Calculating Background Levels for Ecological Risk Parameters in Toxic Harbor Sediment

By Leadon, Christopher J McDonnell, Thomas R; Lear, Janet; Barclift, David

Establishing background levels for biological parameters is necessary in assessing the ecological risks from harbor sediment contaminated with toxic chemicals. For chemicals in sediment, the term contaminated is defined as having concentrations above background and significant human health or ecological risk levels. For biological parameters, a site could be considered contaminated if levels of the parameter are either more or less than the background level, depending on the specific parameter. Biological parameters can include tissue chemical concentrations in ecological receptors, bioassay responses, bioaccumulation levels, and benthic community metrics. Chemical parameters can include sediment concentrations of a variety of potentially toxic chemicals. Indirectly, contaminated harbor sediment can impact shellfish, fish, birds, and marine mammals, and human populations. This paper summarizes the methods used to define background levels for chemical and biological parameters from a survey of ecological risk investigations of marine harbor sediment at California Navy bases. Background levels for regional biological indices used to quantify ecological risks for benthic communities are also described. Generally, background stations are positioned in relatively clean areas exhibiting the same physical and general chemical characteristics as nearby areas with contaminated harbor sediment. The number of background stations and the number of sample replicates per background station depend on the statistical design of the sediment ecological risk investigation, developed through the data quality objective (DQO) process. Biological data from the background stations can be compared to data from a contaminated site by using minimum or maximum background levels or comparative statistics. In Navy ecological risk assessments (ERA’s), calculated background levels and appropriate ecological risk screening criteria are used to identify sampling stations and sites with contaminated sediments. Keywords Contaminated harbor sediments, ecological risk, bioassessment, background reference

Introduction

The United States Navy has many coastal installations at which sediment investigations may be necessary. To avoid costly cleanup of large sediment volumes, the Department of the Navy (DON) relies on thorough investigations of contaminated sediment through their Environmental Restoration (ER) program, a cleanup program similar to Superfund. Information on how background levels were determined for ecological risk assessment (ERA) parameters in harbor sediment in ER reports from several California Navy bases was surveyed and summarized in this paper. The California Navy bases include installations on San Francisco Bay, Port Hueneme, Mugu Lagoon, San Pedro Bay, and San Diego Bay. As well as providing an interesting discussion of various methods, the objective of this paper is to provide a useful reference for readers seeking information on how to calculate background for ecological risk assessments of harbor sediments.

Background provides a reference against which site-specific conditions can be compared. Background levels for chemical and ecological risk parameters must be established during assessments of harbor sediments to distinguish between ecological conditions resulting from the release of a contaminant at a site and ecological conditions resulting from naturally occurring or non-release anthropogenic conditions. This paper will present a limited review of physical and chemical properties and their use in characterizing biological background.

The collection of background samples is necessary to supply data for the calculation of site-specific background levels. Background levels for ecological risk parameters are used in screening ERAs as one of the two fundamental types of screening levels that can eliminate a sediment site from further investigation as a contaminated site. The other major screening tool is ecological risk criteria, which are levels of chemical or biological parameters that have been shown to be associated with ecological risks in similar areas.

The calculation of background levels for ecological parameters is a basic part of ERAs of toxic harbor sediments. The information on calculating background in toxic harbor sediments in this paper is designed to fit within the DON’s ER program, a Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) program. But the information is also relevant to other regulatory ERA processes.

Generally for CERCLA programs, the processes of ERAs are an eight- step process arranged in three general steps: Tier 1 , Screening Risk Assessment; Tier 2, Baseline ERA; and Tier 3, Evaluation of Remedial Alternatives. Calculating background levels for biological parameters is usually part of baseline ERAs in Tier 2. Frequently, applicable background levels for chemical parameters have already been established at another nearby toxic harbor sediment site, so that background screening for chemical parameters can be done in Tier 1 of the ERA process. If established information on background levels for chemical parameters is not available from a nearby sediment site, sampling for chemical and biological parameters and calculation of background levels takes place in Tier 2. Sampling investigations for ERAs in the Navy ER Program follow the seven- step DQO process (U.S. EPA, 1994a) to determine the type, quality, and quantity of data needed to support decisions.

Background levels for ecological risk parameters are used in baseline ERAs to define an incident of an adverse biological condition at a site. Whether an area of concern has an adverse condition is important when using a preponderance-of-evidence approach with multiple lines of evidence, such as the sediment quality triad approach (Chapman, 1990). The triad approach includes assessments of chemical concentrations in sediment or biological tissue, sediment toxicity bioassays, and benthic community analyses. Background levels of chemical contaminants in the tissue of biological receptors can also be important in hazard index calculations for shellfish, fish, and marine birds and mammals. A hazard quotient (HQ) can be calculated for an ecological receptor by dividing the exposure dose from a contaminant to the receptor by a screening-level dose for specific receptors and contaminants. A hazard index (HI) results from the addition of the HQs from all contaminants of potential ecological concern (COPECs) with similar modes of toxicity to a receptor. HIs less than one indicate no ecological risk. Background levels for chemical or biological parameters are the other basic screening criteria used in ecological risk calculations, in addition to screening-level doses for specific receptors and contaminants.

The concentrations of contaminants in the tissue offish and shellfish are often connected to human-health risk assessments of harbor sediment. The food chain in harbor sediment ranges from minute benthic invertebrates in the sediment, through bottom- feeding fish and birds to higher trophic-order fish, birds, marine mammals, and human fishermen.

Identifying Background Stations for Harbor Sediment Investigations

Background stations are chosen in areas outside a contaminated site. For chemical data, the data from the background stations are then statistically compared to the site data according to the procedures in the Navy guidance report on background analysis in sediment (DON, 2003a; U.S. EPA, 200Oa). The procedures are summarized in the section of this paper on “Background Levels for Chemicals in Sediment.” Specific statistical techniques for representing background for biological parameters in comparisons of background station and site data are described in the sections on biological parameters in this paper.

To establish locations of background stations in harbor sediment, the qualities and characteristics of possible locations must be evaluated before data are collected. To interpret background biological data, knowing the physical and chemical characteristics of the background locations is essential. Project designers must determine the number of background station locations. Typically, one sample replicate is taken per background station location for background data sets. The number of background station locations depends on the statistical design, which is developed through the DQO process. This process will determine the type of qualitative or statistical testing that will be used, the error probabilities, and the number of samples to be analyzed.

Background stations and the site should exhibit similar physical characteristics, such as sediment grain size and share general chemical characteristics, such as sediment organic carbon content and sulfides. Hydrological characteristics that should be similar include depth of water, salinity (i.e., freshwater, estuarine, or marine), and flow (i.e., lentie or lotie). It is also desirable to have background stations that are similar to the site with regard to ecological characteristics, such as habitat type and aquatic communities (U.S. EPA, 200Ob). Samples from both the background stations and the contaminated sediment site should be analyzed for the same target analytes, the chemicals of potential ecological concern (COPECs). The number of background stations used in the surveyed ER Program projects in Navy harbor sediments in California, summarized here for ecological risk parameters such as bioassay samples, ranged from one to ten, with five being the most common. This represents the probable number of background stations that would be needed at a new toxic harbor sediment site when statistics, regulatory requirements, and habitat characteristics are all taken into consideration. Five replicate background samples were collected at the ER site that had only one background station. This was the only background station among those surveyed in the survey of California Navy harbor sediment sites where replicate samples were collected in the field. For the remaining stations, replicate samples were prepared in the laboratory from single bioassay samples collected in the field.

Methods for Calculating Background Levels

Background Levels for Chemicals in Sediment

Evaluation of background biological parameters usually accompanies evaluation of background chemical concentrations. While selection of the target chemical compounds, those for which analyses will be run, is typically a site-specific effort, target chemical compounds usually include metals, chlorinated pesticides, polychlorinated biphenyls, volatile organic compounds, and semi- volatile organic compounds. Other general sediment characteristics are often evaluated, such as grain size, total organic carbon, total sulfides, and acid volatile sulfides and simultaneously extracted metals.

There are three general approaches to characterize the background chemical concentrations and to compare them to site chemical concentrations: exploratory data analysis, geochemical background analysis (the geochemical method), and comparative statistical analysis (the comparative method) (DON, 2003a; U.S. EPA, 200Oa). The exploratory data analysis method examines distribution patterns for an upper limit background threshold. A background threshold value is a concentration from a background dataset representing the upper range of background concentrations for a specific chemical. For small background dataseis, the background threshold value may be a mean or upper confidence limit on the mean. For larger background dataseis, the background threshold may be a 95th or 99th percentile (DON, 1999). In the frequently used threshold comparison method, the highest concentration of a chemical from a possibly contaminated site is compared to the concentration of the background threshold value for the chemical. If the highest concentration of a chemical from a site is higher than the background threshold value concentration, the site is considered possibly contaminated by the chemical and the site is screened-in for further study. The use of threshold value levels to represent background levels in screening ERAs, as with the exploratory data analysis method, has been shown to result in a high false-positive rale (i.e., declaring a site contaminated when in fact it is not) (DON, 2003a). Instead of using the exploratory data analysis method to determine background levels by calculating upper threshold limits, use of either the comparative or geochemical method is recommended (DON, 2003a; U.S. EPA, 2000a).

The geochemical method evaluates association relationships between concentrations of COPECs and concentrations of naturally occurring background chemicals, such as aluminum and iron, or natural sediment characteristics, such as grain size or organic carbon. Also known as the “on-site” method, the geochemical method can be used when it is not possible to identify a background area, because the method does not require reference area date for comparison (DON, 2003a; FDER, 1988). The geochemical method uses statistical techniques based on geochemical principles to graphically and numerically distinguish between metal concentrations that reflect natural background conditions and concentrations that may represent a chemical release (DON, 2003a; FDER, 1988; Schropp et al., 1990; Hanson et al., 1993). A classic example of the use of the geochemical method was Windom’s and Schropp’s (FDER, 1988) development of guidelines for distinguishing natural background sediment areas from contaminated sediments in the coastal estuarine waters of Florida from correlations of aluminum concentrations with the concentrations of other metals in estuarine sediments.

The comparative method compares chemical concentration distributions of a specific chemical from the site and from background locations using statistical estimates of the data distributions-in order to determine whether the site and background data distributions are similar. If the distributions of site and background date are shown lo be different after a series of statistical tests, the site is considered contaminated by the chemical. With the comparative method, the statistical comparative tests are either parametric or nonparametric. Parametric tests assume the background and site data can be represented by specific statistical distributions. Nonparametric tests do not make this assumption (DON, 2003a; U.S. EPA, 2000a). Some of the comparative statistical tests evaluate the similarity of extreme site and background concentrations, while other tests evaluate aspects like central tendencies (medians or means of the site and background data).

The procedures for calculating background levels for ecological risk parameters in sediment are similar to the comparative method used to screen chemical concentrations in sediment. With the comparative method for a specific chemical, the whole data distribution from all the background stations for a specific chemical is statistically compared to the data distribution for that chemical from all the site sampling stations as a group. In the comparison of data for an ecological risk parameter to background, the data from each individual site station for the parameter is compared to the combined data for that parameter from all the background stations. The comparison of site chemical concentrations to background concentrations often takes place in the screening phase of an ERA, while the comparison of site ecological risk data to background station data is part of a baseline ERA.

Bioassays

Sediment toxicity tests, which include an array of assays of various organisms and response endpoints (e.g., percent survival, percent normal development), provide specific information for the selected species under laboratory conditions. Bioassay protocols are available for the major taxonomic groups of sediment organisms, such as polychaete worms, crustaceans, mollusks, and echinoderms (ASTM 1990, 1995, 200Oa, 2004; U.S. EPA, 1994b, c, d, 1995, 200Oc, 2001; U.S. EPA and U.S. ACE, 1998). Sediment evaluations from the general region of the site should be reviewed for common types of bioassays used. Using the same bioassay tests from other regional studies provides the benefit of a broad database for comparison and interpretation of results. The DON presents useful summaries of the available bioassay methods for various habitats, media, and response endpoints (DON, 2003b, 2003c). Principal response endpoints include survival, growth, and development. Media alternatives include bulk sediment, water (e.g., pore water), and sediment-water interface. Endpoints, test organisms, and media should be selected in accordance with the project conceptual site model and the DQOs.

Selection of bioassays should be based on applicability to selected assessment endpoints, suitability to site environment (e.g., temperature and salinity), and desired range of sensitivity. The number of reference station samples or bioassay analyses should be determined using the DQO process, assuming that a statistical comparison of the background data and site data is desired. Each background station represents one background sample.

Before selecting the bioassays, potential confounding factors should be considered. Several confounding factors have been reviewed by DON (2000), Ankley et al. (1991, 1994), and Lacy et al. (1998). Ammonia and sulfides are toxic to a wide range of marine organisms, and both occur as a result of natural bacterial action on decaying sedimentorganic matter in marine sediment. General sediment characteristics such as grain size and total organic carbon can interfere with bioassay test results, typically due to the sensitivities of selected test species. Salinity and dissolved oxygen may interfere with test results due to their effect on other parameters such as ammonia. Before implementing the selected bioassays, methods of addressing any potential confounding factors should be considered. Before analyzing bioassay data for background conditions, the bioassay data should be compared to the laboratory control data. Several types of laboratory control data are collected during bioassays tests to aid in the interpretation of results. Positive controls or reference toxicity tests are implemented to assess the responsiveness of the test organisms. The organisms should respond to a given toxicant according to an established dose- response curve. Negative controls are implemented to assess the health of the test organisms. The negative controls should show that in the absence of a toxicant and under normal laboratory conditions, the test organisms display a strong normal response for the measured endpoint. The timeline control is unique to bioaccumulation bioassay tests in that it records the amount of COPECs present in the tissue of the test organisms prior to exposure to site sediment.

Bioassay data exhibit a fundamental difference from chemical data because of the replicates for each sample. Typically, a single sediment sample is collected from an individual background location. Chemical analysis generates a single value per target chemical. Conversely, bioassay samples are analyzed using replicates prepared in the laboratory. The replicate samples are used to test for a statistically significant difference between the mean and distribution of replicate test values for the site and the background location. ER sites have characterized background bioassay results using the arithmetic mean and various threshold values. As previously stated, the use of some threshold levels to represent background levels has been shown to result in a high false-positive rate (declaring a site contaminated when in fact it is not) (DON, 2003a). Comparison of background data with site data can be evaluated using statistical methods. A significant difference between the background results and the site results indicates an adverse response at the site that cannot be attributed to variation in the background sediment conditions. A thorough presentation of statistical methods is available in DON guidance (DON, 2003a).

The most common comparison technique in the ER Program projects summarized here is use of a t-test with pooled background data and pooled site data. However, this technique does not provide a finding for each individual station, which is often the basis of the evaluation. Individual stations have been evaluated using four techniques.

The first technique is the t-test (p < 0.05) comparing the distribution of the laboratory replicates for a site station with the distribution of the pooled replicates from all of the background stations. In this case, the laboratory replicates from individual background stations are grouped together as if they were replicates of a single background station.

The 20 percent threshold value was obtained through negotiation with project parties; RPD values can be expected to be negotiable relative to specific project needs. Some IR Program projects have used an additional comparison to evaluate toxicity based on the protocol-specific precision performance of the bioassay test over a large number of comparisons. The objective of this approach is to avoid a conclusion of statistically significant toxicity based on laboratory negative control data when the actual difference is low.

Thursby et al. (1997) presented an approach for evaluating toxicity test results for Ampelisca abdita based on a minimum significant difference from the test’s past statistical performance. Phillips et al. (2001) used a similar approach to develop threshold values for six species and nine endpoints. Phillips et al. (2001) evaluated more than 1,000 samples from the California Bay Protection and Toxic Cleanup Program using t-tests and calculating a minimum significant difference (MSD). Phillips et al. (2001) selected the 90th percentile MSD as a toxicity threshold value stating that this was equivalent to setting the statistical power at 0.90. Thresholds are given as percent of negative control and are available for three amphipods (survival), two mollusks (development), a polychaete (survival and growth), and a sea urchin (fertilization and development). By presenting this value as a percent of negative control results (i.e., normalizing by the negative control results), organism response bias that may be present in a particular bioassay test can be minimized.

MSDs are calculated for each t-test comparison using the following equation:

MSD = t^sup critical^ + (s^sub 2^^sup 1^/n^sup 1^ + s^sub 2^^sup 2^/n^sup 2^)^sub -1/2^

where

MSD = minimum significant difference

t critical = t value from the standard statistical table, for alpha = 0.05 and the appropriate degrees of freedom

s^sub 2^^sup 1^, s^sub 2^^sup 2^ =variances for treatments (control and field sample)

n^sup 1^, n^sup 2^= numbers of replicates for treatments (control and field sample)

Bioaccumulation

Bioaccumulation studies measure the concentration of chemical residue in an organism’s tissue. Bioaccumulation is the increase of a chemical’s concentration within an organism due to exposure to the chemical in the organism’s environment. In an aquatic environment, mechanisms of bioaccumulation include direct absorption from the sediment or water medium and absorption from ingested food items. If elimination mechanisms such as excretion are more efficient than the uptake due to exposure mechanisms, bioaccumulation does not occur.

Bioaccumulation may be assessed by using models or by direct measurement. However, bioaccumulation models including simple bioaccumulation factors are not available for all contaminants and are only available for a few species (Veith et al., 1979, 1980; Veith and Kosian, 1983; Bintein et al,, 1993; Tracey and Hansen, 1996). Due to the high uncertainties associated with bioaccumulation models, the direct or empirical measurement of bioaccumulation is often selected for baseline ERA. The direct measurement approach may be accomplished by using laboratory-exposed organisms (ASTM, 200Ob; Lee et al., 1993; U.S. EPA and U.S. ACE, 1998), resident site- exposed organisms, or caged transplanted site-exposed organisms (ASTM, 2002; U.S. EPA, 200Ob; Salazar et al., 1995; Salazar and Salazar, 1997, 1998).

Project design objectives determine whether the measured tissue is from laboratory organisms, resident organisms of the site, or transplanted organisms. If laboratory-exposed organisms are used, the species should be selected to minimize the inter-species uncertainties compared to the species of interest at the site. The project design objectives also determine if the measured tissue will be from specific organs, such as liver and muscle, or from a whole- body homogenate. United States Environmental Protection Agency (U.S. EPA) provides a thorough discussion of advantages and disadvantages of the various approaches for assessing bioaccumulation (U.S. ERA, 200Ob).

Laboratory-exposed organisms are selected to model certain species or groups of species that occur at a site. The DON presents useful summaries of the laboratory organisms available for bioaccumulation studies (DON, 2003b,c). Common organisms include a clam (Macoma nasuta) and a worm (Nereis virens). The laboratory studies measure the chemical uptake by standard laboratory organisms exposed to site sediment under laboratory conditions. These results may then be used as surrogate data for site-specific organisms under site-specific conditions.

Alternatively, organisms under site-specific conditions can be measured directly by sampling the resident site-exposed biota. Fish sampling is often conducted to assess the chemical residue in the tissue of resident biota. Even though the biota are collected from the site, it is often difficult to determine what percentage of the exposure has been site-specific. Fish, for example, are generally quite mobile, and the tissue residue may be due to complex exposure conditions from site and off-site sources. The results of the field collection efforts are often inconsistent between sample locations, and target species may be available at one site but not at another (U.S. EPA and U.S. ACE, 1998; U.S. EPA, 200Ob).

Transplanted site-exposed organism studies control some of these uncertainties by confining organisms to cages to assure that exposure is site-specific and that a sufficient quantity of the target species is collected. Special consideration for the design and placement of the field exposure equipment (e.g., cages) should be taken to avoid any interference with site activities, such as vessel traffic, or site organisms that may prey on the test organisms.

A bioaccumulation background condition can be described for each of the available approaches. Selection of background locations is important so that the exposure conditions are as similar as possible to exposure conditions in resident site-exposed organisms or caged, transplanted site-exposed organisms.

Of the sediment IR sites reviewed, most used a 28-day laboratory- exposed clam to assess bioaccumulation. When pooled site data were compared to pooled background data, significantly different bioaccumulation was determined using a t-test or a Wilcoxon rank sum test. The t-test method compares the two groups of data with certain assumptions, such as the data in each group approximating a normal distribution. The Wilcoxon rank sum test method compares the two groups of data with few assumptions and uses the rank value of the data. When site sample locations (n = 1) were individually compared to pooled reference data, elevated bioaccumulation values were determined with a threshold value based on the reference data such as the maximum, 95 percent upper confidence limit (UCL), and 90th percentile. However, as noted earlier, comparison of site chemical concentrations to background threshold values (such as the maximum, 95 percent UCL, or 95th percentile of the background data) may have unacceptable error.

Benthic Community Structure

In its most fundamental presentation, benthic community analysis enumerates and identifies every individual organism within a collected sample. However, because the results of such an analysis can present more than 100 species and more than 1,000 individuals, interpretation of results can be complex. These data can be combined according to various theories to present estimates of diversity. Other indices, often based on regional data, are available that are constructed around estimates of the likelihood of representative species to occur along a gradient of sediment quality.

The benthic invertebrate community can have a very irregular distribution as certain populations respond to varying conditions in small areas. Additionally, benthic invertebrate populations respond to Stressors other than the COPECs, such as temperature, salinity, and physical disturbances. Therefore, interpreting the benthic community data can be a significant challenge. Consequently, this line of evidence, which is the most direct measure of the biological characterization of a site, is often omitted from project designs. This is why background reference stations are so important for the interpretation of benthic invertebrate community structure data from toxic harbor sediment sites. At sites where suitable background stations are available, the non-COPEC responses are matched at the background and site stations and can then be removed from consideration as COPEC effects. When included in ERAs of sediment, benthic community parameters add a direct non-laboratory estimate of benthic community conditions that can be useful for comparisons to other ecological risk parameters at sediment sampling stations. Comparison of distributions for benthic community indices using statistical tests such as the f -test is not often conducted for sediment assessment projects. However, this comparison is not uncommon for monitoring programs. For those benthic community indices with background levels by definition above levels at a contaminated site, threshold values for the indices have been estimated using the minimum background reference threshold value. However, as noted earlier, comparison of site data to background reference values, such as the minimum value of the background data, may have unacceptable error. The LPL presented in the discussion of the bioassay data may present a more appropriate estimate of the background threshold. For those benthic community indices with background levels by definition less than levels at a contaminated site, threshold values for the indices have been estimated using the maximum background threshold value. An upper predictive limit (UPL) may present a more appropriate estimate of the background reference threshold than a maximum.

Confidence in interpreting benthic community data can be gained by using multiple assessment tools based on substantially different approaches. Since most diversity indices are based on the distribution of the abundance across the species richness, they share the same strengths and weaknesses. Two independent indices have been developed in southern California: the relative benthic index based on regional data from San Diego Bay (Anderson et al., 1997; Fairey et al. 1996) and the benthic response index based on regional data from southern California (Smith et al., 1998). Both of these indices present threshold values that identify a degraded benthic community.

The relative benthic index, developed as part of the California Bay Protection and Toxic Cleanup Program, was initially presented by Fairey et al. (1996) and subsequently revised by Anderson et al. ( 1997) and Fairey et al. (1 998). The index averages three subindices that are scaled relative to the range of values addressed.

Only two of the sediment investigations at California Navy bases evaluated the in situ benthic community. Both investigations evaluated the benthic community data from the contaminated site using cluster analysis with the Bray-Curtis similarity index and threshold values (minimum or maximum values) for several community indices. Background levels for benthic community indices were determined by using the appropriate minimum or maximum values for the indices from background stations. One of the studies included calculation of the BRI. Background levels for a regional parameter such as the BRI are based on accumulated data from harbor sediments in southern California that had been used initially to construct the index (Smith et al., 2003).

A weight of evidence approach using multiple lines of evidence may be used to determine the status of the benthic community. The multiple lines of evidence should be selected in consultation with the other stakeholders involved in the site management. An example selection of multiple lines of evidence is a diversity index such as species richness, cluster analysis, and indicator species represented by individual species or by an index. When all lines of evidence indicate the same condition, then the status of the site can be easily determined. When the lines of evidence are not in agreement, then concurrence of the other site stakeholders is necessary to determine the site status.

Hazard Quotient Evaluation for Wildlife Risk Calculations

Potential ecological risk to aquatic wildlife such as birds and mammals is often assessed using the chemical-specific and receptor- specific hazard quotient (HQ) approach: calculating the ratio of the exposure dose to the screening-level dose. The HQ calculations are described in detail in U.S. ERA (1997) and DON (2001) guidance. The exposure dose for aquatic wildlife is based on ingestion of food items that may have bioaccumulated chemicals from sediment at the site. The chemical concentrations in the food items are estimated by literature-based bioaccumulation factors, site-specific bioaccumulation factors, or chemical analysis of site-specific food items. Screening-level doses are frequently identified as toxicity reference values (TRVs).

HQ values not exceeding 1 are generally interpreted as indicative of acceptable ecological risk. HQs exceeding 1 in a screening-level ERA are generally interpreted as sufficient cause for further investigation. HQs exceeding 1 in a baseline risk assessment are generally interpreted as indicative of adverse ecological effects, relative to the specific assumptions made during model development.

Bioaccumulation of potentially toxic chemicals is not, alone, an adverse effect. Even HQs > 1 do not automatically signal that an adverse effect can or has occurred. Only empirical measurements of actual effects can demonstrate that the wildlife have been adversely impacted.

The evaluation of background using HQs is based on the results of the bioaccumulation studies described above. The bioaccumulation studies provide data that may be used to estimate the chemical concentrations present in prey species that larger birds and mammals will consume. Design of the bioaccumulation studies should take into consideration the models that will be used for the wildlife assessment. Relevant design considerations include how well the species used for bioaccumulation measurements represents the wildlife prey species and how well the tissue analyzed represents the portion of the wildlife prey species that is consumed.

Background HQ values, based on background bioaccumulation data, can be compared to HQ values calculated from site data because most of the exposure and toxicity factors, such as ingestion rate and TRVs, would be the same for both HQs. A few exposure factors used to calculate the background HQ and the site HQ (such as site use factor) may be different due to differences in conditions at the background location and at the site. The site assessment often eliminates or screens out inorganic chemical compounds such as metals based on a statistical comparison of the background and site chemical concentrations. Inorganic compounds with concentrations that are not significantly greater than the background concentrations may be eliminated from further evaluation. On the other hand, all organic compounds reported by the bioaccumulation studies should be evaluated with the HQ approach. Characterization of the site-related risk should include a comparison of the site HQ values to the background HQ values. In a screening-level ERA, it is common to have many chemical compounds with background HQs that are greater than 1 due to the very conservative nature of the TRVs.

In some circumstances, HQs for individual chemical compounds are summed for a hazard index value. The intent of summing the individual HQ values is to estimate the cumulative toxic effect of multiple Stressor chemicals. The value of the hazard index as an indicator of potential cumulative toxic effects depends on identification of modes of toxicity for each Stressor chemical. Only those chemicals with similar modes of action will have the potential for additive toxic effects. See U.S. EPA (1997) and DON (2001) guidance for further discussion of the use of the hazard index value.

Discussion

The methods discussed in this paper for determining background for ecological risk parameters in sediment represent practical methods used at California Navy bases in studies of harbor sediment potentially contaminated with toxic chemicals. Although the selection of background sites is irrespective of high laboratory costs, the high cost of the laboratory methods for ecological risk parameters limits the number of background samples collected for the purpose of decreasing uncertainty. The comparative method is used because it allows comparison of sample data from the possibly contaminated part of a harbor with sample data from background stations located in areas thought to represent typical, non- contaminated, ambient conditions.

The statistical parameters discussed in this paper to represent background ecological risk for sediment include threshold values (e.g., lower predictive limits or lower tolerance limits). These threshold values were not considered as susceptible to false positives as other threshold values such as maximums and percentiles. It has been shown that representing background levels with threshold values such as maximums and percentiles leads to a greater likelihood of false-positives (i.e., declaring a site contaminated when in fact it is not) (DON, 2003a).

To avoid possible false positives, a better approach might be to combine the background data set from the background stations with the site data set for exploratory data analysis, as described in Section 2.2 of the DON guidance (DON, 2003a). In general, the exploratory data analysis in Section 2.2 includes the following steps: 1) determine the probability distribution of the investigated data; 2) compute descriptive summary statistics of measured values; 3) compute representative exposure concentrations for risk screening; 4) identify potential outliers; and 5) determine background ranges (DON, 2003a). The univariate, post plot, and probability plot analyses presented in Section 2.2.4 of the DON guidance could then be used to estimate more accurately the lower, or an upper, limit of the background range for a particular ecological risk parameter. This estimate of the lower, or upper, background limit for a particular ecological risk parameter would be compared with single data points from the site area during a baseline ERA. In ERAs of sediment, site data are first screened with a variety of ecological risk screening criteria, such as threshold effect levels (TELs), effects range-lows (ERLs), probable effect levels (PELs), or effects range-medians (ERMs) (MacDonald, 1 994; Long et al., 1995). TELs and ERLs are statistically derived sediment quality guidelines (SQGs) representing chemical contaminant concentrations in marine sediments below which adverse biological effects in benthic invertebrate ecosystems are infrequently observed (MacDonald, 1994; Long et al., 1995). PELs and ERMs are statistically derived SQGs representing chemical contaminant concentrations in marine sediments.

If the maximum concentration of the chemical data in the site sediment is greater than the maximum or average concentration of the ecological risk screening criteria, the site fails the screening- level ERA and a baseline ERA is performed. In the baseline ERA, a one-to-one comparison is made between a lower or an upper limit representing the background level for a particular ecological risk parameter and a single value of the ecological risk parameter at a site sampling location or from an area of potential concern (AOPC) at the site. Those sampling stations or AOPCs with values of ecological risk parameters indicating more ecological risk than background levels are considered for possible remediation. Sampling stations or AOPCs with values indicating less ecological risk than background limits are recommended for no further action.

Conclusions

Methods used at several California Navy bases to determine background levels for ERA parameters in marine harbor sediment were surveyed and summarized in this paper. The methods included in this paper are not meant to represent a battery of laboratory tests or “in tandem” lab methods (i.e. paired lab samples).

The general design of the biological measurements in a sediment investigation should closely parallel the physical and chemical sediment measurements. The similarity of sampling design and sample location will allow for the investigation of association relationships among the data. Selection of background stations should follow the recommendations of the DON (2003a). When planning the background locations, particular attention should be given to sediment grain size and organic carbon content. The number of background stations should be developed according to the project DQO process. Based on a survey of sediment investigations for California Navy bases, five background station locations may be considered a minimum number of background stations when statistics, regulatory requirements, habitat characteristics, and costs are all considered.

Analysis of background sediment samples for physical characteristics and chemical constituents should be guided by the project DQO process. Evaluation of the background sediment physical and chemical data should follow guidance of the DON (2003a) and U.S. EPA (2000a).

Bioassays should be selected through the project DQO process and include those test species that are applicable to the site conceptual model. Background conditions can be described by the data distribution or by a calculated threshold value. The background data distribution can be statistically compared to the data distribution for site locations using a test such as a t-test. Alternatively, background threshold values may be calculated as the LPL of the background data. Statistical significance identified by either the comparisons of data distributions or a threshold value should be confirmed for relevance by reviewing overall protocol performance with a test such as the MSD threshold.

Bioaccumulation data should be collected according to project DQOs to estimate the potential bioaccumulation for exposure pathways identified in the conceptual site model. The bioaccumulation data can be evaluated with exploratory data analysis and the comparative method.

Benthic community assessment is more complex than the other sediment investigation measurements because none of the available benthic community measurements includes the state of the benthic community. Two fundamental types of data are generated by benthic community measurements: a list of species (or best possible taxonomic level) and an enumeration or count of the individual organisms belonging to the listed species. These data can be summarized into various indices considered reflective of the condition of the benthic community. The results of these indices are typically similar in the resulting interpretation of conditions in benthic communities at a harbor sediment station because they are based on the same two-variable mathematical structure of the original data set.

Therefore, calculating additional indices does not necessarily increase confidence in the data interpretation. The benthic community indices range from simple to complex integrations of the two fundamental benthic community measurements and have from traditional to contemporary origins. However, none of the indices is considered a definitive measurement of the benthic community.

Additionally, a number of factors can affect the state of the benthic community at a given point in time. Therefore, setting narrow, stand-alone measurement endpoints, such as the number of crustacean individuals or the number of molluscan species, is impractical. However, setting several measurement endpoints for benthic community performance can overcome the uncertainty attributed to a single endpoint. Selecting measurement endpoints that are based on different data analysis strategies can overcome the uncertainty that may occur with a single strategy. A concurrence of findings from a multivariate technique such as cluster analysis, a regional benthic community index such as the BRI in southern California, and conventional community indices such as abundance and diversity, can identify background conditions.

Potential risk to wildlife such as mammals and birds is often evaluated according to the HQ approach. HQ values are based on estimated exposures developed from measurement of various site media such as soil, water, or biota (tissue). An evaluation of the background conditions and comparison to site conditions should be completed before estimating the exposure dose and calculating HQ values. Inorganic compounds can be eliminated if site data are statistically less than background reference data. HQs should be prepared for each remaining COPEC and compared qualitatively to HQ values from background reference data.

References

American Society for Testing and Materials. 1990. Standard Guide for Conducting 10-day Static Sediment Toxicity Tests with Marine and Estuarine Amphipods. ASTM designation E- 1367 -90. American Society for Testing and Materials, Philadelphia, PA.

American Society for Testing and Materials. 1995. Standard Guide for Conducting Static Acute Toxicity Tests with Echinoid Embryos. E- 1563-95. American Society for Testing and Materials, Philadelphia, PA.

American Society for Testing and Materials. 2000a. Standard Guide for Conducting Sediment Toxicity Tests with Polychaetous Annelids. E 1611-00. American Society for Testing and Materials, Philadelphia, PA.

American Society for Testing and Materials. 200Ob. Standard Guide for Determination of the Bioaccumulation of Sediment-associated Contaminants by Benthic Invertebrates. E 1688-0Oa. American Society for Testing and Materials, Philadelphia, PA.

American Society for Testing and Materials. 2002. Standard Guide for Conducting In-situ Field Bioassays with Caged Bivalves. E 2122- 02. American Society for Testing and Materials, Philadelphia, PA.

American Society for Testing and Materials. 2004. Standard Guide for Conducting Static Toxicity Tests with Microalgae. E 1218-04. American Society for Testing and Materials, Philadelphia, PA.

Anderson, B., Hunt, J., Tudor, S., Newman, J., Tjeerdema, R., Fairey, R., Oakden, J., Bretz, C., Wilson, C., La Caro, F., Kapahi, G., Stephenson, M., Puckett, M., Anderson, J., Long, E., Fleming, T., and Summers, K. 1997. Chemistry, Toxicity and Benthic Community Conditions in Sediments of Selected Southern California Bays and Estuaries. Final Report. California State Water Resources Control Board. Sacramento, CA.

Ankley, G.T., Phipps, G.L., Leonard, E.N., Benoit, D.A., Mattson, V.R., Kosian, P.A., Cotter, A.M., Dierkes, J.R., Hansen, DJ. , and Mahony J.D. 1991. Acid-volatile sulfide as a factor mediating cadmium and nickel bioavailability in contaminated sediment. Environ. Toxicol. Chem. 10, 1299-1307.

Ankley, G.T., Benoit, D.A., Balough, J.C., Reynoldson, T.B., Day, K.E., and Moke, R.A. 1994. Evaluation of potential confounding factors in sediment toxicity tests with three freshwater benthic invertebrates. Environ. Toxicol. Chem. 13, 627-635.

Bintein, S., Devillers, J., and Karcher, W. 1993. Nonlinear dependence of fish bioconcentration on n-Octanol/water partition coefficients. SAR and QSAR in Em. Res. 1, 29-39.

Chapman, P.M. 1990. The sediment quality triad approach to determining pollution-induced degradation. Sci. Total Environ. 97(8), 815-825.

DON (Department of the Navy). 1999. Handbook for Statistical Analysis of Environmental Background Data. Southwest Division, Naval Facilities Engineering Command (NAVFAC) and Engineering Field Activity West (EFA West), San Diego, CA, July.

DON (Department of the Navy). 2000. Confounding Factors in Sediment Toxicology, Navy Guidance for Conducting Ecological Risk Assessments, Issue Papers. Naval Facilities Engineering Command (NAVFAC), (October 24, 2000); http://web.ead.anl.gov/ecorisk/issue/ . DON (Department of the Navy). 2001. Navy Guidance for Conducting Ecological Risk Assessments. Revised January 30, 2001; http:// web.ead.anl.gov/ecorisk/process/pdf/index.cfm.

DON (Department of the Navy). 2003a. Guidance for Environmental Background Analysis. Volume II: Sediment. Department of the Navy, Naval Facilities Engineering Command, Washington, DC. UG-2054-ENV. April.

DON (Department of the Navy). 2003b. Implementation Guide for Assessing and Managing Contaminated Sediment at Navy Facilities. Naval Facilities Engineering Command (NAVFAC). Prepared by SPAWAR Systems Center (SSC) San Diego, CA.

DON (Department of the Navy). 2003c. Toxicity Tests for Various Habitats and Media, Navy Guidance for Conducting Ecological Risk Assessments, Methods and Tools. Naval Facilities Engineering Command (NAVFAC); http://web.ead.anl.gov/ecorisk/index.cfm.

Fairey, R., Bretz, C., Lamerdin, S., Hunt, J., Anderson, B., Tudor, S., Wilson, C., La Caro, F., Stephenson, M., Puckert, M., and Long, E. 1996. Chemistry, Toxicity, and Benthic Community Conditions in Sediments of the San Diego Bay Region, Final Report. California State Water Resources Control Board, Sacramento, CA.

Fairey, R., Downing, J., Roberts, C., Landrau, E., Hunt, J., Anderson, B., Wilson, C., Kapahi, G., La Caro, F., Michael, P., Stephenson, M., and Puckert, M. 1998. Chemistry, Toxicity, and Benthic Community Conditions in Sediments of the San Diego Bay Region, Final Addendum Report. California State Water Resources Control Board, Sacramento, CA.

Florida Department of Environmental Regulation (FDER). 1-988. A Guide to the Interpretation of Metal Concentrations in Estuarine Sediments. Eds. Schropp, S. J., and Windom, H. L. Tallahassee, FL. April.

Hanson, P., Evans, D., Colby, D., and Zdanowics, V. 1993. Assessment of elemental contamination in estuarine and coastal environments based on geochemical and statistical modeling of sediments. Marine Environmental Research. 36, 237-266.

Hunt, J.W., Anderson, B.S., Phillips, B.M., Newman, J., Tjeerdema, R.S., Fairey, R., Puckett, H.M., Stephenson, M., Smith, R. W, Wilson, C.J., andTaberski, K. 2001 . Evaluation and use of sediment toxicity reference sites for statistical comparisons in regional assessments. Environ. Toxicol. Chem. 20(10), 1266-1275.

Hunt, J.W., Anderson, B.S., Phillips, B.M., Newman, J., Tjeerdema, R., Stephenson, M., Puckett, M., Fairey, R., Smith, R.W., and Taberski, K. 1998. Evaluation and Use of Sediment Reference Sites and Toxicity Tests in San Francisco Bay, Final Report. California State Water Resources Control Board, Sacramento, California, San Francisco Bay Regional Water Quality Control Board, California Department of Fish and Game, Institute of Marine Sciences, University of California Santa Cruz, Santa Cruz, CA.

Lacy, R., Watzin, M.C., Mclntosh, A.W. 1998. Sediment organic matter content as a confounding factor in toxicity tests with Chironomus tentons. Environ. Toxicol. Chem. 18, 231-236.

Lee II, H., Boese, B.L., Pelletier, J., Winsor, M., Specht, D.T., and Randall, R.C. 1993. Guidance manual: Bedded sediment bioaccumulation tests. U.S. EPA Report No. 600/R-93/183, U.S. EPA, ERL-Narragansett, RI, p. 231.

Long, E.R., MacDonald, D.D., Smith, S.L., and Calder, F.D. 1995. Incidence of adverse biological effects within ranges of chemical concentrations in marine and estuarine sediments. Environmental Management. 19(1), 81-97.

MacDonald, D.D. 1994. Approach to the Assessment of Sediment Quality in Florida Coastal Waters, Volume 1 – Development and Evaluation of Sediment Quality Assessment Guidelines. Florida Department of Environmental Protections, Tallahassee, FL.

Margalef, R. 1951. Diversidad de especies en las comunidades naturales. Pubi. Inst. Biol. Api. Barcelona. 9, 5-27.

Phillips, B.M., Hunt, J.W, Anderson, B.S., Puckett, H.M., Fairey, R., Wilson, C.J., and Tjeerdema, R. 2001 . Statistical significance of sediment toxicity test results: threshold values derived by the detectable significance approach. Environ. Toxicol. Chem. 20(2), 371- 373.

Pielou, E. C. 1977. Mathematical Ecology, John Wiley and Sons, Inc., New York.

Salazar, M.H., Duncan, P.B., Salazar, S.M., and Rose, K.A. 1995. In-situ bioassays using transplanted mussels: II. Assessing contaminated sediment at a Superfund site in Puget Sound. Environmental Toxicology and Risk Assessment, Third Volume, ASTM STP 1218. Eds. Hughes, J. S., Biddinger, G. R., and Mones, E. American Society for Testing and Materials, Philadelphia, PA.

Salazar, M.H., and Salazar, S.M. 1997. Using caged bivalves to characterize exposure and effects associated with pulp and paper mill effluents. Water Sd. Technol. 35, 213-220.

Salazar, M.H., and Salazar, S.M. 1998. Using caged bivalves as part of an exposure-dose-response triad to support an integrated risk assessment strategy. In: Proceedings, Ecological Risk Assessment: A Meeting of Policy and Science, SETAC Special Publication. Eds. de Peyster, ?., and Day. K., SETAC 3 Press, pp. 167-192.

Schropp, S., Lewis, G., Windom, H., Ryan, J., Caldner, F., and Burney, L. 1990. Interpretation of metal concentrations in estuarine sediments of Florida using aluminum as a reference element. Estuaries 13, 227-235.

Shannon, C.H., and Weaver, W. 1962. The Mathematical Theory of Communication. University of Illinois Press, Urbana, IL.

Smith, R.W., Bergen, M., Weisberg, S.B., Cadien, D.B., Dalkey, A., Montagne, D.E., Stull, J.K., and Velarde, R.G. 1998. Benthic response index for assessing infaunal communities on the mainland shelf of Southern California. In: Southern California Bight 1994 Pilot Project: IV Benthic lnfauna. Eds. Bergen, M., Weisberg, S. B., Cadien, D., Dalkey, A., Montagne, D., Smith, R. W., Stull, J. K., and Velarde, R. G., Southern California Coastal Water Research Project, Westminster, CA, pp. E1-E86.

Smith, R.W., Bergen, M., Weisberg, S.B., Cadien, D.B., Dalkey, A., Montagne, D.E., Stull, J.K., and Velarde, R.G. 2001 . Benthic response index for assessing infaunal communities on the Southern California mainland shelf. Ecol. Appi. 11, 1073-1087.

Smith, R.W., Ranasinghe, J.A., Weisberg, S. B., Montagne, D.E., Cadien, D., Mikel, T., Velarde, R., and Dalkey, A. 2003. Extending the Southern California benthic response index to assess benthic conditions in bays. In: Southern California Bight 1998 Regional Monitoring Program: VlI Benthic Macrofauna. Eds. Ranasinghe, J. A., Montagne, D. E., Smith, R. W, Mikel, T. K., Weisberg, S. B., Cadien, D., Velarde, R., and Dalkey, A. Southern California Coastal Water Research Project, Westminster, CA, pp. C1-C42.

Swartz, R.C., Schulte, D.W., Ditsworth, G.R., DeBen, W.A., and Cole, F.A. 1985. Sediment toxicity, contamination, and macrobenthic communities near a large sewage outfall. In: Validation and Predictability of Laboratory Methods for Assessing the Fate and Effects of Contaminants in Aquatic Ecosystems. Ed. Boyle, T. T. American Society for Testing and Materials, Philadelphia, PA.

Tracy, O.A. and Hansen, D. J. 1996. Use of biota sediment accumulation factors to access similarity of non-ionic exposure to benthically-coupled organisms of differing trophic mode. Arch. Envir. Contamin. Toxicol. 30, 467-475.

Thursby, G.B., Heltshe, J., and Scott, K.J. 1997. Revised approach to toxicity test acceptability criteria using a statistical performance assessment. Environ. Toxicol. Chem. 16, 1322-1329.

U.S. EPA (U. S. Environmental Protection Agency). 1994a. Guidance for the Data Quality Objectives Process. Office of Research and Development. EPA/600/R-96/055. September.

U.S. EPA (U. S. Environmental Protection Agency). 1994b. Methods for Assessing the Toxicity of Sediment-Associated Contaminants with Estuarine and Marine Amphipods. Office of Research and Development, U.S. Environmental Protection Agency. EPA/600/R-94/025. June.

U.S. ERA (U. S. Environmental Protection Agency). 1994c. Standard Test Methods for Measuring the Toxicity of Sediment-Associated Contaminants with Fresh Water Invertebrates. U.S. EPA Office of Science and Technology, Washington, DC. EPA/600/R-94/024.

U.S. ERA (U. S. Environmental Protection Agency). 1994d. Short- Term Methods for Estimating the Chronic Toxicity of Effluents and Receiving Waters to Marine and Estuarine Organisms, Second Edition. EPA/600/4-91/003. July.

U.S. ERA (U. S. Environmental Protection Agency). 1995. Short- Term Method for Estimating the Chronic Toxicity of Effluents and Receiving Waters to West Coast Marine and Estuarine Organisms. EPA/ 600/R-95/136. August.

U.S. ERA (U. S. Environmental Protection Agency). 1997. Ecological Risk Assessment Guidance for Superfund: Process for Designing and Conducting Ecological Risk Assessments, Interim Final. EPA/540/R-97/006. OSWER 9285.7-25. June.

U.S. EPA (U. S. Environmental Protection Agency). 200Oa. Guidance for Data Quality Assessment. EPA QA/G-9 Practical Methods for Data Analysis. QAOO Update. U.S. Environmental Protection Agency. Office of Environmental Information, Washington, DC. EPA 600/R-96/084. July.

U.S. ERA (U. S. Environmental Protection Agency). 200Ob. Bioaccumulation Testing and Interpretation for the Purpose of Sediment Quality Assessment, Status and Needs. Office of Water, Office of Solid Waste, Washington, DC. EPA-823 -R-00-001. February.

U.S. ERA (U. S. Environmental Protection Agency). 200Oc. Methods for Measuring the Toxicity and Bioaccumulation of Sediment- associated Contaminants with Freshwater Invertebrates. EPA600-R-99- 064. U.S. Environmental Protection Agency, Office of Water, Office of Science and Technology, Washington, DC. March.

U.S. EPA (U.S. Environmental Protection Agency). 2001 . Methods for Assessing the Chronic Toxicity of Marine and Estuarine Sediment- Associated Contaminants with the Amphipod Leptocheirus plumulosus. Office of Research and Development and Office of Water, U.S. Environmental Protection Agency, Washington, DC. EPA/600/R-0 1/020. March. U.S. EPA (U. S. Environmental Protection Agency) and U.S. ACE (U. S. Army Corps of Engineers). 1998. Evaluation of Dredged Material Proposed for Discharge in Water of the U.S. – Testing Manual (Inland Testing Manual). EPA-823-B-98-004. Washington, DC.

Veith, G.D., Defoe, D.L., and Bergstedt, B.V. 1979. Measuring and estimating the bioconcentration factor of chemicals on fish. J. Fish Res. Board Can. 36,1040-1048.

Veith, G.D., Macek, K.J., Petrocelli, S.R., and Carroll J., 1980. An evaluation of using partition coefficients and water solubility to estimate bioconcentration factors for organic chemicals in fish. In: Aquatic Toxicology. Eds. Eaton, J. G., Parrish, P. R., Hendricks, A. C. ASTN STP 707. American Society for Testing and Materials, Philadelphia, PA.

Veith, G.D., and Kosian, P. 1983. Estimating bioconcentration potential from octanol/water partition coefficients. In: Physical Behavior of PCBs in the Great Lakes. Eds. Mackay, D., Paterson, S., Eisenreich, S. J., Simons, M. S. Ann Arbor Sciences Publishers, Ann Arbor, MI, pp. 269-282.

CHRISTOPHER J. LEADON,1 THOMAS R. McDONNELL,2 JANET LEAR,3 AND DAVID BARCLIFT4

1 Naval Facilities Engineering Command Southwest (NAVFACSW), San Diego,

CA, USA

2 Brown and Caldwell, Irvine, CA, USA

3 Brown and Caldwell, San Diego, CA, USA

4 Formerly NAVFAC, Engineering Field Activity (EFA) Northeast, Lester, PA,

USA, and currently with U.S. Geological Survey, Patuxent Wildlife Research

Center, Duty Station: ERA Region III, Philadelphia, PA, USA

The authors would like to thank the anonymous reviewers with Soil and Sediment Contamination : An International journal who reviewed this paper. The views expressed herein are the authors’ own and not the official views of the U.S. Navy, the Naval Facilities Engineering Command, or the U.S. Geological Survey.

Address correspondence to Christopher J. Leadon, Naval Facilities Engineering Command Southwest (NAVFACSW), San Diego, CA, USA. E- mail: christopher.leadon@navy.mil

Copyright Taylor & Francis Ltd. 2007

(c) 2007 Soil & Sediment Contamination. Provided by ProQuest Information and Learning. All rights Reserved.