WI DNR Field Procedures Manual
Intranet Edition

Introduction

Contaminated sediments are a chronic and sometimes toxic problem affecting some of Wisconsin's inland lakes and streams and Great Lakes coastal areas, especially where industrial and other human activity has been most intense. Contaminated sediments can become a storage sink and possible resource for contaminants to be released to the environment. Suspended contaminated sediments can move through a water system to be deposited, sometimes over a wide area, in previously clean waters. Contaminants from sediments have been shown to move into the food chain, and thus be available for consumption by fish, wildlife and humans. Contaminants in sediments can also affect the base of the food chain by affecting existing resident biota communities.

Specific consequences of contaminated sediments in the state of Wisconsin include for example: contaminated fish and the need for advisories on the consumption of these fish; reduced vitality and diversity of species in an ecosystem; restrictions on dredging activities and costly handling of dredge spoils; expense to businesses and the taxpayers for the remediation of contaminated sites; unpleasant aesthetics and reduced recreational useability of some water bodies.

Although there are many adequate methods for sampling sediments, the intent of this guidance is to present the few most desirable methods to be used for sediment work performed for or by the Wisconsin Department of Natural Resources. The methods presented herein are, at this time, well standardized (i.e., EPA), reliable, effective, repeatable, and deemed most economical. Standardization of sediment assessment procedures should help increase the reliability and quality of the sediment data generated for Wisconsin waters, and will consequently benefit those who must use the data for decision making.

This guidance presents information generally in the order in which one would approach a sediment assessment project. However, later stages of a project including sampling and analysis must be collectively considered in the initial planning stage to insure that all conditions for the assessment and data quality will be met throughout the study. A well planned assessment is most likely to produce complete and reliable data.

Planning a Sediment Survey

1. Scope

   The goal of this section is to provide a short summary of the Department's sediment assessment activities as well as a basic framework (including where to go for additional information) for designing and planning the types of sediment quality assessments most often carried out by the Department. In addition to what is presented here, much literature exists that describes in detail principles that should be followed during the planning of a sediment assessment study. This literature as well as Sediment Management and Remediation Techniques program (SMART Program) staff (in the Bureau of Watershed Management) should be consulted when planning a sediment assessment project to assure a well planned and effective study that provides quality data.

2. Sediment Assessment Activities within WDNR

   There are generally three types of Department activities under which sediment assessment studies are conducted by the Department:

   1. Condition monitoring - Condition monitoring of sediments are conducted within the framework of the Department's annual water resources monitoring plan and basin assessment cycle whenever possible and appropriate. Monitoring strategy activities include: 1) the discovery of suspected and previously unknown contaminated sediment sites; 2) confirmation and definition of the extent of contamination and ecological and/or human health threat at contaminated sites; and 3) monitoring to evaluate effectiveness of a remediation project.

   2. Special projects linked to basin assessments - These are special projects usually designed to fully evaluate the spatial and chemical extent of sediment contamination including the possible ecological and human health impacts. These projects are implemented through the annual surface water monitoring plan in conjunction with the basin assessment cycle.

   3. Special projects - Any special project for research or other purposes that does not naturally fall within the auspices of the annual surface water monitoring plan and basin assessment cycle.
3. Basic Principles of Study Design

   "The specific design of a sediment assessment depends on the objectives of the study. Therefore, it is essential that the study objectives be sufficiently detailed to adequately guide a sediment toxicity evaluation (or invertebrate survey or chemical survey)." (EPA 1994). Because the study design governs the successful outcome of the study, improper study design can nullify the best field collection, laboratory, and data analysis methods (EPA 1994).

   Following are ten principles of study design quoted directly from Green (1979). They describe the different aspects of study design that should be adhered to to assure a resultant quality data set that answers the questions posed in the study.

   1. "Be able to state concisely to someone else what question you are asking. Your results will be as coherent and as comprehensible as your initial conception of the problem."

      Seek statistical help during the design of the experiment, before any samples are taken. This also helps to ensure that all collected data is used to a meaningful end and that raw data is not collected without a purpose in mind. Money is saved by not wasting field time and analysis dollars.

   2. "Take replicate samples within each combination of time, location, and any other controlled variable. Differences among can only be demonstrated by comparison to differences within."

   3. "Take an equal number of randomly allocated replicate samples for each combination of controlled variables. Putting samples in "representative" or "typical" places is not random sampling."

   4. "To test whether a condition has an effect, collect samples both where the condition is present and where the condition is absent but all else is the same. An effect can only be demonstrated by comparison with a control."

   5. "Carry out some preliminary sampling to provide a basis for evaluation of sampling design and statistical analysis options. Those who skip this step because they do not have enough time usually end up losing time."

   6. "Verify that your sampling device or method is sampling the population you think you are sampling, and with equal and adequate efficiency over the entire range of sampling conditions to encountered. Variation in efficiency of sampling from area to area biases among-area comparisons."

   7. "If the area to be subsampled has a large-scale environmental pattern, break the area up into relatively homogeneous subareas and allocate samples to each in proportion to the size of the subarea. If it is an estimate of total abundance over the entire area that is desired, make the allocation proportional to the number of organisms in the subarea."

   8. "Verify that your sample unit size is appropriate to the sizes, densities and spatial distributions of the organisms you are sampling. Then estimate the number of replicate samples required to obtain the precision you want."

   9. "Test your data to determine whether the error variation is homogeneous, normally distributed, and independent of the mean. If it is not, as will be the case for most field data, then (a) appropriately transform the data, (b) use a distribution-free (non-parametric) procedure, (c) use an appropriate sequential sampling design, or (d) test against simulated Ho data."

   10. "Having chosen the best statistical methods to test your hypothesis, stick with the result. An unexpected or undesired result is not a valid reason for rejecting the method and hunting for a 'better' one."
4. Forming Study Objectives

   Coherent study objectives must be formulated during the planning stage of a sediment assessment, before field work is started. They are the questions that the data is expected to answer with some determined degree of accuracy (see also Data Quality Objectives in 7. Quality Assurance/Quality Control). The study objectives should be concise statements of the questions that the study results are expected to answer. They should include as many specifics as possible about the problem being studied, and may include possible actions that will be taken as a result of the study results. For more information about study objective and hypothesis formulation see Green (1979).

5. Statistical Needs

   Statistics can be a useful tool for objective data analysis in many sediment assessment studies. Because many statistical techniques contain requirements that the data must meet to be valid, such as randomization and replication of samples, it is important to plan the statistical tests to be used in the data analysis during the planning stage of the study. Consultation with a statistician during the planning stage of the study is highly recommended, but be sure to have the study objectives in hand. Planning ahead of sampling and field work ensures that enough quality data from replicated and well placed samples will be collected to permit valid statistical techniques to be utilized on the data.

   The decision to use qualitative or quantitative samples for invertebrate studies should be made on the basis of which type can meet the needs of the study objectives. Quantitative samples allow "...the use of simple but powerful statistical tools that aid in maintaining the objectivity of the data evaluation process." (Klemm et al. 1990).

   An explanation and discussion of specific statistical applications is beyond the scope of this guidance, but plenty of textbooks and literature exist to provide necessary information for most purposes. Many of the specific environmental sampling methods literature contain sections reviewing basic statistical applications and statistical techniques directly applicable to the method (Klemm et al. 1990; EPA 1992; Burton 1992). These can be a valuable source of data analysis information for a specific subject.

6. Sampling Design

   A complete discussion of sampling design is not possible here, but a summary of important points to be considered are listed below. A variety of reference materials exist that discuss sampling design in detail for a variety of study types (Klemm et al. 1990; EPA 1992; Burton 1992; EPA 1994). It is recommended that relevant references be considered before designing a study (see part 8. References of this section). Some important considerations about study design derived from these references as well as recommendations by the Department are summarized below.

   1. Reference and Control Samples

      Always use reference site samples to compare to the test site samples. Without a reference or control sample, the test data may have little or no meaning because there is no base level with which to compare test site results. For example, when sediment is collected for toxicity or bioaccumulation tests or benthic invertebrate surveys, at least one reference site that is ecologically similar to the test sites should be included into the study. Ideally, only the variable being tested (mercury concentration in the sediment for example) should differ between test and reference sites. If many test sites are involved in a study where the ecological conditions vary greatly, a reference site should, if possible, be chosen for each ecological type of site. Since conditions in the real environment are rarely ideal, the investigator must use his/her own biological knowledge to choose the best possible reference sites for any particular study, and then to further keep in mind the drawbacks any reference site may have when interpreting the data.

      The reference site can be the most important site in some cases. Probable causes of detected impact can often be determined to be more or less likely because of the data collected from the reference site. Further and more detailed information on reference site selection can be found in Green (1979), Burton (1992), Klemm et al. (1990), EPA (1994), to name a few.

      Control samples are commonly used in laboratory tests or field sampling efforts for quality control (see 7. Quality Assurance/Quality Control below).

   2. Overall Site and Sampling Station Selection

      A sediment site can be a homogeneous or heterogeneous sediment deposit or sampling area of almost any size, and can be comprised of just one or many sampling stations (see below). The area of a site is defined by the investigator (and nature), and should be chosen to best answer the questions posed in the study objectives with the degree of accuracy and precision deemed necessary for the purpose of the study.

      Sampling stations are sampling points located within a sediment site so as to best characterize the sediment as per the requirements of the study objectives. In general, the more precision and accuracy desired, and/or the larger and more heterogeneous a site is, the more sample stations that will be necessary at a site to obtain adequate data.

      Physical and chemical factors to consider when selecting sampling sites, especially for invertebrate collections, include: substrate or sediment type (particle size, organic carbon content); water depth; submerged vegetation, temperature average and fluctuation; sunlight; turbidity; pH; dissolved oxygen; current velocity.

      Besides a historical review of site information, a reconnaissance study of potential sites may be advisable to explore for pollution sources, human disturbances, and physical characteristics as well as take preliminary samples for the purpose of estimating the minimum number and placement of samples to provide the necessary accuracy and precision for the main study.

      To minimize uncertainty in any environmental data, variability between sites (physical and chemical characteristics of overall sampling sites), except for the variable(s) in question (for example, contaminant concentrations, location from point source, etc.) should be minimized. Taking aquatic invertebrate samples from physically similar sites and stations will decrease the variability between samples and thus decrease the number of samples required to detect population differences or allow detection of less obvious differences (Klemm et al. 1990).

      Always have at least one reference or control station to compare to suspected "impaired sites". Ideally, the reference site is ecologically similar to the test sites except for the variable in question (i.e., pollutant discharge, contaminant concentration, etc.).

      For any site, invertebrate and sediment samples for chemical and physical analysis should be taken in close proximity to one another, without interfering with one another, so that chemical concentration can be legitimately compared to the benthic invertebrate population.

      Replicate samples should always be taken, although the number and placement should be depend on the study objectives data quality objectives, collection methods, and resources available. A minimum of three replicates are recommended for any collection method and is deemed adequate for artificial substrate type samplers (Mason et al., 1973 in Klemm et al. 1990). Five replicate samples for benthic invertebrate collections are recommended and will increase statistical precision and accuracy, allowing better definition of less obvious population differences between sites (Klemm et al. 1990).

      A random sampling method should be used when quantitative data is required. For qualitative or semi-quantitative data, a non-random sampling scheme is acceptable (Klemm et al. 1990).

      Suitable references should be consulted for detailed descriptions of acceptable random and non-random sampling methods for different study and site types (Klemm et al. 1990; Green, 1979; EPA, 1985).

7. Quality Assurance/Quality Control

   1. Data Quality Objectives

      Data quality objectives (DQO) are statements describing the level of uncertainty or accuracy required for any given study, and should be developed at least to some degree for every sediment assessment undertaken by the WDNR. This may be done informally, but any person managing a study should understand and be able to summarize on paper for all staff involved the level of certainty the data must meet to accomplish the goals of the study. Klemm et al. (1990) summarizes three stages of DQO development:

      "During the first stage, the decision-maker determines what information is needed, reasons for the need, how the information will be used, and specifies time and resource constraints. The second stage involves the technical staff and decision-maker interacting to establish a detailed and clarified specification of the problem. how the information will be used, any constraints imposed on the data collection, and what limitations of the information will be acceptable. The third stage involves the analysis of possible approaches to collection and analysis of the data and a determination of the quality of the data that can be expected to result from each approach. The best approach is selected based on the criteria agreed upon in the second stage."

      Details about the development of data quality objectives can be found in three Environmental Protection Agency documents (1984, 1986 and 1994).

   2. Measurement Quality Objectives

      EPA (1994), in its Assessment and Remediation of Contaminated Sediments (ARCS) program "Assessment Guidance Document", lists six measurement quality objectives that should be considered when developing a sampling design and quality assurance plan for a study. The EPA document discusses each one of the following in greater detail than is allowed here. "The measurement quality objectives should be defined according to the following six quality assurance objectives and attributes:

      - Detection Limit-The lowest concentration of an analyte that a specified analytical procedure can reliably detect

      - Bias-The difference between an observed value and the 'true' value (or known concentration) of the parameter being measured; bias is the first component of accuracy, which is the ability to obtain precisely a non-biased (true) value

      - Precision-The level of agreement among multiple measurements of the same characteristic; precision is the second component of accuracy

      - Representativeness-The degree to which the data collected accurately represent the population of interest (e.g., contaminant concentrations)

      - Comparability-The similarity of data from different sources included within individual or multiple data sets; the similarity of analytical methods and data from related projects across AOCs [Areas of Concern]

      - Completeness-The quantity of data that is successfully collected with respect to the amount intended in the experimental design."

      See also WDNR (1990) "Quality Assurance Guidance for Inplace Pollutant Monitoring".

   3. Quality Assurance and Quality Control Samples

      Various types of samples taken before and during sampling can and should be used to check for error in the study results that may occur from sampling efforts or in the laboratory. These include field or laboratory replicate samples and field blank or laboratory control samples. Replicate field samples can either be individual (fresh sediment for each sample) or split (mix and split a single sample) samples that are then submitted to the laboratory as separate or blind samples. Individual samples will be less likely to be similar depending on the types of samples taken (i.e., invertebrate, bulk chemistry sediment, etc.) and the homogeneity of the site. Split samples, if well mixed, should be very similar and should be an indication of variability within the single sample, assuming there is no error occurring from the laboratory analysis. Laboratory replicate samples which are collected by sub-sampling a single sample in the lab and analyzing the two new samples separately, can be used to assess analytical precision within the laboratory.

      A field or trip blank can be used to check for contamination of samples collected in the field. A verified "clean" sample of water or sample reagents are taken into the field and subjected to all of the manipulations and equipment that a site sample would go through. Contamination of this blank sample would indicate that the other samples are probably likely contaminated. To measure the variability of contamination from sample to sample, replicate blanks would need to be "taken".

      When collecting samples on a time scale such as during exposure of test organisms to a sediment to assess bioaccumulation potential, a time=0 or "blank" sample (two are preferred) must be taken to show what concentration of contaminants the test organism contained immediately before the beginning of the test. It is important to collect the time=0 sample when the test begins rather than earlier. An example of the importance of this occurred when minnows were shipped to a laboratory two weeks prior to the beginning of a bioaccumulation test exposure, and maintained and fed frozen brine shrimp for those two weeks (WDNR 1993b). A sample of fish was collected upon arrival to the lab as well as at the beginning of the tests (two weeks later). It was discovered that the minnows accumulated significant amounts of mercury during the two week holding period, and remained at that level until the end of the test period, suggesting that the tank, water and/or food contributed the contaminating mercury. The brine shrimp was then analyzed and found to contain significant concentrations of mercury. If the minnows had not been analyzed from both the arrival and test start dates, the contamination from the food would not have been discovered. If only minnows from arrival time and test end time had been analyzed, the results would have shown an increase in mercury between the two times, but uncertainty would exist as to whether the test sediment or the holding tank was the source.

   4. Basic Quality Control Measures

      The following are measures or actions that should be taken during a sediment sampling and analysis to assure quality data (1. through 5. from Klemm et al. 1990).

      1. Follow approved methods for sampling and sample handling and analysis procedures whenever possible.

      2. Calibrate and test all sampling and measuring devices following the manufacturers instructions. Document tests and calibrations for future reference.

      3. Determine and complete replicated samples at all phases of sampling deemed necessary.

      4. Collect representative samples for each site such that the needs of the study are met.

      5. Record accurate and complete field information. Make sure all data is accurately recorded and stored so as to be readily retrievable.

      6. Communicate clearly with, and include in any contract, all expectations you have for quality control practices to be followed by any lab that will be performing sample analysis work.

      7. Keep very good records of all samples collected, including sample identifications and purpose, and destination. Lost samples can result in difficulty in data analysis and a decreased statistical power to detect differences.

      8. Keep accurate and detailed records of all field data and measuring instrument calibrations and tests.

      9. Provide standard operating procedures to all staff involved to insure all samples are handled similarly and without introduced contamination.

      10. Be aware of and prevent possible routes of sample contamination during all phases of sample collection and handling. Examples include: boat motors, vehicle surfaces, dirty sampling equipment, boat surfaces.

          A summary of quality assurance/quality control information for sediment sampling can also be found in Chapter 2 of EPA's (1992) "Sediment classification methods compendium". The chapter discusses data quality objectives, sampling design, terminology, sources of error, quality assurance project plans, and sample collection and handling.

   5. Chain of Custody

      Refer to section VI for information about chain of custody procedures to be followed for samples that may be used as legal evidence for the Department.

8. References

   Baudo, R., Giesy, J., and H. Muntau, (Eds.). 1990. Sediments: Chemistry and Toxicity of In-Place Pollutants. Lewis Publishers, Boca Raton, FL.

   Burton, G.A. Jr. (Ed.) 1992. Sediment Toxicity Assessment. Lewis Publishers, Boca Raton, Florida. 457pp.

   EPA. 1984. The development of data quality objectives. Prepared by the EPA Quality Assurance Management Staff and DQO Workshop. U.S. Environmental Protection Agency, Office of Research and Development, Washington, DC 20460.

   EPA. 1985. Sediment sampling quality assurance user's guide. Environmental Monitoring Systems Laboratory. Las Vegas, Nevada. EPA/600/4-85/048.

   EPA. 1986. The development of data quality objectives. Description of stages I and II. Prepared by the EPA Quality Assurance Management Staff. U.S. Environmental Protection Agency, Office of Research and Development, Washington, DC 20460.

   EPA. 1992a. Sediment classification methods compendium. Office of Water, Washington, DC. EPA 823-R-92-006.

   EPA. 1992b. Draft. Biological criteria: Technical guidance for survey design and statistical evaluation of biosurvey data. U.S. Environmental Protection Agency, Office of Water, Office of Science and Technology, Washington, DC.

   EPA. 1994. Assessment and remediation of contaminated sediments (ARCS) program, Assessment Guidance Document. Great Lakes National Program Office, Chicago, Illinois. EPA 905-B94-002.

   Green, Roger H. 1979. Sampling design and statistical methods for environmental biologists. John Wiley & Sons. New York. 257 pp.

   Klemm, D.J., P.A. Lewis, F. Fulk, and J.M. Lazorchak. 1990. Macroinvertebrate field and laboratory methods for evaluating the biological integrity of surface waters. Environmental Monitoring Systems Laboratory, U.S. Environmental Protection Agency, Cincinnati, OH 45268. EPA/600/4-90/030.

   Mason, W.T., Jr., C.I. Weber, P.A. Lewis, and E.C. Julian. 1973. Factors affecting the performance of basket and multiplate macroinvertebrate samplers. Freshwater Biology. 3:409-436.

   WDNR. 1990 (draft). Quality Assurance Guidance for Inplace Pollutant Monitoring Activities. Unpublished document on file at Office of Technical Services, Bureau of Watershed Management.

   WDNR. 1993. In situ caged fish and laboratory bioaccumulation study with Lake Superior and Green Bay coastal sediments. PUBL WR 317-93. Sediment Management and Remediation Techniques program in Bureau of Watershed Management, Wisconsin DNR, Madison, Wisconsin.

Rev. 1, April 1995

This document is intended solely as guidance and does not contain any mandatory requirements except where requirements found in statute or administrative rule are referenced. This guidance does not establish or affect legal rights or obligations and is not finally determinative of any of the issues addressed. This guidance does not create any rights enforceable by any party in litigation with the State of Wisconsin or the Department of Natural Resources. Any regulatory decisions made by the Department of Natural Resources in any matter addressed by this guidance will be made by applying the governing statutes and administrative rules to the relevant facts. (From Manual Code 1210.1)