Department of Health Seal

TGM for the Implementation of the Hawai'i State Contingency Plan
Section 21.6

21.6 Baseline Ecological Risk Assessment

After completing the SLERA, including the Step 3a refinement, the risk assessor is ready to begin Step 4: the BERA. The first task of Step 4 is to prepare a BERA work plan (WP). If additional field data collection is required, the WP may include a field sampling and analysis plan (SAP). Typically, a combined WP/SAP is prepared to streamline the planning and approval process before BERA data collection begins.

The purpose of preparing a BERA WP is two-fold: (1) it compels the risk assessor to thoroughly evaluate existing data, describe site conditions, formulate DQOs, identify data gaps, and anticipate issues that may arise during later risk characterization and data interpretation phases; and (2) it provides a site-specific framework for discussions with the HEER Office during which information can be shared and common goals can be established. This subsection guides the risk assessor through the tasks typically included in the BERA, describes best practices, and reviews technical references to support the process. This subsection assumes that a combined WP/SAP is being prepared. The process of developing the BERA WP/SAP is described below.

  1. Review the SLERA and ensure that you have access to all available data that contributed to the conclusions of the SLERA.
  2. Compile any pertinent information collected since the SLERA was prepared. If any new information leads you to question the need for a BERA, present the information and your rationale to the HEER Office for discussion.
  3. Once you are sure that a BERA is appropriate, prepare a BERA WP/SAP using the outline in Appendix 21-G. The rest of this subsection will provide templates and examples to help you develop the BERA WP.
  4. Notify the HEER Office that you are preparing a BERA WP/SAP and request additional guidance as needed.
  5. Submit the draft BERA WP/SAP to the HEER Office well before you expect to begin field work.

As described in previous subsections, the SLERA usually relies on literature-based toxicity and bioaccumulation factors and conservative default assumptions about exposure because site-specific data are not available. The purpose of the BERA is to replace literature or default values with site-specific data so that risk can be more accurately characterized. Site-specific data collection may include toxicity and bioaccumulation tests, collection of organisms, passive sampling of water or sediment, analysis of TOC and grain size, and other types of information. In addition to collection of new data, a more detailed analysis of data available during the SLERA may be warranted.

The components of the BERA mirror those of the SLERA. First, the problem formulation is refined to better describe the environmental setting, ecological receptors, and complete exposure pathways, resulting in a revised CSM (Subsection 2 of the BERA WP/SAP). Then, exposure and effects estimates are updated using site-specific information. The study design for collecting and analyzing new data is in Subsection 3 of the BERA WP/SAP (Study Design and DQOs). Elements of the BERA are presented in Subsections 21.6.1 through 21.6.4 below.

Although each BERA WP/SAP will represent site-specific conditions and address unique considerations, most or all can be prepared using the template in Appendix 21-G. The template provides general direction on which elements should be included in a site-specific BERA WP/SAP and includes useful tips. The HEER Office does not require that the risk assessor follow the template exactly, but it is important that all the necessary components of the BERA be included in the WP/SAP. The full set of topics to be included in the BERA will be determined by the location and geophysical features of the site, the site-specific COPECs, the selected assessment and measurement endpoints, and complete exposure pathways.

21.6.1 BERA Refined Problem Formulation

The problem formulation subsection serves as the “backbone” of the ERA. The SLERA problem formulation (described in Subsection 21.3.3) included a description of the environmental setting, including ecological receptors, potential sources of contamination, and potential exposure pathways, which were used to develop the preliminary CSM. At the start of the BERA, the problem formulation is refined to reflect the conclusions from the SLERA.

The result of Step 3a is a list of COPECs that require further evaluation in the BERA and a list of chemicals eliminated from further evaluation because they were found not likely to cause significant risk. Ideally, the BERA will focus only on chemical-receptor pairs posing potential risk. Careful completion of this step will prevent the risk assessor from wasting time and effort evaluating chemicals in the BERA that should have been screened out during Step 3A.

The refined problem formulation should also identify any data gaps necessary to characterize site-specific risk at the end of the BERA. In some case, information obtained since the SLERA was written may warrant inclusion of chemicals, receptors, or exposure pathways that were not evaluated in the SLERA. For example, the risk assessor may have learned of a historical spill at the site, or a unique habitat with receptors not considered during the SLERA may have been identified. Data gaps identified during review of the SLERA may also require additional lines of investigation. In general, the refined problem formulation should include the environmental setting, COPECs, and assessment and measurement endpoints. Each of these is discussed below.

This subsection of the BERA should describe the environmental setting, COPECs, and sources identified in Step 3a, and ecological receptors. Although much of the site characterization will remain as described in the SLERA, it should be updated with any new information, especially on habitats that will be the focus of the BERA. Sediment Dynamics

The SLERA may have relied on assumptions about sediment grain size based on regional geology, as described in the introduction to Section 21. For example, the area may have been described as depositional based on regional data, habitat, or conservative assumptions. For the BERA, it may be necessary to confirm substrate type and grain size at the site to determine whether the area is depositional to better predict chemical behavior and presence of receptors when refining the CSM. Grain size and wave energy must also be considered when selecting an appropriate reference location for the BERA.

Beaches are eroding more than accreting across Hawaiʻi (Fletcher et al. 2012) and coastal erosion is expected to nearly double over the next few decades across the state (except Kailua Beach on Oʻahu). Nevertheless, sediment dynamics are spatially variable, and areas of erosion and accretion may be separated by only a few hundred meters.  Each small embayment created by rocky headlands is influenced by local wave energy and terrestrial processes, creating a patchwork of erosion and accretion along the shore. The most recent data on coastal erosion and accretion of shorelines on Kauaʻi, Oʻahu, and Maui are available at (Fletcher et al. 2012).  This USGS information should be consulted during the site characterization phase of the BERA. Chemicals of Potential Ecological Concern

The list of COPECs developed at the end of Step 3A should include only those chemicals that exceed background or reference concentrations and ecotoxicological effect levels for receptors at the site. If new information suggests the presence of additional chemicals that were not analyzed during the SLERA, then new chemicals should be added to the BERA WP/SAP. Ecological Receptors (Assessment and Measurement Endpoints)

Measurement Endpoints)

Based on the results of Step 3a, some receptors considered in the SLERA may be eliminated from further evaluation, and others may be added. The refined problem formulation should include only receptors that will be evaluated in the BERA, based on their known or expected presence at the site or their selection as surrogates for species of interest. The HEER Office has prepared species profiles for selected marine species in Hawaiʻi (Appendix 21-A). The appropriate receptors from this list should be considered for evaluation in the BERA, noting that additional exposure information may be needed to quantify risks to some receptors. Note that the list of species in Appendix 21-A is not comprehensive; other species may be evaluated in the BERA if approved in advance by the HEER Office. In the BERA WP/SAP, explain any changes to the list of receptors in the SLERA.

Assessment and measurement endpoints that are commonly evaluated in marine sediment ERAs are summarized in Table 21-5 (see Subsection 21.3.3). This subsection of the BERA should provide rationale for the selected assessment endpoints and describe how each assessment endpoint will be evaluated using the selected measurement endpoints. A table similar to Table 21-5, including the following elements, should be developed for the BERA:

  • Ecological Guild: The functional niche of the receptor (such as benthic invertebrate)
  • Assessment Endpoint: The specific attributes of value for the ecological guild at the organism or population level.
  • Species Evaluated: Table 21-8 lists typical species included in each ecological guild. In the BERA, identify the species that were used to represent the ecological guild, along with the rationale for selecting the species. In some cases, species other than those listed in Table 21-8 may be used based on available data. Use of other species should be presented in the BERA WP/SAP and approved in advance by the HEER Office.
  • Measurement Endpoint: Table 21-5 lists common measurement endpoints for each of the assessment endpoints. In the BERA, present the specific measurement endpoints that were used to evaluate the assessment endpoints, along with the rationale for selecting those endpoints. The measurement endpoints may include some or all the endpoints listed in Table 21-5, and endpoints not listed in the table that are deemed appropriate for the site. Refined Conceptual Site Model

The screening level CSM was developed as part of the SLERA based on what was known about the site at that time, without regard to potential ecological risks. As described in Step 1b, Task 5 (Subsection 21.3.3), the elements of the CSM include (1) ecological receptors present at the site; (2) sources of chemicals in the environment; (3) contaminant transport pathways; and, (4) exposure pathways to the ecological receptors. The same elements are included in the refined CSM, which represent the chemicals, receptors, and exposure pathways evaluated in the BERA.

Appendix 21-C describes the approach for defining the ecological DU. DUs set the boundaries for where the BERA investigations will be conducted. The refined conceptual site model should describe the DUs that were selected for each assessment endpoint evaluated in the BERA and the rationale for selecting them. Refer to the discussion of sediment types at the beginning of Section 21 before identifying DUs. Note also that the size of the DUs is determined in part by the receptors, as home range is an important variable in the evaluation of exposure and effects. The site may contain several DUs designated by sediment type, wave energy, preliminary contaminant concentrations, receptor distribution, and other factors.

21.6.2 BERA Study Design and Data Quality Objectives

The BERA should describe the investigations conducted to evaluate each assessment endpoint, such as chemical analysis, toxicity testing, bioaccumulation studies, biological surveys, and tissue analyses. The DQO process that was followed during the SLERA (see TGM Sections 3, 4, and 5) should be revisited when preparing the BERA WP/SAP. The study design and DQOs should be presented in the BERA WP/SAP and cited in the BERA. Because the BERA WP/SAP will be included as an appendix to the BERA, it is not necessary to repeat the DQO subsection. A Quality Assurance Project Plan (QAPP) should also be prepared as part of the BERA planning effort (see TGM Section 10). Laboratory Analyses

Additional data collected for the BERA are likely to include field samples of sediment, sediment pore water, surface water, groundwater, or even soil (in case where terrestrial erosion is suspected as a transport pathway to the marine site). The BERA WP/SAP should identify analytical methods and detection limits to ensure that detection limits lower than selected screening levels can be achieved.

The HEER Office recommends evaluating chemicals with similar modes of toxicity as “total” concentrations, but analysis of individual constituents may also be necessary. Total concentrations are commonly calculated for HMW PAHs, LMW PAHs, total PAHs, total PCBs, DDT and its breakdown products (total DDTx), and dioxin toxic equivalency quotients (TEQs). Methods for calculating total PCBs and dioxin TEQs are discussed later in this subsection, but the risk assessor is encouraged to review the current literature and determine the most appropriate method for the site. No specific list of constituents or summation method is prescribed because methods are rapidly changing as new technical literature is published, methods are vetted, and best practices are disseminated within the risk assessment community. The BERA WP/SAP should describe the proposed methods of summing constituents and clearly identify the individual constituents to be included in the sum. Relevant literature should be cited to support the proposed methods.

In general, HDOH requires the following when calculating total values:

  • Non-detected values should be assigned a value of zero provided the detection limits were acceptable, as described above.
  • The mean of replicate samples (i.e. triplicates or duplicates) should be used for the calculation.
  • The list of individual constituents included in the total calculations must be given (e.g. see notes at the bottom of Table 21-7 for a list of HMW and LWM PAH totals).

Risks from dioxins/furans should be evaluated by using Toxicity Equivalence Factors (TEFs) to calculate toxicity equivalence concentration (TEQ) as described in the Framework for Application of the Toxicity Equivalence Methodology for Polychlorinated Dioxins, Furans, and Biphenyls in Ecological Risk Assessment (USEPA 2008g). The detected concentration of each dioxin (or furan) in a sample is multiplied by its TEF. The resulting values for each sample are summed to calculate the TEQ Dioxins/Furans for each sample. TEQs should be calculated for birds, mammals, and fish using chemical-specific TEFs for each group; no dioxin TEFs are available for plants and invertebrates.

PCB results historically have been reported as Aroclors (i.e., Aroclor-1254, Aroclor-1260) in BERAs because the early ecotoxicological studies were based on total PCBs expressed as the sum of Aroclors. Although some current studies continue to report effects of total PCBs, newer literature is increasingly focused on one or a small set of the 209 PCB congeners. Each Aroclor originally contained a specific combination of PCB congeners and could be identified by its distinctive chromatographic pattern when is analyzed by gas chromatography. However, as Aroclors age and weather, the chromatographic patterns may change and not be recognizable as standard patterns. Such degradation of Aroclors may cause the laboratory to underestimate the concentration of total PCBs in a sample. (See USEPA 2013c) for more detail on this issue.

Analysis of PCB congeners is considerably more expensive than Aroclors, so the decision of analytical method must be made with care. The HEER Office recommends that PCBs be analyzed as Aroclors during the SLERA. However, if total PCBs are detected at concentrations exceeding the screening level in the SLERA samples, a subset of samples (no less than 10 percent) should be analyzed for all 209 congeners. Note that twelve of the PCB congeners have been designated by the World Health Organization (WHO) as having “dioxin-like” toxicity (Van den Berg et al. 1998). The same process described above to calculate the TEQs for dioxins (USEPA 2008g) can be used to sum the dioxin-like PCBs when site conditions warrant. The BERA WP/SAP should describe the rationale for the selected analytical methods for PCBs (Aroclors, congeners, or a combination of the two) and discuss how the dioxin-like PCBs will be summed if samples are analyzed for PCB congeners. Sediment Sampling

The objectives of the study and availability of existing data play an important role in dictating the sampling design, methods, and equipment. For example, MI sampling should be conducted to determine representative average contaminant concentrations in sediment across a designated DU (see Sections 3, 4, and 5. Section 5.7 of the TGM (Sediment Sampling) discusses issues affecting sediment sampling in more detail.

A wide variety of sampling equipment is available for collecting sediment, but not all equipment is suitable for all sites. For example, grab samplers such as a ponar dredge or Van Veen grabs are capable of sampling only the top several inches of sediment, while sediment corers and vibracores can be used to collect deeper samples if historical chemical concentrations are needed. Other considerations include whether the sediment sample must be undisturbed (as it should be for analyzing volatile organic compounds). Water depth, currents, sediment volume, bottom firmness, and other parameters also influence the likelihood of success of each collection method. When acid volatile sulfides [AVS] are to be analyzed, exposure of the sample to oxygen must be limited. A thorough discussion of the various sediment sampling devices, including advantages and disadvantages of each and the best samplers to use for different types of sediment is presented in Chapter 3 in Methods for Collection, Storage and Manipulation of Sediments for Chemical and Toxicological Analyses: Technical Manual (USEPA 2001i). The BERA WP/SAP should include a complete description of equipment, techniques, and standard operating procedures (SOPs) for all sediment collection methods; references should be cited as needed to support the proposed methods.

The BERA WP/SAP should describe the procedures for any representative sub-sampling of sediment samples in the field. This is a critical component of sample processing and should be based on the objective of the investigation, the COPECs, and the sediment matrix. Typically, processing and representative sub-sampling of MI samples are conducted in the laboratory following an established SOP (see TGM Section 4).

Sediment samples must be collected from the appropriate depth to address the goals of the BERA (as identified in the DQO analysis). General guidance on selecting the appropriate depth for collecting sediment samples in the biologically active zone is in Determination of the Biologically Relevant Sampling Depth for Terrestrial and Aquatic Ecological Risk Assessments (USEPA 2015e). Table 21-12 summarizes the depths of the biotic zone associated with different sediment substrates and lists habitats in Hawaiʻi that may contain that substrate.

Table 21-12. Typical Depths of Biotic Zone  
Depth Sediment Substrate Example Habitat Type
5 cm oligohaline/polyhaline mud Mudflats
5 cm oligohaline sand and marine coastal sand Sandy Beach
10 cm marine coastal mixed and marine offshore sand Seagrass beds
10 to 15 cm estuarine and tidal freshwater environments Stream-fed Estuarine Wetlands

The HEER Office recommends taking the above-referenced guidance into consideration when determining appropriate sampling depths to capture the biotic zone. However, depending on the objective of the investigation, deeper samples (below the biotic zone) may also be needed to characterize vertical extent of contamination.

Special sediment sampling consideration may be warranted for target receptors that ingest sediment directly, as sediment effect levels may not account for the ingestion pathway. Ingestion is the basis for the food chain modeling used to evaluate risk to birds and mammals, but many benthic invertebrates and fish also consume sediment as part of a typical diet. Tissue concentrations of benthic invertebrates may reflect chemicals adsorbed to ingested sediment particles as well as chemicals absorbed directly from sediment and water (Lee et al. 2006; Belzunce-Segarra et al. 2015). To evaluate the sediment ingestion pathway, sample collection methods must ensure that the top layer of fine particles is retained for analysis.

When developing the BERA WP/SAP, sediment sample collection log sheets from the SLERA should be reviewed to determine whether they contain useful information to guide the BERA. For example, if sulfide odors were detected during sediment sampling, then AVS may be present in the sediment. Methods for evaluating bioavailability of metal mixtures in sediment containing AVS are discussed in Procedures for the Derivation of Equilibrium Partitioning Sediment Benchmarks (ESBs) for the Protection of Benthic Organisms: Metal Mixtures (Cadmium, Copper, Lead, Nickel, Silver, and Zinc) (USEPA 2005e).

At some sites, it may be appropriate to use a Dynamic Sampling Approach, in which field analytical methods such as x-ray fluorescence (XRF), immunoassays, or other mobile screening approaches help make quick decisions regarding the need to collect samples in a location. This approach is discussed briefly in Sections 3.10 and 5.5.8 of the TGM. Basically, this approach allows samples to be collected and sites to be characterized more efficiently and quickly than traditional sampling. The costs and benefits of a dynamic sampling approach should be discussed in the BERA WP/SAP. See A Guideline for Dynamic Workplans and Field Analytics: The Keys to Cost-Effective Site Characterization and Cleanup (Robbatt 1997 and USEPA 1997i) for more information on field assessment techniques. Pore Water Sampling

In sediments where sediment pore water is relatively static, contaminants in the pore water are expected to be at thermodynamic equilibrium with the sediment (solid phase), making pore water useful for assessing contaminant levels and associated toxicity (USEPA 2001i). The utility of collecting sediment pore water at a site is influenced by a variety of factors, including the solubility of the chemicals, ongoing sources of chemicals in groundwater, grain size and organic content of sediment, and other factors. Sites where pore water analysis may be appropriate include fine-grained sediments in low energy depositional areas (such as bays and harbors) and nearshore sites where contaminated groundwater is known or suspected to discharge to sediment. The suitability of sediment pore water as an exposure pathway to ecological receptors should be evaluated as part of the DQO process and documented in the BERA WP/SAP.

Pore water collection methods should be tailored to the site and the contaminants of concern. No single method is clearly superior in all cases. For example, peepers are suitable for collecting small volumes of pore water for one or two analyses but are not practical for collecting large volumes required to analyze for numerous chemicals. Fine-grained sediments may be collected in buckets and taken to the laboratory for extraction of pore water by centrifugation. However, centrifugation may overestimate concentrations of freely dissolved contaminants (Cfree) in sediment porewater. Depending on the target receptors, the freely dissolved concentration may be a more appropriate exposure point concentration than the total concentration in pore water. Pore water samples also can be collected directly from the sediment using drive points and pushpoint samplers (Henry samplers). In coarser-grained sediments, especially where contaminants are being continuously discharged, in situ measures may be more practical because coarse grained sediment does not retain much water when collected. Traditional collection of sediment followed by centrifugation would require substantial effort because of the large volume of sediment needed to yield an adequate volume of pore water.

Passive in situ sampling methods may be suitable in cases where collecting large volumes of sediment for centrifugation is impractical. or when other limitations of centrifugation are of concern. For example, when chemicals of concern are volatile or unstable, concentrations in pore water may change as the sediment is transported to the lab and centrifuged.

Pore water in situ sampling methods for coarse-grained sediments are under development. The Laboratory, Field, and Analytical Procedures for Using Passive Sampling in the Evaluation of Contaminated Sediments: User’s Manual (USEPA/SERDP/ESTCP. 2017) provides the most comprehensive review of methods for passive sampling of contaminated sediments. The manual provides guidance on selecting and implementing passive sampling technology to evaluate PCBs, PAHs, and selected metals (cadmium, copper, nickel, lead and zinc) in sediment. Earlier technical reviews of passive sampling are provided in (Ghosh et al. 2014), (Greenberg et al. 2014), (Lydy et al. 2014), (Mayer et al. 2014), and (Peijnenburg et al. 2014).

Passive sampling consists of inserting various materials such as polydimethylsiloxane (PDMS), low-density polyethylene (LDPE), or other similar materials into the sediment for a period of time (usually several weeks or months). The materials are typically mounted on frames and may be enclosed by screens or tubes for protection. The samplers are cleaned with an organic solvent to remove oligomers, plasticizers, and contaminating organic chemicals prior to deployment in the field. In some cases, performance reference compounds (PRCs) are added to the sampler as a quality control for estimating the extent of equilibrium of the target contaminant. After the samplers are retrieved from the sediment, the sampling material is cleaned, the contaminants are extracted, the extract is analyzed for contaminants, and concentrations of cfree are calculated. The sampler can be sectioned prior to extraction, if desired, to investigate vertical concentration gradients.

The BERA WP/SAP should specify the methods of collecting and analyzing sediment pore water will be used and provide rationale for selecting the methods. It is essential that the same collection procedures be used and the pore water be collected at the same depth across the site so that appropriate comparisons can be made (USEPA 2001i). Likewise, the same methods must be used at the reference location. If the pore water concentrations will be compared with water quality criteria, the WP must specify how the cfree concentrations will be interpreted with respect to the dissolved criteria for protection of aquatic life. In some cases, side-by-side analysis of standard dissolved concentrations may be required to establish that the passive sampling methods are representative. Additional methods are discussed in several comprehensive technical references:

  • USEPA 2001i: Methods for Collection, Storage and Manipulation of Sediments for Chemical and Toxicological Analyses: Technical Manual (Chapter 6).
  • Carr et al. 2001: SETAC Technical Workshop on Porewater Toxicity Testing: Biological, Chemical, and Ecological Considerations with a Review of Methods and Applications, and Recommendations for Future Areas of Research
  • Various authors 2014: “Passive Sampling Methods for Contaminated Sediments,” in the SETAC Technical Workshop “Guidance on Passive Sampling Methods to Improve Management of Contaminated Sediments” in Integrated Environmental Assessment and Management (Volume 10) reviews the use of passive samplers to quantify concentrations of chemicals in sediment pore water. (Ghosh et al. 2014, Greenberg et al. 2014, Lydy et al. 2014, Mayer et al. 2014, Parkerton et al. 2014, and Peijnenburg et al. 2014). Surface Water Sampling

The surface water pathway is evaluated by comparing chemical concentrations in surface water with water quality standards based on ecotoxicity. However, surface water should be evaluated only in places where the water has a relatively long residence time so that the exposure duration is meaningful. For example, surface water is not considered a measurable pathway at sites where high energy wave action mixes the water constantly. The HEER Office generally does not recommend collecting surface water samples from high energy environments or areas where considerable flushing occurs. In contrast, surface water could be an important pathway in a protected bay contaminated by a surface release, stream input, or groundwater flow. Surface water samples should be collected if chemicals in groundwater are known or suspected to discharge through sediment into protected surface water areas.

Surface water samples may be analyzed for total or dissolved chemicals, depending on the proposed use of the results. Samples that will be compared with the HEER Office EALs for aquatic life for metals should be analyzed for dissolved fractions, represented by samples passed through a 0.45 micrometer (μm) filter. The filtering step typically takes place in the lab, although field-filtering is an option under special circumstances. The USGS provides comprehensive guidance on proper methods for collecting water samples in the National Field Manual for the Collection of Water-Quality Data: U.S. Geological Survey Techniques of Water-Resources Investigations, Book 9, Chapters A1-A10 (USGS 2018).

Both freshwater and saltwater (marine) standards are available. Freshwater and saltwater standards apply to waters with a dissolved inorganic ion concentration less than and greater than 0.5 parts per thousand (ppt), respectively. Saltwater samples can be analyzed for dissolved constituents only. In freshwater habitats, however, total concentrations from unfiltered samples are better indicators of the concentrations ingested by animals as drinking water and are preferred as inputs to the food chain model (see Step 2, Task 3). Freshwater samples may be split and analyzed as both total and dissolved concentrations. The BERA WP/SAP should clearly indicate and provide rationale for which water quality standards will be applied and which water samples will be filtered.

Sample numbers and locations, sampling equipment, and proposed analyses should be presented in the BERA WP/SAP. Equipment should be selected based on the depth of water to be sampled, volume of water needed, strength of currents, and other logistical factors. For example, if the objective is to collect surface water samples at the surface water-sediment interface to determine whether groundwater discharge is transporting chemicals to surface water, a horizontal water bottle sampler may be appropriate. Alternatively, passive sampling devices can be deployed at the sediment-water interface to measure concentrations over time in a specific area. Passive sampling devices are newer and less standardized but may be acceptable for use at some sites. Regardless of the methods and equipment selected, it is important that site samples and reference area samples be collected in the same way.

The BERA WP/SAP should present a rationale for the selection of devices, equipment, and methods. The procedure should be designed so as to minimize incidental collection of suspended solids with the water sample, as solids can artificially inflate measured chemical concentrations. Such interference can be especially important when relatively hydrophobic chemicals such as PCBs and pesticides are being analyzed. In such cases, side-by-side analyses of filtered and unfiltered samples may be warranted. Biological Surveys

Biological surveys may be conducted as part of a BERA for many reasons:

  1. To document the presence and abundance of ecological receptors at the site, including protected or rare species;
  2. To compare the distribution or abundance of species with reference areas or historical records;
  3. To evaluate the health or integrity of the ecological community; and
  4. To collect tissue samples for chemical analysis (described below) for use in food chain models or critical body residue analyses.

Field surveys may be designed for the reasons listed above, as well as simply to ground-truth the CSM. Unlike sediment and water sampling, which may be conducted by general field teams, biological surveys should be conducted by experienced biologists or ecologists who are prepared to document and interpret what they see in the field. Although a single species or type of organism may be targeted for collection, the presence and condition of other species may inform the BERA. Well-designed biological surveys focus on structured data collection, but a competent field biologist will also make opportunistic findings, such as the presence of unanticipated species; the relative scarcity of individuals where abundance was expected; evidence of degraded habitat such as algal overgrowth, stressed vegetation, or chemical sheens and odors; and other features that are not the direct target of the survey.

The BERA WP/SAP should describe the proposed survey as thoroughly as possible, including but not limited to the elements below:

  • Objectives of the survey
  • Qualifications of the field team
  • Locations to be surveyed (with rationale), and process for adjusting the location when field conditions warrant
  • Relation of survey locations to established DUs
  • Intended dimensions of each survey location (length and width)
  • Survey methods (areal grid, transect, etc.)
  • Sample field forms
  • Protocol for avoiding habitat degradation during survey
  • Protocol for unintended encounters with protected species
  • Temporal requirements of the survey: time of day, season, restrictions based on weather
  • Health and safety issues (to be documented in a separate health and safety plan)
  • Use of survey data (species richness, taxonomic diversity, percent dominant taxa, frequency and dominance of stressor tolerant taxa, etc.)

Surveys at the site should be repeated to the extent feasible at reference locations. The reference locations should be similar in size, substrate (grain size distribution), wave energy, surrounding habitat/land use (i.e. urban, rural, forested, etc.). Field-Collected Tissue Sampling

Organisms may be collected from the site (and reference location) during a biological survey or as a separate activity. Field-collected organisms may contribute in several ways to the BERA:

  1. Whole organisms or body parts may be analyzed for selected chemicals. When appropriate, chemical concentrations in the organisms can be compared with concentrations in sediment to evaluate bioavailability and uptake by the organism. Note that this approach requires that both the organisms and the sediment be relatively immobile.
  2. Organisms may be collected as part of a biological inventory focused on characterizing the health of the community in a given area. Species distribution and abundance, species diversity, age or size class distribution, reproductive condition, and other parameters may be measured.
  3. Organisms may be collected for evidence of disease, which may then be linked to chemical contaminants in the sediment or water. External tumors or lesions may indicate exposure to PAHs, for example. Internal examination may reveal parasites, liver damage, or other evidence of degraded health.

Note that a Special Activity Permit may be required for collecting marine organisms for the BERA, even if the organisms are returned to the water unharmed. The Hawaiʻi Division of Aquatic Resources should be contacted during the BERA planning stages to identify necessary permits (DLNR DAR 2018). Other permits may be required for collecting protected species or certain native species, or for collecting in parks or other specified areas. The risk assessor should coordinate with the Hawaiʻi Division of Forestry and Wildlife (DLNR DFW 2018) to obtain the required permits.

In addition to the elements listed above for biological surveys of any kind, the BERA WP/SAP should fully describe the proposed rationale and methods for collecting and analyzing organisms, including at a minimum the following details:

  • Objective of the collection effort
  • Target species to be collected and alternate species in the event that the target species cannot be collected
  • Locations, relative to DU, and protocol for field adjustment of locations
  • Number of individuals of each species to be collected (by sex and size, if relevant), per location, including reference location
  • Number of organisms to be composited in each sample (single species only)
  • Body part(s) to be tested (whole body, liver, eggs, blood, etc.)
  • Other parameters to be measured (lesions, parasites, fin rot, etc.)

Selection of Appropriate Species for Field Collection

Selecting the appropriate species for field collection is critical to the defensibility of the BERA. Not all species are suitable for answering all questions. The three principal reasons for collecting organisms from the site are (1) chemical analysis; (2) community metrics; and (3) evidence of disease.

All three of these lines of evidence require species with the following characteristics:

  • Exposure: The species is exposed to the site (and the reference area) for a substantial period of time relative to its lifespan, so that observed effects can be linked to the site. Year-round residency is desired but not required.
  • Ecological Relevance: Organisms should be ecologically relevant to the evaluation. For example, if risk to the wedge-tailed shearwater from fish consumption is being evaluated, individual fish of the appropriate species and size should be collected. Seasonality should also be considered (see below).
  • Abundance: Field-collected species should be abundant enough at the site and reference area to support collection of specimens for the intended use. See Table 21-13 for tissue volumes generally required for chemical analyses.

Table 21-13. Typical Tissue Volumes Required for Selected Chemical Analysis  
Chemical Group Tissue Volume Required (grams wet weight)
Low Level Detection Standard Level Detection
Metals 2 2
Pesticides 15 1.5
PCBs 15 1.5
Dioxins/Furans 10 10
SVOCs 30 2
Percent Lipids and Moisture 10 3

Species collected specifically for chemical analysis must meet the following additional criteria:

  • Ability to accumulate the chemical: Many metals are accumulated by both plants and animals, but most organic chemicals are not likely to be accumulated in plants. Metals that are essential nutrients may be actively regulated by the organism and thus not suitable for use as indicators of bioavailability. Verify that the COPECs being evaluated are known or expected to accumulate in the organism targeted for collection.
  • Limited ability to metabolize the chemical: Some organisms metabolize certain organic chemicals, which makes the compounds less likely to accumulate in tissues. For example, PAHs induce mixed function oxidase enzymes (and thus their own biotransformation) in fish and other vertebrates, but not in mollusks or crustaceans (USEPA 2000i). Although fish may show signs of PAH exposure, such as lesions or tumors, tissues may not contain elevated concentrations of PAHs relative to sediments.
  • Sex and Seasonal Variability: Chemical concentrations in a species may vary by sex, often influenced by reproductive processes. For example, a female fish or invertebrate may transfer some organic chemicals to her eggs, thus reducing her body burden. Chemical analysis of composite samples made up of several individuals may vary from one another simply because the sex ratios in the samples differed. This situation would confound the analysis of site-related bioavailability and compromise the findings of the BERA. Whenever possible, the sex and reproductive condition (pre- or post-spawning) of individuals in a composite (and across the site and reference area) should be matched. Likewise, chemical concentrations in organisms may vary by season. A study of tissue concentrations at Ordnance Reef reported that metals were higher in goatfish samples in the fall, but higher in octopus samples in the spring. The BERA WP/SAP should include a review of published findings on factors affecting seasonal variability to support the proposed sampling approach.

Collection and analysis of organisms can be time consuming and costly, as well as potentially affecting the habitats and communities at the site and reference area. The rationale for tissue collection should be clearly explained in the BERA WP/SAP so that the most appropriate organisms are collected to address the study objectives. Appendix 21-A presents profiles of 22 common Hawaiian species, including information on previous tissue analysis. The HEER Office recommends that these 22 species be used whenever possible so that a more robust statewide dataset can be developed.

Tissue Sample Handling and Processing

The BERA WP/SAP should describe methods for handling and processing field-collected organisms, including preservation (freezing or refrigerating), dissection (body parts to be analyzed), homogenization techniques, and other procedures. No single approach is appropriate for all tissue samples. If the tissue concentrations will be used in a food chain model, then the whole body should generally be analyzed. If a COPEC is known to differentially accumulate in a single organ, such as the liver, then an organ-specific analysis may be more appropriate. In some cases, only a part of an organism (blood, eggs, feathers) is collected.

The approach to preparing laboratory analysis replicates of tissue samples should be described in the QAPP. In most cases, replicate samples for tissue samples (i.e. triplicates or duplicates) will be prepared by the laboratory after the sample is homogenized and replicates need to be collected in separate random locations across the homogenized tissue, not co-located. Separate samples collected in the field, even from a single location, are considered different samples for analysis, not replicates.

In most cases, the HEER Office recommends that the laboratory report the results in dry weight and also measure and report percent moisture. Various uses of the results may require wet weight or dry weight concentrations. For example, if the results will be used as inputs into a food chain model, and the predator’s ingestion rate is on a dry-weight basis, then the results should be in dry-weight. However, if the tissue results will be compared to critical body residues that are presented in wet-weight, then the results should be presented in wet-weight. In either case, if percent moisture in the tissue samples is reported, concentrations can be converted between wet-weight and dry-weight by the risk assessor as needed. Percent lipids should also be measured in any tissue analyzed for organic chemicals.

Spatial Correlation with Sediment Samples

As mentioned above, tissue concentrations can provide a strong line of evidence for bioavailability and potential toxicity of chemicals in sediment. However, the strength of this line of evidence is dependent on the degree to which the organisms are linked to the area of known sediment concentrations. It is essential that tissue samples be co-located in space and time with sediment samples, and that both are relevant to the DUs previously established.

Collection of Reference Samples

Because most sites are affected by general human activity apart from any site-related chemical release, the use of reference locations is essential to a strong ERA. Reference samples are used as a basis of comparison so that site-related chemical concentrations can be interpreted in the context of ambient or background chemical concentrations. The designation of reference areas was discussed previously (Step 3A, Task 1 in Subsection 21.3.4 and Appendix 21-E, Step 3). The HEER Office is compiling a database of tissue concentrations collected across the state for numerous purposes. During the BERA WP/SAP review process, the HEER Office may make these data available to the risk assessor to support a more robust analysis of ambient tissue concentrations.

The HEER Office must approve the reference area prior to sample collection. A minimum of three tissue samples should be collected at the reference site, using the following guidelines:

  • Reference samples should be of the same species, size (+20%), and sex as the site samples.
  • Site samples should be collected first, followed immediately by reference samples. This ensures that the reference species can be matched to the site samples.
  • Reference areas should reflect general regional conditions (air deposition, general land use) but not be affected by site contaminants or other known sources of contamination. Physical habitat must be comparable to the site (wave energy, grain size, salinity, etc.) Toxicity Testing

Chemicals detected in sediment, surface water, or pore water are not necessarily in a form that can cause adverse effects on receptors. To directly measure the bioavailability and potential toxicity of a sample, a test organism is exposed to the sample under controlled conditions. Laboratory bioassays are used to test the reactions of living organisms to water or sediment collected from a potentially contaminated site. Because interpretation of in situ field bioassays using native organisms can be confounded by multiple factors, standardized laboratory bioassay tests with a small number of well-studied species are typically used instead. Whether the suite of bioassay organisms and the particular test protocols that have become the norm in the mainland U.S. are reasonable tools for tropical marine assessments has been the topic of discussion during the past 20 years (Peters et al. 1997; Batley and Simpson 2008; Simpson et al. 2007).

Need for specific protocols to address ecological risk in tropical marine ecosystems was identified during the early stages of the USEPA ERA framework process because differences in the geochemistry and physical nature of sediment, climatic conditions, and other features of tropical ecosystems suggest that the exposure pathways may not be adequately represented by protocols developed for temperate ecosystems. Tropical marine ecosystems are not well represented by standard USEPA bioassays or exposure models.

Since that initial review paper (Peters et al. 1997), which thoroughly described the steps necessary to develop a tropical marine program, progress has been slow. Despite substantial advances in assessing ecological risk in general, the focus is still on temperate ecosystems (Batley and Simpson 2008).

Tropical marine species can be substituted for temperate species in some cases. Examples of new bioassay protocols developed to address tropical regions of Australia include the following (based on Adams and Stauber 2008):

  • Tests using native benthic unicellular microalgae measure enzyme activity rather than growth; the test can be used in a wide range of grain sizes.
  • A native polychaete (Scoloplos sp.) was substituted in an ASTM method; the native polychaete is an infaunal tunneler that lives in sediment of a wide range of grain sizes.
  • No tropical amphipod test has been developed, but these authors suggest that amphipods exposed to typical coarse-grained sediments of coralline habitats may have to be fed during the test. (Tests with the freshwater Australian amphipod Melita plumulosa were not compromised by feeding.)
  • The tropical hermit crab (Diogenes sp.) can be used for whole sediment bioassays. (Although this genus may not occur in Hawaiʻi, other members of the Family Diogenidae may be equally useful as test organisms.)
  • A standard bivalve bioassay can be modified to use the widespread tropical Donax cuneata.

The Sediment Evaluation Framework for the Pacific Northwest (Northwest Regional Sediment Evaluation Team (RSET 2016) provides a useful overview of toxicity testing and approaches for evaluating the results. The HEER Office continues to work with researchers to identify appropriate test organisms for contaminated marine sediment sites in Hawaiʻi. The BERA WP/SAP should provide rationale for the proposed toxicity tests, including and modifications to standard protocols that would make the tests more representative of site conditions (water temperature, day length, etc.). The HEER Office will discuss other options with the risk assessor as needed.

In addition to the test species, the BERA WP/SAP Work Plan should describe the overall approach to the toxicity tests, including the duration of exposure; feeding regime; endpoints evaluated (growth, survival, reproduction, other); number of replicates; parameters measured during the test and frequency of measurements (pH, ammonia, other); and other specific test procedures. The criteria for sample selection should also be described. Issues that must be considered in the design of toxicity test samples include, but are not limited to, those below:

  • What is the purpose of the toxicity test? What is the null hypothesis?
  • Will toxicity testing run concurrent with or after chemical analysis?
  • If chemical results are known, will samples for toxicity tests be selected randomly or purposefully to represent a range of concentrations?
  • If purposefully selected, how will concentration bins be determined? What if more than one chemical is detected at the site (the most typical situation)?
  • What types of correlation or regression analyses are planned? How many samples are required for robust analysis?
  • How will variability among endpoints be interpreted? (For example, the test may show no effect on growth but a significant decrease in reproductive output, or vice versa.)
  • How will samples form the reference area be selected?
  • How will toxicity in site samples be evaluated with respect to the reference area?

The questions above, and any other relevant issues, should be thoroughly discussed in the BERA WP/SAP. Well-designed toxicity tests can provide a strong line of evidence to the BERA, but poorly designed tests waste time and money while only adding to the uncertainty in the BERA. Laboratory Bioaccumulation Testing

Several limitations with field collected organisms can be addressed by conducting laboratory bioaccumulation tests. For example, while field collected organisms can answer questions about exposure to chemicals in the wild, it is never possible to identify with certainty when or how the chemicals were taken into the organism’s tissues. In other cases, organisms may be too scarce or difficult to collect from the site. Laboratory bioaccumulation tests also have disadvantages, such as using test organisms that are not native to the site, misrepresenting conditions in overlying water at the site, and interrupting normal feeding habits of the test organisms. Even when the same species is tested in the field and in the laboratory, results may vary. For example, tests comparing bioaccumulation in an estuarine bivalve (Tellina deltoidalis) under lab and field conditions reported that important parameters differed between lab and field over the 31-day exposure period. Percent fines at the surface of the test sediment increased in the field but not in the lab. AVS increased in lab but not in the field (Belzunce-Segarra et al. 2015). This and other studies serve as a caution against extrapolating or over-interpreting both lab and field results. Despite these caveats, laboratory bioaccumulation tests can provide an independent line of evidence to the ERA. The Sediment Evaluation Framework for the Pacific Northwest (Northwest Regional Sediment Evaluation Team (RSET 2016) provides a useful overview of bioaccumulation testing.

The proposed laboratory bioaccumulation test should be thoroughly described in the BERA WP/SAP, referencing protocols when available. Include information on the exposure duration, test organisms, depuration, parameters measured during the test, frequency of measurements, endpoints, replacement of overlying water, feeding, and any other variable that could affect the usefulness of the test. The BERA WP/SAP should describe how samples will be selected for bioaccumulation testing, in keeping with the discussion above for toxicity tests.

Prior to initiating the test, at least one representative tissue sample of test organisms must be collected and either immediately analyzed or frozen for analysis with the test samples after the test is completed. This sample will serve as the baseline concentration for comparison of test samples.

Depending on the study objective, organisms may or may not be depurated to eliminate sediment from the gut prior to chemical analysis. When the goal of the test is to derive a BSAF, or to compare bioaccumulation among several samples, then the organisms are typically depurated. If the goal of the bioaccumulation test is to provide concentrations in prey organisms for use in the food chain model, then the test organisms should not be depurated. The rationale for depurating (or not) should be given in the BERA WP/SAP.

After the exposure period, test organisms are processed and analyzed for chemical constituents. The BERA WP/SAP should provide details on which samples (if not all) will be analyzed, how they will be homogenized, whether they will be frozen or otherwise preserved, and which analyses will be performed.

As mentioned above, tissue analytical results should be reported as dry weights. Percent moisture and percent lipids should be measured whenever organic compounds are analyzed. The BERA WP/SAP should specify how tissue results will be interpreted with respect to laboratory controls and reference area samples. For example, what does it mean when tissue concentrations at the site are 10 times concentrations at the reference area?

21.6.3 Data Analysis and Interpretation

The results of the chemical sampling, biological surveys, toxicity testing, bioaccumulation studies, and any other data collected are evaluated in the data analysis subsection of the BERA. The HEER Office expects that the risk assessor will follow current practice and adhere to professional standards in analyzing and interpreting the data. If the risk assessor is not familiar with the general process of preparing an ERA or would like a review, numerous publications available to the public offer guidance and assistance on specific topics. Current USEPA ERA guidance can be accessed online (USEPA 2018b). Older ERA guidance documents have been made easily accessible by Oak Ridge National Laboratory (ORNL 2018).

In general, all field-generated data and records (such as the field data sheets) should be reviewed for completeness and accuracy by the risk assessment technical lead. All field-generated data, including photographs and videos, should be maintained in the project file and included (as appropriate) in the final BERA. Notes on selected topics important to the HEER Office are presented below. The risk assessor should contact the HEER Office to request additional support if needed. Field Notes

Descriptions of the sediment, surface water, and habitat such as odors, colors, sheens, debris, presence of organisms, sediment substrate (i.e., sand, silt, gravel), signs of scouring, water depth, outfalls, and other features can be helpful when interpreting results of site-specific studies. Therefore, all observations should be documented in a field log book and photographs should be taken of the sediment and sample locations. Any field variances of the SAP should be clearly documented in the field log book. These observations should be presented in a summary table to aid the reviewer of the BERA (Table 21-14).

Table 21-14. Example of Qualitative Field Notes  
Station Redox Discontinuity (cm) Sediment Description Biota Present Other Comments/Notes Photos
SD01 < 1 black color, silt/clay with some sand worm burrows, iron secretions Collected samples in the mudflat located on the peninsula on the side facing the bridge. 2
SD02 no redox red-brownish color sand with some silt none observed   1
SD03 3 medium brown sand with medium grey silt below worm burrows Moved 14 feet toward water because riprap was present at proposed sample location. 1
SD04 < 1 black silt one mussel shell (open) Collected sample 30 feet south of 2nd wooden pier. 1
SD05 1 to 4 brown sand at surface, brown/dark grey to black silt below limu, eelgrass, some live gastropods   3
Redox Discontinuity - Depth at which the color changes from brown to gray/black Analytical Results

Data that will be used in a risk assessment should undergo a Stage 4 data validation in accordance with the USEPA National Functional Guidelines to ensure that the data are of good quality and are legally defensible. Methods for validating the data should be given briefly in the BERA WP/SAP and explained fully in the QAPP, along with the criteria for determining the acceptability of the data. Guidance on data validation is available from USEPA through the Contract Laboratory Program National Functional Guidelines for Data Review (USEPA 2018).

Data packages should also be reviewed to determine whether any data should be rejected and whether any data qualifiers assigned during the validation process affect the usability of the data as defined in the QAPP. The validated analytical data packages should contain a summary of all data qualifier flags and their explanations.

Analytical results for all media should be presented in summary statistics tables including the following information: chemical name and CAS number, number of samples analyzed, maximum and minimum detected concentrations, data qualifiers, range of detection limits, and frequency of detection. When samples sizes are large enough (n>10), estimates of the mean such as the 95 percent upper confidence limit on the mean concentration (UCL95) may be used to represent the exposure point concentration in the DU. See TGM Section 4 for more information on calculating UCL95. When appropriate, separate tables that show results only for chemicals detected in at least one sample may be presented to focus the BERA. However, whenever a result is listed as “not detected,” the sample-specific detection limit must be given in the table.

The sample-specific detection limits reported by the laboratory should be reviewed prior to using the data in the BERA. If the laboratory was not able to meet the detection limits presented in the WP/SAP, the data may not be useable for the BERA. Regardless of the format of tables chosen by the risk assessor, all data for all analyzed parameters, including parameters not detected in any sample, must be included as appendices to the BERA.

Pay close attention to concentration units (e.g., µg/kg, mg/kg) in all tables and throughout the text. Laboratory results, regulatory criteria, and published literature may use different units. It is the risk assessor’s responsibility to convert all units to a uniform standard so that meaningful comparisons can be made. Many components of the BERA incorporate ratios (such as hazard quotients and bioaccumulation factors) that are rendered meaningless when units are not consistent. Likewise, double-check that the dry-weight or wet-weight concentrations are properly represented. In peer-reviewed publications, this detail may appear only in a table or figure legend rather than stated explicitly in the text. When in doubt, contact the HEER Office for assistance. Toxicity and Bioaccumulation Tests

The BERA should refer to the description of toxicity and bioaccumulation tests proposed in the WP/SAP and explain any variances to the proposed procedures. When results of the laboratory toxicity tests are presented, the reasons for variances and potential effects of results should be explained. For example, the laboratory technician may have decided to aerate the samples because the dissolved oxygen level decreased below a certain threshold.

The full laboratory toxicity test report should be included as an appendix to the BERA and the results summarized in the BERA. The format of results may vary with the type of test; Table 21-15 is provided as an example only. Any potentially confounding factors, such as high ammonia or low dissolved oxygen, should be discussed in the text. The laboratory control sample results should be evaluated to determine whether the test met acceptability criteria.

Table 21-15. Summary of Leptocheirus plumulosus Toxicity Test Results  
Sample Number Mean Survival (%) Mean Dry Weight (mg/organism) Mean Overall Juvenile Production (juveniles/ amphipod) Mean Juvenile Production per Surviving Female (juveniles/female amphipod)
Lab Control Sample 85 1.47 6 13
Reference Samples
RF-SD01 83 1.40 7 13
RF-SD02 84 1.48 6 14
RF-SD03 80 1.52 6 12
Site Samples
SD101 63 0.99 6 12
SD102 77 1.57 5 17
SD103 81 1.27 5 9
SD104 53 1.30 7 13
SD105 71 1.55 9 17

Several methods can be used to evaluate toxicity test results. The three most common endpoints for sediment toxicity testing include (1) mortality as measured by survival of the amphipods; (2) growth as measured by weight and biomass; and (3) reproduction as measured by overall juvenile production and juvenile production per surviving female. The BERA WP/SAP should provide details on how results will be interpreted for any endpoints other than these.

Site samples are identified as toxic relative to the reference samples using a statistical test. The laboratory control samples are included simply to determine whether the test organisms were healthy; laboratory controls are not used to evaluate site-specific toxicity. Methods to interpret toxicity test results should have been specified in the BERA WP/SAP and discussed with the HEER Office. (Guidance on statistical tests appropriate for analyzing toxicity test results is under development and will be added to this TGM when available.)

Laboratory bioaccumulation studies should be treated in the same way as toxicity studies, with the added component of final tissue concentrations. As discussed previously, tissue concentrations should be provided in dry weight, along with percent moisture and percent lipid results. Tissue results from laboratory bioaccumulation tests should be presented in the same way as field-collected tissues, with the additional component of calculated BSAFs, if warranted.

Risks to Receptors from Food Chain Exposure

Tissue concentrations are used in food chain models to estimate daily doses to consumers, as described in Subsection 21.3.3. While the SLERA intentionally biased the estimated daily dose high using conservative exposure parameters, the average dose is used in the BERA to represent a more realistic exposure scenario. The focus of the BERA is risk to populations of receptors, not to individual organisms. Therefore, average exposure assumptions are used. For example, the estimated daily dose in the BERA should incorporate the components below:

  • Mean chemical concentrations in sediment and food;
  • Mean ingestion rates for sediment and food;
  • Mean body weight;
  • Appropriate site use factor; and
  • Most sensitive life stage present at the site.

In the SLERA, concentrations in food are estimated from concentrations in sediment using BSAFs as described in Appendix 21-E. However, if site-specific tissue samples were analyzed in the BERA, those concentrations should be substituted in the dose equation. Alternatively, if site-specific BSAFs are determined in the BERA, they should be used instead of BSAFs from the literature to estimate tissue concentrations at the site.

21.6.4 Risk Characterization

The risk characterization subsection of the BERA is where all available data are evaluated holistically to determine whether the site poses unacceptable risk to any of the assessment endpoints. The risk characterization should present both quantitative and qualitative characterizations of risk, to the extent supported by available data. As described in Step 3a (Task 6), the risk characterization focused on interpreting exposure and effects data within the context of other site-specific information. Risk characterization in the BERA is similar, in that it synthesizes all available data and various sources of uncertainty, while acknowledging data gaps that may limit conclusions.

When multiple measures of effect are available for a single assessment endpoint, then a weight-of-evidence approach should be used to interpret the implications of different datasets. For example, as discussed in Subsection, biological surveys are often collected as part of a sediment triad approach where three lines of evidence (sediment chemistry, sediment toxicity test data, and benthic community data) are evaluated to as part of an overall investigation of the benthic community. This can be done by assigning each line of evidence a score and associated weighting factors.

The risk assessment results can be presented graphically to highlight locations where chemical concentrations exceed toxicity screening levels that were identified in the BERA WP/SAP. Maps and graphs may be used to illustrate spatial distribution of risk using various measures. The HEER Office can offer examples of effective data presentation methods, as needed.