786 670 1387 Yes (p = <0.001)



Table 4. Continued

Chemical Surface:

N Subsurface:

N Surface: Median*

Subsurface: Median*


Benz(a)anthracene 652 779 690 850 No (p = 0.262)

Benzo(a)pyrene 651 757 588 549 No (p = 0.611)

Chrysene 653 779 740 827 No (p = 0.480)

Dibenz(a,h)anthracene 546 610 97.5 71.5 No (p = 0.445)

Fluoranthene 660 787 1265 1774 No (p = 0.064)

Pyrene 660 788 1100 1640 No (p = 0.083)

Total PAHs (13) (exclude high ND) 661 789 6943 11710 Yes (p = 0.004)

PCBs (in µg/kg dry wt.)

Total PCBs - reported (can be recalculated) 71 113 150 77.2 Yes (p = 0.001)

Total PCBs - reported only 53 180 50 17.0 Yes (p = <0.001)

Total Aroclor PCBs (exclude high ND) 82 48 200 193 No (p = 0.458)

Total congener PCBs (exclude high ND) 80 120 147 78.0 Yes (p = 0.003)

Pesticides (in µg/kg dry wt.)

Total Chlordane (exclude high ND) 18 28 3.9 0.35 Yes (p = 0.006)



Table 4. Continued

Chemical Surface:

N Subsurface:

N Surface: Median*

Subsurface: Median*


Total Chlordane - reported only 18 36 2.5 2.5 Yes (p = 0.020)

Dieldrin 63 59 6.5 3.5 Yes (p = <0.001)

Sum DDD (exclude high ND) 39 44 3.4 1 Yes (p = <0.001)

Sum DDE (exclude high ND) 40 44 3.4 2.4 Yes (p = 0.014)

Sum DDT (exclude high ND) 65 43 7.5 0.78 Yes (p = <0.001)

Total DDT (exclude high ND) 67 70 12.8 9.0 Yes (p = <0.001)

Endrin 72 60 11.3 2.5 Yes (p = <0.001)

Heptachlor epoxide 27 34 2 4.6 Yes (p = <0.001)

Lindane (gamma-BHC) 28 39 2.0 1.1 Yes (p = 0.039)

Toxaphene 14 32 24.5 10.0 No (p = 0.185)

Mean PEC-Q 891 1156 0.25 0.29 No (p = 0.480)

TOC (%) 580 770 3.2 2.9 No (p = 0.286)

BHC = hexachlorocyclohexane; N = number of samples; ND = nondetect; PAHs = polycyclic aromatic hydrocarbons; PAH (13) = sum of 13 low molecular weight and high molecular weight PAHs; PCBs = polychlorinated biphenyls; PEC-Q = probable effect concentration quotient; and TOC = total organic carbon. * Values in italics and yellow shading exceed the corresponding Level I SQT; values in bold italics and orange shading exceed the corresponding Level II SQT. ** All chemical pairs failed the test for normality (p < 0.050), except for heptachlor epoxide (p = 0.108). Statistical significance (pink shading) was determined

using the nonparametric Mann-Whitney Rank Sum test.



Table 5. Results of Regression Analyses of Chemical Parameters Included in the Phase III Sediment Quality Database for the St. Louis River AOC. Regression Relationships are in the Form of Y = b0 + b1(X), with Variables Defined Below

Intercept* Slope

Y Variable (b0) (b1) X Variable r2 value N

Mercury 0.0017 0.0011 Zinc 0.57 221

Lead 7.5 0.34 Zinc 0.547 227

Mercury 0.09 1.9 Lead 0.413 416

Zinc 99 14 TOC 0.312 171

Mercury 0.12 0.025 TOC 0.24 516

Lead 35 6.6 TOC 0.227 291

Total PAHs (13) -159000 104000 TOC 0.0414 433

Total PAHs (13) -196000 103000 TOC (0.2 - 12%) 0.0362 374

Total PAHs (13) 23500 28200 TOC (0.5 - 10%) 0.00813 345

Total PAHs (13) 41400 152000 Mercury 0.00558 452

Total PAHs (13) 81400 200 Lead 0.000806 312

Total PAHs (13) 124000 82 Zinc 0.000398 203 TOC = total organic carbon; PAHs (13) = sum of the 13 low molecular weight and high molecular weight PAHs. *Metals are in units of mg/kg dry wt.; Total PAHs (13) are in units of µg/kg dry wt.



Table 6. Comparison of PAH Source Ratios for Various Sediment Depth Intervals in the St. Louis River AOC. P/A = Phenanthrene/Anthracene and F/P = Fluoranthene/Pyrene.

P/A F/P

Depth Interval (cm) N Median N Median

0 - 5 114 3.4 114 1.2

0 - 15 250 2.8 250 1.2

0 - 30 303 2.1* 303 1.3**

15 - 30 49 4.4 49 1.2

30 - 45 36 4.6 36 1.2

>30 377 2.4* 377 1.2** N = number of samples. A Mann-Whitney Rank Sum test was run between the P/A groups of 0 - 30 cm and >30 cm, as well as between the F/P groups of the same depth intervals. * For P/A ratios, the difference in the median values between the two groups was greater than would be

expected by chance; there was a statistically significant difference (p = 0.002). ** For F/P ratios, the difference in the median values between the two groups was not great enough to

exclude the possibility that the difference was due to random sampling variability; there was not a statistically significant difference (p = 0.134).



Table 7. Comparison of Surficial Sediment Chemistry Data to Four Types of SQTs Used by the MPCA (Crane et al. 2000)

Number of SQTs Exceeded

Type of SQT N Average Minimum Maximum

Level I SQT 1047 9 0 30

Level II SQT 1047 3.9 0 21

Human Health SQT 779 1.8 1 9

Wildlife SQT 257 2.8 0 9 N = number of samples; SQT = sediment quality target. Table 8. Percentage of Samples in Three Surficial Depth Intervals with Low, Moderate, or High

Mean PEC-Q Ranges

Percentage (%) of Samples in Mean PEC-Q Risk Category

Low Moderate High

Depth Interval (cm) (<0.1)* (≥0.1 to ≤0.6) (>0.6)**

0 - 5 40.7 47.7 11.6

0 - 15 12.7 56.5 30.8

0 - 30 15.9 42.6 41.5 PEC-Q = probable effect concentration quotient. * Mean PEC-Qs <0.1 correspond with the Level I SQT, which is intended to identify contaminant

concentrations below which harmful effects on sediment-dwelling organisms are unlikely to be observed (Crane et al. 2000, 2002a).

** Mean PEC-Qs >0.6 correspond with the Level II SQT, which is intended to identify contaminant

concentrations above which harmful effects on sediment-dwelling organisms are likely to be frequently or always observed (Crane et al. 2000, 2002a).



Table 9. Compilation of Sediment Toxicity Test Results from the Phase III GIS-based Sediment Quality Database for the St. Louis River AOC

Nontoxic or Toxic or

Toxicity Test Species and Endpoint No Effect Effect

N (%) (%)

Amphipod: Hyalella azteca

10-d survival 197 88.3 11.7

28-d survival 30 46.7 53.3

28-d growth (weight) 30 50 50

28-d growth (length) 20 40 60

28-d survival (UV-exposure) 20 80 20

Midge: Chironomus tentans

10-d survival 225 89.8 10.2

10-d growth (ash free dry weight) 48 85.4 14.6

20-d survival 20 65 35

20-d growth 20 65 35

20-d emergence 20 100 0

20-d survival (UV-exposure) 20 75 25

Oligochaete: Lumbriculus variegatus

28-d growth 20 30 70

Cattail: Typha latifolia

17-d survival 23 78.3 21.7 d = day; N = number of samples; UV = ultraviolet.



APPENDIX A

COPY OF LEVEL I AND LEVEL II SEDIMENT QUALITY TARGETS FOR THE MINNESOTA SIDE OF THE ST. LOUIS RIVER AOC



APPENDIX A Table A-1. Recommended Level I and Level II Sediment Quality Targets for the Protection of

Sediment-dwelling Organisms (Crane et al. 2000, 2002a)

Aquatic Life

Level I Level II

Chemical SQT SQT

Source†

Metals (in mg/kg DW) Arsenic§ 9.8 33 MacDonald et al. (2000) Cadmium*§ 0.99 5.0 MacDonald et al. (2000) Chromium§ 43 110 MacDonald et al. (2000) Copper*§ 32 150 MacDonald et al. (2000) Lead*§ 36 130 MacDonald et al. (2000) Mercury 0.18 1.1 MacDonald et al. (2000) Nickel§ 23 49 MacDonald et al. (2000) Zinc*§ 120 460 MacDonald et al. (2000)

PAHs (in µg/kg DW)

2-Methylnaphthalene 20 200 CCME (1999) Acenaphthene 6.7 89 CCME (1999) Acenaphthylene 5.9 130 CCME (1999) Anthracene* 57 850 MacDonald et al. (2000) Fluorene 77 540 MacDonald et al. (2000) Naphthalene*§ 180 560 MacDonald et al. (2000) Phenanthrene*§ 200 1200 MacDonald et al. (2000) Benz(a)anthracene*§ 110 1100 MacDonald et al. (2000) Benzo(a)pyrene*§ 150 1500 MacDonald et al. (2000) Chrysene*§ 170 1300 MacDonald et al. (2000) Dibenz(a,h)anthracene 33 140 MacDonald et al. (2000); CCME (1999) Fluoranthene* 420 2200 MacDonald et al. (2000) Pyrene*§ 200 1500 MacDonald et al. (2000) Total PAHs*§ 1600 23000 MacDonald et al. (2000)

PCBs (in µg/kg DW)

Total PCBs*§ 60 680 MacDonald et al. (2000)

A-1 MPCA Document Number: tdr-fg05-04


Table A-1. Continued

Aquatic Life

Level I Level II

Chemical SQT SQT

Source†

Pesticides (in µg/kg DW)

Chlordane* 3.2 18 MacDonald et al. (2000) Dieldrin* 1.9 62 MacDonald et al. (2000) Sum DDD* 4.9 28 MacDonald et al. (2000) Sum DDE*§ 3.2 31 MacDonald et al. (2000) Sum DDT* 4.2 63 MacDonald et al. (2000) Total DDT* 5.3 570 MacDonald et al. (2000) Endrin 2.2 210 MacDonald et al. (2000) Heptachlor epoxide* 2.5 16 MacDonald et al. (2000) Lindane (gamma-BHC) 2.4 5 MacDonald et al. (2000) Toxaphene 0.1 32 NYSDEC (1999)¥

Mean PEC-Q 0.1 0.6 USEPA 2000

DW = dry weight; SQT = sediment quality target; PEC-Q = probable effect concentration quotient. † Some SQT values were rounded to two significant figures from the original source. * Reliable consensus-based TEC values that were adopted as Level I SQTs [i.e., predictive ability ≥75% and ≥20 samples below the TEC (MacDonald et al. 2000)].

§ Reliable consensus-based PEC values that were adopted as Level II SQTs [i.e., predictive ability ≥75% and ≥20 samples predicted to be toxic (MacDonald et al. 2000)].

¥ originally based on μg/g OC; assumed TOC = 1%.

A-2 MPCA Document Number: tdr-fg05-04


APPENDIX B

DESCRIPTIVE STATISTICS OF SEDIMENT CHEMISTRY PARAMETERS WITH CORRESPONDING SEDIMENT QUALITY TARGETS, AS WELL AS

TOTAL ORGANIC CARBON



APPENDIX B

Descriptive Statistical Information for Selected Chemicals from SigmaStat 3.1 Analyses

Descriptive Statistics: Sunday, December 11, 2005, 1:36:42 PM Data source: Arsenic: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Arsenic 228 0 6.746 11.554 0.765 1.508 Subsurf: Arsenic 294 0 8.773 11.110 0.648 1.275 Column Range Max Min Median 25% 75% Surface: Arsenic 142.090 142.140 0.0500 3.250 2.025 6.300 Subsurf: Arsenic 66.390 66.440 0.0500 4.000 2.200 11.100 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Arsenic 7.642 83.334 0.287 <0.001 1538.072 40678.208 Subsurf: Arsenic 2.260 5.520 0.257 <0.001 2579.124 58792.133 Descriptive Statistics: Sunday, December 11, 2005, 1:38:59 PM Data source: Cadmium: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Cd 230 0 1.016 1.200 0.0791 0.156 Subsurf: Cd 271 0 1.074 1.153 0.0701 0.138 Column Range Max Min Median 25% 75% Surface: Cd 7.193 7.203 0.00999 0.565 0.250 1.200 Subsurf: Cd 7.417 7.430 0.0130 0.520 0.250 1.513 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Cd 2.548 8.534 0.211 <0.001 233.651 567.198 Subsurf: Cd 1.890 4.561 0.206 <0.001 291.131 671.883

B-1 MPCA Document Number: tdr-fg05-04


Descriptive Statistics: Sunday, December 11, 2005, 1:40:45 PM Data source: Chromium: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Cr 309 0 33.956 29.950 1.704 3.353 Subsurface: Cr 360 0 31.850 17.212 0.907 1.784 Column Range Max Min Median 25% 75% Surface: Cr 244.800 246.700 1.900 30.000 17.300 42.000 Subsurface: Cr 107.900 110.000 2.100 31.000 19.600 41.370 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Cr 3.675 19.441 0.167 <0.001 10492.288 632541.721 Subsurface: Cr 0.851 1.472 0.0609 0.003 11466.007 471543.572 Descriptive Statistics: Sunday, December 11, 2005, 11:03:41 AM Data source: Copper: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Copper 206 0 37.435 42.847 2.985 5.886 Subsurf: Copper 235 0 37.461 60.822 3.968 7.817 Column Range Max Min Median 25% 75% Surface: Copper 289.800 291.000 1.200 25.000 15.300 42.748 Subsurf: Copper 656.660 660.000 3.340 24.000 13.868 40.000 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Copper 3.411 14.697 0.228 <0.001 7711.616 665037.306 Subsurf: Copper 6.958 60.658 0.287 <0.001 8803.392 1195413.116



Descriptive Statistics: Sunday, December 11, 2005, 1:42:33 PM Data source: Lead: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Lead 424 0 62.510 82.158 3.990 7.843 Subsurf: Pb 510 0 93.303 146.407 6.483 12.737 Column Range Max Min Median 25% 75% Surface: Lead 719.700 720.000 0.300 33.950 13.954 81.000 Subsurf: Pb 1499.950 1500.000 0.0500 42.000 13.000 110.000 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Lead 3.576 19.002 0.224 <0.001 26504.394 4512000.646 Subsurf: Pb 4.321 29.577 0.262 <0.001 47584.620 15350285.626 Descriptive Statistics: Sunday, December 11, 2005, 1:43:55 PM Data source: Mercury: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Hg 768 0 0.219 0.264 0.00953 0.0187 Subsurf: Hg 916 0 0.293 0.412 0.0136 0.0267 Column Range Max Min Median 25% 75% Surface: Hg 2.729 2.730 0.000950 0.149 0.0605 0.283 Subsurf: Hg 3.899 3.900 0.000500 0.159 0.0520 0.372 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Hg 3.785 22.382 0.205 <0.001 168.152 90.326 Subsurf: Hg 3.675 19.404 0.239 <0.001 268.449 234.067



Descriptive Statistics: Sunday, December 11, 2005, 1:45:18 PM Data source: Nickel: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Nickel 200 0 22.179 18.988 1.343 2.648 Subsurf: Nickel 220 0 22.010 13.580 0.916 1.804 Column Range Max Min Median 25% 75% Surface: Nickel 163.700 166.000 2.300 20.000 12.900 26.000 Subsurf: Nickel 114.230 118.000 3.770 20.679 14.400 26.250 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Nickel 4.186 25.372 0.189 <0.001 4435.751 170126.002 Subsurf: Nickel 3.169 16.561 0.146 <0.001 4842.255 146967.682 Descriptive Statistics: Sunday, December 11, 2005, 1:46:48 PM Data source: Zinc: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Zn 227 0 170.569 183.080 12.151 23.945 Subsurface: Zn 325 0 198.705 287.998 15.975 31.428 Column Range Max Min Median 25% 75% Surface: Zn 1493.500 1498.000 4.500 120.000 63.250 229.006 Subsurface: Zn 3779.600 3780.000 0.400 124.000 59.825 240.750 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Zn 3.755 20.310 0.182 <0.001 38719.205 14179450.396 Subsurface: Zn 7.177 77.534 0.251 <0.001 64579.187 39705725.841



Descriptive Statistics: Sunday, December 11, 2005, 1:48:28 PM Data source: 2-Methylnaphthalene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: 2-Methylnaph. 335 0 21550.336 152056.566 8307.738 16342.084 Subsurf: 2-Meth. 491 0 44250.925 467857.601 21114.118 41485.381 Column Range Max Min Median 25% 75% Surface: 2-Methylnaph. 2063491.690 2063491.940 0.250 266.660 55.000 1288.830 Subsurf: 2-Meth. 9411764.525 9411764.650 0.125 320.000 53.755 1999.037 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Surface: 2-Methylnaph. 10.001 113.381 0.449 <0.001 7219362.655 Subsurf: 2-Meth. 17.467 335.605 0.462 <0.001 21727204.135 Column Sum of Squares Surface: 2-Methylnaph. 7.878E+012 Subsurf: 2-Meth. 1.082E+014 Descriptive Statistics: Sunday, December 11, 2005, 1:50:12 PM Data source: Acenaphthene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Acenaphthene 549 0 3973.957 26779.730 1142.931 2245.062 Subsurf: Acenapthen 600 0 16630.148 97841.627 3994.368 7844.668 Column Range Max Min Median 25% 75% Surface: Acenaphthene 399999.950 400000.000 0.0500 70.000 16.500 600.000 Subsurf: Acenapthen 1269230.520 1269230.710 0.190 220.575 35.000 1394.240 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Surface: Acenaphthene 11.316 141.775 0.441 <0.001 2181702.273 Subsurf: Acenapthen 9.706 104.367 0.433 <0.001 9978088.950 Column Sum of Squares Surface: Acenaphthene401670358088.334 Subsurf: Acenapthen 5.900E+012



Descriptive Statistics: Sunday, December 11, 2005, 1:52:09 PM Data source: Acenaphthylene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Acenaph. 552 0 6732.046 67251.443 2862.411 5622.574 Subsurf: Acenaptyle 613 0 18236.651 338034.861 13653.099 26812.609 Column Range Max Min Median 25% 75% Surface: Acenaph. 1099999.500 1100000.000 0.500 67.000 16.250 407.500 Subsurf: Acenaptyle 8235293.905 8235293.950 0.0450 85.000 19.000 621.715 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Acenaph. 14.108 211.937 0.461 <0.001 3716089.282 2.517E+012 Subsurf: Acenaptyle 23.685 573.731 0.478 <0.001 11179067.055 7.014E+013 Descriptive Statistics: Sunday, December 11, 2005, 1:54:20 PM Data source: Anthracene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Anthra. 648 0 27269.514 249948.950 9818.922 19280.802 Subsurf: Anthrac 781 0 31475.448 198706.954 7110.294 13957.578 Column Range Max Min Median 25% 75% Surface: Anthra. 5299998.600 5300000.000 1.400 305.000 58.000 1800.000 Subsurf: Anthrac 4117646.780 4117646.970 0.190 555.550 68.000 4525.000 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Anthra. 16.735 323.580 0.457 <0.001 17670645.090 4.090E+013 Subsurf: Anthrac 14.529 260.605 0.437 <0.001 24582325.010 3.157E+013



Descriptive Statistics: Sunday, December 11, 2005, 1:55:52 PM Data source: Fluorene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Fluorene 632 0 23019.598 232363.062 9242.906 18150.578 Subsurf: Fluorene 776 0 30553.280 210192.135 7545.458 14811.958 Column Range Max Min Median 25% 75% Surface: Fluorene 5199999.500 5200000.000 0.500 165.500 42.000 932.135 Subsurf: Fluorene 4313725.400 4313725.590 0.190 415.330 57.445 3925.000 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Fluorene 18.763 399.335 0.461 <0.001 14548386.155 3.440E+013 Subsurf: Fluorene 14.722 260.293 0.442 <0.001 23709345.445 3.496E+013 Descriptive Statistics: Sunday, December 11, 2005, 1:57:37 PM Data source: Naphthalene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Naphth. 552 0 148696.724 927664.971 39484.043 77557.665 Subsurf: Naphth. 671 0 583026.395 10636216.064 410606.567 806230.510 Column Range Max Min Median 25% 75% Surface: Naphth. 10793649.890 10793650.390 0.500 285.500 82.500 3520.750 Subsurf: Naphth. 274509812.280 274509812.500 0.220 735.840 105.000 14582.090 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Naphth. 8.629 80.900 0.446 <0.001 82080591.765 4.864E+014 Subsurf: Naphth. 25.579 659.523 0.478 <0.001 391210711.230 7.602E+016



Descriptive Statistics: Sunday, December 11, 2005, 1:59:12 PM Data source: Phenanthrene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Phenan. 657 0 66495.228 691148.817 26964.283 52946.712 Subsurf: Phenan 786 0 95186.988 655886.458 23394.695 45923.566 Column Range Max Min Median 25% 75% Surface: Phenan. 15999996.300 16000000.000 3.700 670.000 165.000 4749.105 Subsurf: Phenan 14313725.190 14313725.590 0.400 1387.445 165.000 14677.420 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Phenan. 19.828 439.559 0.462 <0.001 43687364.890 3.163E+014 Subsurf: Phenan 16.102 312.595 0.442 <0.001 74816972.920 3.448E+014 Descriptive Statistics: Sunday, December 11, 2005, 2:00:53 PM Data source: Benz(a)anthracene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: benz(a)a 652 0 12134.079 66711.300 2612.616 5130.171 Subsurf: benz(a)an 779 0 20769.655 145296.520 5205.788 10219.055 Column Range Max Min Median 25% 75% Surface: benz(a)a 1099996.500 1100000.000 3.500 690.450 164.000 3915.000 Subsurf: benz(a)an 3725490.040 3725490.230 0.190 850.000 135.328 5837.555 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: benz(a)a 11.232 150.488 0.428 <0.001 7911419.330 2.993E+012 Subsurf: benz(a)an 21.859 546.497 0.443 <0.001 16179561.210 1.676E+013



Descriptive Statistics: Sunday, December 11, 2005, 2:02:21 PM Data source: Benzo(a)pyrene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: benzo(a)p 651 0 7771.057 50107.715 1963.876 3856.307 Subsurf: benzo(a)py 757 0 15203.141 147722.402 5369.062 10540.043 Column Range Max Min Median 25% 75% Surface: benzo(a)p 892853.620 892857.120 3.500 588.000 130.990 2920.000 Subsurf: benzo(a)py 3921568.490 3921568.600 0.110 549.010 90.000 3712.500 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: benzo(a)p 13.345 201.872 0.438 <0.001 5058958.060 1.671E+012 Subsurf: benzo(a)py 24.701 649.550 0.459 <0.001 11508778.035 1.667E+013 Descriptive Statistics: Sunday, December 11, 2005, 2:04:02 PM Data source: Chrysene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Chryse. 653 0 11236.877 78833.517 3084.994 6057.722 Subsurf: Chrysene 779 0 20690.674 126548.617 4534.075 8900.470 Column Range Max Min Median 25% 75% Surface: Chryse. 1699995.000 1700000.000 5.000 740.000 172.933 3700.000 Subsurf: Chrysene 2745097.950 2745098.140 0.190 827.000 136.502 5688.043 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Chryse. 16.928 336.863 0.443 <0.001 7337680.830 4.134E+012 Subsurf: Chrysene 16.365 317.398 0.435 <0.001 16118034.950 1.279E+013



Descriptive Statistics: Sunday, December 11, 2005, 2:16:32 PM Data source: Dibenz(a,h)anthracene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Dibenz(ah). 546 0 775.944 2888.047 123.597 242.785 Subsurf: Dibenz(ah) 610 0 2375.468 17741.624 718.337 1410.719 Column Range Max Min Median 25% 75% Surf: Dibenz(ah). 30644.960 30645.160 0.200 97.500 22.900 550.000 Subsurf: Dibenz(ah) 411764.640 411764.710 0.0700 71.500 22.500 855.260 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Dibenz(ah). 7.840 68.998 0.394 <0.001 423665.400 4874484470.439 Subsurf: Dibenz(ah) 20.541 468.049 0.447 <0.001 1449035.570 195134163496.534 Descriptive Statistics: Sunday, December 11, 2005, 2:18:15 PM Data source: Fluoranthene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Fluoranth. 660 0 45710.431 406538.173 15824.477 31072.473 Subsurf: Fluoranth 787 0 65916.960 368368.813 13130.930 25775.841 Column Range Max Min Median 25% 75% Surf: Fluoranth. 9199993.800 9200000.000 6.200 1265.365 280.000 7059.800 Subsurf: Fluoranth 7254901.460 7254901.860 0.400 1774.190 176.602 14000.000 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Fluoranth. 18.775 401.699 0.455 <0.001 30168884.350 1.103E+014 Subsurf: Fluoranth 13.726 232.180 0.429 <0.001 51876647.450 1.101E+014



Descriptive Statistics: Sunday, December 11, 2005, 2:19:37 PM Data source: Pyrene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Pyrene 660 0 29209.605 232744.802 9059.579 17789.121 Subsurf: Pyrene 788 0 44967.080 220256.037 7846.296 15402.146 Column Range Max Min Median 25% 75% Surface: Pyrene 5099994.400 5100000.000 5.600 1100.000 274.000 5976.190 Subsurf: Pyrene 3921568.200 3921568.600 0.400 1639.785 165.000 11402.175 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Pyrene 17.463 355.715 0.450 <0.001 19278339.000 3.626E+013 Subsurf: Pyrene 11.370 162.292 0.419 <0.001 35434059.280 3.977E+013 Descriptive Statistics: Sunday, December 11, 2005, 2:20:47 PM Data source: T. PAHs (13): surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: T. PAHs (13) 661 0 365027.956 2645123.008 102883.361 202018.150 Subsurf: T.PAHs(13) 789 0 874232.990 12198990.837 434295.574 852513.112 Column Range Max Min Median 25% 75% Surf: T. PAHs (13) 53499965.050 53500000.000 34.950 6942.910 1542.000 40971.000 Subsurf: T.PAHs(13) 337274512.775 337274517.660 4.885 11710.000 1484.165 90736.958 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: T. PAHs (13) 14.587 262.003 0.445 <0.001 241283478.990 4.706E+015 Subsurf: T.PAHs(13) 26.743 736.598 0.471 <0.001 689769829.070 1.179E+017



Descriptive Statistics: Sunday, December 11, 2005, 2:22:07 PM Data source: T. PCBs-reported: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: T. PCBs report 71 0 198.660 145.669 17.288 34.479 Subsurf: T.PCBs-rep 113 0 194.011 260.452 24.501 48.546 Column Range Max Min Median 25% 75% Surf: T. PCBs report 772.970 777.170 4.200 150.000 108.135 267.380 Subsurf: T.PCBs-rep 1268.150 1272.730 4.580 77.220 39.743 252.025 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: T. PCBs report 1.453 2.871 0.154 <0.001 14104.860 4287430.316 Subsurf: T.PCBs-rep 2.197 5.021 0.276 <0.001 21923.280 11850937.846 Descriptive Statistics: Sunday, December 11, 2005, 2:23:42 PM Data source: T. PCBs-reported only: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: T. PCBs-rep only 53 0 65.265 79.632 10.938 21.949 Subsurf: T.PCBs-ro 180 0 49.579 84.860 6.325 12.481 Column Range Max Min Median 25% 75% Surf: T. PCBs-rep only 459.285 460.000 0.715 50.000 20.938 68.563 Subsurf: T.PCBs-ro 611.997 612.000 0.00327 17.000 9.900 48.500 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: T. PCBs-rep only 3.214 12.414 0.249 <0.001 3459.033 555500.459 Subsurf: T.PCBs-ro 3.597 16.141 0.280 <0.001 8924.188 1731480.443



Descriptive Statistics: Sunday, December 11, 2005, 2:25:31 PM Data source: T. Aroclor PCBs: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: T. Arocl. PCBs 82 0 206.711 93.234 10.296 20.486 Subsurf: T. Ar.PCBs 48 0 190.813 87.627 12.648 25.444 Column Range Max Min Median 25% 75% Surf: T. Arocl. PCBs 360.500 402.500 42.000 199.750 155.000 287.000 Subsurf: T. Ar.PCBs 299.000 327.500 28.500 192.500 123.375 277.500 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: T. Arocl. PCBs 0.237 -0.854 0.149 <0.001 16950.300 4207913.175 Subsurf: T. Ar.PCBs -0.281 -1.138 0.123 0.065 9159.000 2108542.125

Descriptive Statistics: Sunday, December 11, 2005, 2:26:58 PM Data source: T. Congener PCBs: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: T. Cong. PCBs 80 0 188.377 144.973 16.209 32.262 Subsurf: T.Cong.PCBs 120 0 187.174 253.821 23.171 45.880 Column Range Max Min Median 25% 75% Surf: T. Cong. PCBs 766.700 777.200 10.500 146.695 91.743 260.945 Subsurf: T.Cong.PCBs 1268.149 1272.730 4.581 77.975 40.235 203.900 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: T. Cong. PCBs 1.442 2.853 0.131 0.002 15070.125 4499220.624 Subsurf: T.Cong.PCBs 2.308 5.607 0.271 <0.001 22460.826 11870645.666



Descriptive Statistics: Sunday, December 11, 2005, 2:28:33 PM Data source: Total Chlordane (MESL): surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: T. Chlordane 18 0 4.033 1.647 0.388 0.819 Subsurf: T.Chlordane 28 0 2.323 3.392 0.641 1.315 Column Range Max Min Median 25% 75% Surf: T. Chlordane 5.900 8.100 2.200 3.850 3.200 4.300 Subsurf: T.Chlordane 8.377 8.400 0.0230 0.346 0.126 7.100 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: T. Chlordane 1.710 2.976 0.277 <0.001 72.600 338.940 Subsurf: T.Chlordane 1.013 -1.018 0.418 <0.001 65.055 461.835

Descriptive Statistics: Sunday, December 11, 2005, 2:30:02 PM Data source: Total Chlordane, reported only: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: T Chlor, rep only 18 0 4.314 2.691 0.634 1.338 Subsurf: T.Chlor, ro 36 0 2.668 1.625 0.271 0.550 Column Range Max Min Median 25% 75% Surf: T Chlor, rep only 7.250 9.000 1.750 2.500 2.100 6.500 Subsurf: T.Chlor, ro 6.850 7.500 0.650 2.450 1.600 3.725 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: T Chlor, rep only 0.574 -1.429 0.281 <0.001 77.650 458.077 Subsurf: T.Chlor, ro 1.354 1.544 0.211 <0.001 96.050 348.707



Descriptive Statistics: Sunday, December 11, 2005, 2:31:55 PM Data source: Dieldrin: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Dieldrin 63 0 11.405 10.565 1.331 2.661 Subsurf: Dieldrin 59 0 3.039 2.352 0.306 0.613 Column Range Max Min Median 25% 75% Surface: Dieldrin 30.860 31.000 0.140 6.500 3.150 20.875 Subsurf: Dieldrin 8.990 9.000 0.01000 3.500 0.449 5.375 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Dieldrin 0.822 -0.997 0.251 <0.001 718.540 15115.990 Subsurf: Dieldrin 0.181 -0.995 0.155 0.001 179.330 865.987

Descriptive Statistics: Sunday, December 11, 2005, 2:33:58 PM Data source: Sum DDD: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Sum DDD 39 0 4.705 3.781 0.606 1.226 Subsurf: Sum DDD 44 0 1.779 1.898 0.286 0.577 Column Range Max Min Median 25% 75% Surf: Sum DDD 12.370 12.500 0.130 3.350 2.112 8.588 Subsurf: Sum DDD 7.295 7.300 0.00500 1.000 0.335 2.500 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Sum DDD 0.993 -0.472 0.281 <0.001 183.480 1406.592 Subsurf: Sum DDD 1.554 2.128 0.182 <0.001 78.285 294.255



Descriptive Statistics: Sunday, December 11, 2005, 2:35:28 PM Data source: Sum DDE: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Sum DDE 40 0 4.989 4.041 0.639 1.292 Subsurf: Sum DDE 44 0 3.444 4.529 0.683 1.377 Column Range Max Min Median 25% 75% Surf: Sum DDE 14.660 15.000 0.340 3.350 2.125 10.000 Subsurf: Sum DDE 26.588 26.600 0.0120 2.440 0.825 4.005 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Sum DDE 1.039 -0.313 0.288 <0.001 199.580 1632.666 Subsurf: Sum DDE 3.468 15.697 0.227 <0.001 151.537 1403.772

Descriptive Statistics: Sunday, December 11, 2005, 2:36:44 PM Data source: Sum DDT: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Sum DDT 65 0 11.895 11.090 1.376 2.748 Subsurf: Sum DDT 43 0 1.723 3.227 0.492 0.993 Column Range Max Min Median 25% 75% Surf: Sum DDT 31.495 31.500 0.00500 7.500 2.675 25.000 Subsurf: Sum DDT 19.995 20.000 0.00500 0.780 0.170 2.500 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Sum DDT 0.716 -1.176 0.245 <0.001 773.145 17067.035 Subsurf: Sum DDT 4.618 25.277 0.297 <0.001 74.095 564.998



Descriptive Statistics: Sunday, December 11, 2005, 2:38:09 PM Data source: Total DDT: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: T. DDT 67 0 17.619 10.525 1.286 2.567 Subsurf: T. DDT 70 0 10.549 8.702 1.040 2.075 Column Range Max Min Median 25% 75% Surface: T. DDT 30.960 31.500 0.540 12.750 8.025 29.500 Subsurf: T. DDT 35.968 36.100 0.132 9.045 3.055 16.500 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: T. DDT 0.0738 -1.664 0.190 <0.001 1180.505 28111.030 Subsurf: T. DDT 1.054 1.025 0.116 0.021 738.417 13014.237

Descriptive Statistics: Sunday, December 11, 2005, 2:39:30 PM Data source: Endrin: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: Endrin 72 0 15.262 12.609 1.486 2.963 Subsurf: Endrin 60 0 2.787 2.480 0.320 0.641 Column Range Max Min Median 25% 75% Surface: Endrin 45.400 45.500 0.1000 11.250 3.350 28.000 Subsurf: Endrin 8.990 9.000 0.01000 2.500 0.215 5.250 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: Endrin 0.454 -1.194 0.209 <0.001 1098.830 28058.089 Subsurf: Endrin 0.300 -1.156 0.226 <0.001 167.195 828.819



Descriptive Statistics: Sunday, December 11, 2005, 2:41:01 PM Data source: Heptachlor epoxide: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Hepta.epox 27 0 2.063 0.682 0.131 0.270 Subsurf: Hepta_epo. 34 0 4.494 1.330 0.228 0.464 Column Range Max Min Median 25% 75% Surf: Hepta.epox 2.950 4.050 1.100 2.000 1.750 2.188 Subsurf: Hepta_epo. 4.000 6.500 2.500 4.625 3.600 5.500 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Hepta.epox 1.715 4.192 0.253 <0.001 55.700 127.000 Subsurf: Hepta_epo. -0.163 -1.257 0.187 0.004 152.800 745.080

Descriptive Statistics: Sunday, December 11, 2005, 2:43:40 PM Data source: Lindane: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Lindane 28 0 1.683 0.646 0.122 0.250 Subsurf: Lindane 39 0 1.715 3.603 0.577 1.168 Column Range Max Min Median 25% 75% Surf: Lindane 2.490 2.500 0.01000 1.950 1.275 2.100 Subsurf: Lindane 22.990 23.000 0.01000 1.050 0.414 2.038 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Lindane -1.356 1.252 0.239 <0.001 47.120 90.556 Subsurf: Lindane 5.692 34.303 0.388 <0.001 66.890 608.002



Descriptive Statistics: Sunday, December 11, 2005, 2:44:55 PM Data source: Toxaphene: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Toxaphene 14 0 28.954 20.800 5.559 12.010 Subsurf: Toxaphene 32 0 34.763 47.940 8.475 17.284 Column Range Max Min Median 25% 75% Surf: Toxaphene 64.150 69.000 4.850 24.500 10.000 42.000 Subsurf: Toxaphene 202.550 204.000 1.450 10.000 7.250 62.000 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Toxaphene 0.682 -0.627 0.177 0.263 405.350 17360.772 Subsurf: Toxaphene 1.968 3.996 0.359 <0.001 1112.400 109915.795

Descriptive Statistics: Sunday, December 11, 2005, 2:46:07 PM Data source: Mean PEC-Q: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surf: Mean PEC-Q 891 0 11.046 98.438 3.298 6.472 Subsurf: Mean PEC-Q 1156 0 24.613 438.738 12.904 25.318 Column Range Max Min Median 25% 75% Surf: Mean PEC-Q 2329.998 2330.000 0.00245 0.253 0.126 0.758 Subsurf: Mean PEC-Q 14700.000 14700.000 0.000380 0.286 0.0859 1.550 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surf: Mean PEC-Q 17.387 367.770 0.455 <0.001 9842.148 8732824.919 Subsurf: Mean PEC-Q 32.533 1086.681 0.478 <0.001 28452.846 223027285.873



Descriptive Statistics: Sunday, December 11, 2005, 2:47:17 PM Data source: TOC: surface and subsurface in Phase III DB descriptive stats.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean Surface: TOC 580 0 4.252 5.126 0.213 0.418 Subsurf: TOC 770 0 4.446 5.602 0.202 0.396 Column Range Max Min Median 25% 75% Surface: TOC 50.110 50.120 0.01000 3.200 1.575 4.800 Subsurf: TOC 47.990 48.000 0.01000 2.920 1.200 4.870 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares Surface: TOC 4.178 25.460 0.215 <0.001 2465.970 25697.790 Subsurf: TOC 3.196 13.696 0.224 <0.001 3423.380 39354.028



APPENDIX C

RESULTS OF MANN-WHITNEY RANK SUM TESTS BETWEEN SURFACE AND SUBSURFACE SEDIMENTS FOR THE LIST OF

CHEMICALS WITH SQTS, AS WELL AS TOC



APPENDIX C

MANN-WHITNEY RANK SUM TEST RESULTS

Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:38:03 PM Data source: Arsenic: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Arsenic 228 0 3.250 2.025 6.300 Subsurf: Arsenic 294 0 4.000 2.200 11.100 T = 56177.000 n(small)= 228 n(big)= 294 (P = 0.044) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.044)

Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:39:34 PM Data source: Cadmium: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Cd 230 0 0.565 0.250 1.200 Subsurf: Cd 271 0 0.520 0.250 1.513 T = 56766.500 n(small)= 230 n(big)= 271 (P = 0.551) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.551)

C-1 MPCA Document Number: tdr-fg05-04


Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:41:16 PM Data source: Chromium: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Cr 309 0 30.000 17.300 42.000 Subsurface: Cr 360 0 31.000 19.600 41.370 T = 101541.500 n(small)= 309 n(big)= 360 (P = 0.429) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.429) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 11:07:34 AM Data source: Copper: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Copper 206 0 25.000 15.300 42.748 Subsurf: Copper 235 0 24.000 13.868 40.000 T = 46426.500 n(small)= 206 n(big)= 235 (P = 0.500) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.500)

Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:43:01 PM Data source: Lead: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Lead 424 0 33.950 13.954 81.000 Subsurf: Pb 510 0 42.000 13.000 110.000 T = 191701.000 n(small)= 424 n(big)= 510 (P = 0.112) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.112)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:44:24 PM Data source: Mercury: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Hg 768 0 0.149 0.0605 0.283 Subsurf: Hg 916 0 0.159 0.0520 0.372 T = 628406.500 n(small)= 768 n(big)= 916 (P = 0.061) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.061) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:45:49 PM Data source: Nickel: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Nickel 200 0 20.000 12.900 26.000 Subsurf: Nickel 220 0 20.679 14.400 26.250 T = 40697.000 n(small)= 200 n(big)= 220 (P = 0.259) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.259) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:47:31 PM Data source: Zinc: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Zn 227 0 120.000 63.250 229.006 Subsurface: Zn 325 0 124.000 59.825 240.750 T = 61957.500 n(small)= 227 n(big)= 325 (P = 0.661) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.661)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:48:53 PM Data source: 2-Methylnaphthalene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: 2-Methylnaph. 335 0 266.660 55.000 1288.830 Subsurf: 2-Meth. 491 0 320.000 53.755 1999.037 T = 135037.500 n(small)= 335 n(big)= 491 (P = 0.301) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.301) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:50:52 PM Data source: Acenaphthene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Acenaphthene 549 0 70.000 16.500 600.000 Subsurf: Acenapthen 600 0 220.575 35.000 1394.240 T = 287944.500 n(small)= 549 n(big)= 600 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:52:40 PM Data source: Acenaphthylene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Acenaph. 552 0 67.000 16.250 407.500 Subsurf: Acenaptyle 613 0 85.000 19.000 621.715 T = 309066.500 n(small)= 552 n(big)= 613 (P = 0.026) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.026)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:54:51 PM Data source: Anthracene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Anthra. 648 0 305.000 58.000 1800.000 Subsurf: Anthrac 781 0 555.550 68.000 4525.000 T = 439550.000 n(small)= 648 n(big)= 781 (P = 0.002) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.002) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:56:31 PM Data source: Fluorene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Fluorene 632 0 165.500 42.000 932.135 Subsurf: Fluorene 776 0 415.330 57.445 3925.000 T = 406940.000 n(small)= 632 n(big)= 776 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:58:04 PM Data source: Naphthalene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Naphth. 552 0 285.500 82.500 3520.750 Subsurf: Naphth. 671 0 735.840 105.000 14582.090 T = 314262.000 n(small)= 552 n(big)= 671 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 1:59:46 PM Data source: Phenanthrene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Phenan. 657 0 670.000 165.000 4749.105 Subsurf: Phenan 786 0 1387.445 165.000 14677.420 T = 447459.000 n(small)= 657 n(big)= 786 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:01:27 PM Data source: Benz(a)anthracene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: benz(a)a 652 0 690.450 164.000 3915.000 Subsurf: benz(a)an 779 0 850.000 135.328 5837.555 T = 458094.000 n(small)= 652 n(big)= 779 (P = 0.262) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.262) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:02:51 PM Data source: Benzo(a)pyrene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: benzo(a)p 651 0 588.000 130.990 2920.000 Subsurf: benzo(a)py 757 0 549.010 90.000 3712.500 T = 462501.500 n(small)= 651 n(big)= 757 (P = 0.611) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.611)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:04:34 PM Data source: Chrysene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Chryse. 653 0 740.000 172.933 3700.000 Subsurf: Chrysene 779 0 827.000 136.502 5688.043 T = 462369.000 n(small)= 653 n(big)= 779 (P = 0.480) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.480) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:17:05 PM Data source: Dibenz(a,h)anthracene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Dibenz(ah). 546 0 97.500 22.900 550.000 Subsurf: Dibenz(ah) 610 0 71.500 22.500 855.260 T = 311532.500 n(small)= 546 n(big)= 610 (P = 0.445) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.445) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:18:45 PM Data source: Fluoranthene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Fluoranth. 660 0 1265.365 280.000 7059.800 Subsurf: Fluoranth 787 0 1774.190 176.602 14000.000 T = 463170.500 n(small)= 660 n(big)= 787 (P = 0.064) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.064)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:20:03 PM Data source: Pyrene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Pyrene 660 0 1100.000 274.000 5976.190 Subsurf: Pyrene 788 0 1639.785 165.000 11402.175 T = 464444.000 n(small)= 660 n(big)= 788 (P = 0.083) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.083) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:21:16 PM Data source: T. PAHs (13): surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: T. PAHs (13) 661 0 6942.910 1542.000 40971.000 Subsurf: T.PAHs(13) 789 0 11710.000 1484.165 90736.958 T = 456733.500 n(small)= 661 n(big)= 789 (P = 0.004) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.004) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:22:33 PM Data source: T. PCBs-reported: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: T. PCBs report 71 0 150.000 108.135 267.380 Subsurf: T.PCBs-rep 113 0 77.220 39.743 252.025 T = 7711.000 n(small)= 71 n(big)= 113 (P = 0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.001)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:24:12 PM Data source: T. PCBs-reported only: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: T. PCBs-rep only 53 0 50.000 20.938 68.563 Subsurf: T.PCBs-ro 180 0 17.000 9.900 48.500 T = 7676.500 n(small)= 53 n(big)= 180 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:25:57 PM Data source: T. Aroclor PCBs: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: T. Arocl. PCBs 82 0 199.750 155.000 287.000 Subsurf: T. Ar.PCBs 48 0 192.500 123.375 277.500 T = 2989.500 n(small)= 48 n(big)= 82 (P = 0.458) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.458) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:27:28 PM Data source: T. Congener PCBs: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: T. Cong. PCBs 80 0 146.695 91.743 260.945 Subsurf: T.Cong.PCBs 120 0 77.975 40.235 203.900 T = 9233.000 n(small)= 80 n(big)= 120 (P = 0.003) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.003)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:29:05 PM Data source: Total Chlordane (MESL): surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: T. Chlordane 18 0 3.850 3.200 4.300 Subsurf: T.Chlordane 28 0 0.346 0.126 7.100 T = 545.000 n(small)= 18 n(big)= 28 (P = 0.006) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.006) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:30:35 PM Data source: Total Chlordane, reported only: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: T Chlor, rep only 18 0 2.500 2.100 6.500 Subsurf: T.Chlor, ro 36 0 2.450 1.600 3.725 T = 622.500 n(small)= 18 n(big)= 36 (P = 0.020) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.020) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:32:21 PM Data source: Dieldrin: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Dieldrin 63 0 6.500 3.150 20.875 Subsurf: Dieldrin 59 0 3.500 0.449 5.375 T = 2692.000 n(small)= 59 n(big)= 63 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:34:30 PM Data source: Sum DDD: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Sum DDD 39 0 3.350 2.112 8.588 Subsurf: Sum DDD 44 0 1.000 0.335 2.500 T = 2120.500 n(small)= 39 n(big)= 44 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:35:57 PM Data source: Sum DDE: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Sum DDE 40 0 3.350 2.125 10.000 Subsurf: Sum DDE 44 0 2.440 0.825 4.005 T = 1974.500 n(small)= 40 n(big)= 44 (P = 0.014) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.014) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:37:12 PM Data source: Sum DDT: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Sum DDT 65 0 7.500 2.675 25.000 Subsurf: Sum DDT 43 0 0.780 0.170 2.500 T = 1293.000 n(small)= 43 n(big)= 65 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:38:36 PM Data source: Total DDT: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: T. DDT 67 0 12.750 8.025 29.500 Subsurf: T. DDT 70 0 9.045 3.055 16.500 T = 5467.500 n(small)= 67 n(big)= 70 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:40:01 PM Data source: Endrin: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: Endrin 72 0 11.250 3.350 28.000 Subsurf: Endrin 60 0 2.500 0.215 5.250 T = 2684.500 n(small)= 60 n(big)= 72 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:41:26 PM Data source: Heptachlor epoxide: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Passed (P = 0.108) Equal Variance Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Hepta.epox 27 0 2.000 1.750 2.188 Subsurf: Hepta_epo. 34 0 4.625 3.600 5.500 T = 414.000 n(small)= 27 n(big)= 34 (P = <0.001) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = <0.001)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:44:05 PM Data source: Lindane: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Lindane 28 0 1.950 1.275 2.100 Subsurf: Lindane 39 0 1.050 0.414 2.038 T = 1114.500 n(small)= 28 n(big)= 39 (P = 0.039) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.039) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:45:22 PM Data source: Toxaphene: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Toxaphene 14 0 24.500 10.000 42.000 Subsurf: Toxaphene 32 0 10.000 7.250 62.000 T = 385.000 n(small)= 14 n(big)= 32 (P = 0.185) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.185) Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:46:29 PM Data source: Mean PEC-Q: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surf: Mean PEC-Q 891 0 0.253 0.126 0.758 Subsurf: Mean PEC-Q 1156 0 0.286 0.0859 1.550 T = 903021.000 n(small)= 891 n(big)= 1156 (P = 0.480) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.480)



Mann-Whitney Rank Sum Test Sunday, December 11, 2005, 2:47:51 PM Data source: TOC: surface and subsurface in Phase III DB descriptive stats.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% Surface: TOC 580 0 3.200 1.575 4.800 Subsurf: TOC 770 0 2.920 1.200 4.870 T = 399352.000 n(small)= 580 n(big)= 770 (P = 0.286) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.286)



APPENDIX D

STATISTICAL ANALYSIS OF PAH SOURCE RATIOS



APPENDIX D

DESCRIPTIVE STATISTICS

Descriptive Statistics: Monday, December 12, 2005, 4:27:58 PM Data source: 0 - 5 cm in Ph III_PAH_Source_Ratios.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean P/A 114 0 3.554 1.418 0.133 0.263 F/P 114 0 1.141 0.188 0.0176 0.0349 Column Range Max Min Median 25% 75% P/A 7.120 8.200 1.080 3.410 2.430 4.470 F/P 1.210 1.590 0.380 1.150 1.060 1.260 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares P/A 0.614 0.492 0.0591 0.403 405.160 1667.167 F/P -0.742 2.534 0.100 0.007 130.130 152.550

Descriptive Statistics: Monday, December 12, 2005, 4:31:02 PM Data source: 0 - 15 cm in Ph III_PAH_Source_Ratios.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean P/A 250 0 3.138 2.187 0.138 0.272 F/P 250 0 1.261 1.046 0.0662 0.130 Column Range Max Min Median 25% 75% P/A 19.480 19.600 0.120 2.785 1.520 4.450 F/P 16.510 16.550 0.0400 1.210 1.030 1.380 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares P/A 2.256 12.807 0.0876 <0.001 784.570 3653.220 F/P 12.708 185.061 0.316 <0.001 315.250 669.934

D-1 MPCA Document Number: tdr-fg05-04


Descriptive Statistics: Monday, December 12, 2005, 4:34:47 PM Data source: 0 - 30 cm in Ph III_PAH_Source_Ratios.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean P/A 303 0 2.560 1.749 0.100 0.198 F/P 303 0 1.354 1.791 0.103 0.202 Column Range Max Min Median 25% 75% P/A 15.590 15.690 0.1000 2.120 1.550 2.990 F/P 31.670 31.670 0.000 1.260 1.070 1.410 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares P/A 2.783 13.270 0.163 <0.001 775.600 2909.095 F/P 16.208 274.583 0.382 <0.001 410.380 1524.106

Descriptive Statistics: Monday, December 12, 2005, 4:38:07 PM Data source: 15 - 30 cm in Ph III_PAH_Source_Ratios.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean P/A 49 0 4.320 1.775 0.254 0.510 F/P 49 0 1.126 0.264 0.0378 0.0759 Column Range Max Min Median 25% 75% P/A 10.820 11.820 1.000 4.400 3.210 5.170 F/P 1.450 1.570 0.120 1.160 1.007 1.313 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares P/A 1.199 5.607 0.130 0.037 211.690 1065.806 F/P -1.477 4.080 0.135 0.026 55.160 65.450



Descriptive Statistics: Monday, December 12, 2005, 4:41:07 PM Data source: 30 - 45 cm in Ph III_PAH_Source_Ratios.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean P/A 36 0 4.649 2.339 0.390 0.791 F/P 36 0 1.232 0.231 0.0385 0.0782 Column Range Max Min Median 25% 75% P/A 11.220 12.220 1.000 4.590 3.645 5.490 F/P 1.030 1.850 0.820 1.225 1.065 1.385 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares P/A 0.969 2.344 0.164 0.016 167.370 969.569 F/P 0.327 0.178 0.0887 0.616 44.350 56.505

Descriptive Statistics: Monday, December 12, 2005, 4:42:52 PM Data source: >30 cm in Ph III_PAH_Source_Ratios.SNB Column Size Missing Mean Std Dev Std. Error C.I. of Mean P/A 377 0 3.265 3.937 0.203 0.399 F/P 377 0 1.232 0.309 0.0159 0.0312 Column Range Max Min Median 25% 75% P/A 61.130 61.280 0.150 2.380 1.725 3.780 F/P 3.930 4.380 0.450 1.220 1.000 1.380 Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares P/A 9.358 127.312 0.272 <0.001 1231.040 9847.586 F/P 2.947 28.178 0.123 <0.001 464.370 607.778



MANN-WHITNEY RANK SUM TEST RESULTS Mann-Whitney Rank Sum Test Monday, December 12, 2005, 4:47:48 PM Data source: Rank Sum Test Comparison in Ph III_PAH_Source_Ratios.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% P/A: 0 - 30 cm 312 2 2.120 1.550 3.000 P/A: >30 cm 386 2 2.415 1.755 3.795 T = 99617.000 n(small)= 310 n(big)= 384 (P = 0.002) The difference in the median values between the two groups is greater than would be expected by chance; there is a statistically significant difference (P = 0.002)

Mann-Whitney Rank Sum Test Monday, December 12, 2005, 4:49:48 PM Data source: Rank Sum Test Comparison in Ph III_PAH_Source_Ratios.SNB Normality Test: Failed (P < 0.050) Group N Missing Median 25% 75% F/P: 0 - 30 cm 312 2 1.260 1.070 1.420 F/P: > 30 cm 386 2 1.220 1.000 1.380 T = 111660.000 n(small)= 310 n(big)= 384 (P = 0.134) The difference in the median values between the two groups is not great enough to exclude the possibility that the difference is due to random sampling variability; there is not a statistically significant difference (P = 0.134)


Documents

Phase III GIS-based Sediment Quality Database for the St. Louis … · 2012. 10. 4. · Phase III GIS-based Sediment Quality Database for the St. Louis River Area of Concern (AOC)—