Analytical Requirements for Dredged Material Assessment

　

1. This Technical Annex covers the analytical requirements necessary to implement paragraphs 5.4 - 5.9 of the OSPAR Guidelines for the Management of Dredged Material.

2. A tiered approach to testing is recommended. At each tier it will be necessary to determine whether sufficient information exists to allow a management decision to be taken or whether further testing is required.

3. As a preliminary to the tiered testing scheme, information required under section 5.3 of the Guidelines will be available. In the absence of appreciable pollution sources and if the visual determination of sediment characteristics lead to the conclusion that the dredged material meets one of the exemption criteria under paragraph 5.2 of the Guidelines, then the material will not require further testing. However, if all or part of the dredged material is being considered for beneficial uses, then it will usually be necessary, in order to evaluate these uses, to determine at least some of the physical properties of the material indicated in Tier I.

4. The sequence of tiers is as follows:

A pool of supplementary information, determined by local circumstances may be used to augment each tier (cf. section 5.5 of the Guidelines).

5. At each stage of the assessment procedure account must be taken of the method of analysis. Analysis should be carried out on the whole sediment (<2 mm) or in a fine-grained fraction. If analysis is carried out in a finegrained fraction, the results should be appropriately converted to whole sediment (<2 mm) concentrations for establishing total loads of the dredged material. Additional information (e.g. as regards storage and pre-treatment of samples, analytical procedures, analytical quality assurance) can be obtained in the JAMP Guidelines for Monitoring Contaminants in Sediments.

6. The physical composition of samples, and therefore the chemical and biological properties, can be strongly influenced by the choice of sampling sites, the method of sampling and sampling handling. These possible influences should be taken into account when evaluating data.

　

Tier I:　PHYSICAL PROPERTIES

　

Physical analyses are important because they help to indicate how the sediment may behave during dredging and disposal operations and indicate the need for subsequent chemical and/or biological testing. In addition to the visual determination of sediment characteristics required in section 5.3 of the Guidelines, it is strongly recommended that the following determinations be carried out:

When dredged material is being considered for beneficial uses, it will also usually be necessary to have available details of the engineering properties of the material eg. permeability, settling characteristics, plasticity and mineralogy.

　

Tier II:　CHEMICAL PROPERTIES

　

The following trace metals should be determined in all cases:

　Cadmium (Cd)　Copper (Cu)　Mercury (Hg)　Zinc (Zn)

　Chromium (Cr)　Lead (Pb)　Nickel (Ni)

The following organic/organo-metallic compounds should be determined:

However, the determination of PCBs, PAHs and Tri-Butyl Tin compounds and its degradation products will not be necessary when:

When PCB analyses are undertaken, information on each of the congeners on the ICES primary list should be reported to the Commission.

Based upon local information of sources of contamination (point sources or diffuse sources) or historic inputs, other determinants may require analysis, for instance:

　

arsenic	other	organophosphorus	petroleum
	chlorobiphenyls2	pesticides	hydrocarbons
	organochlorine	other organotin	Polychlorinated
	pesticides	compounds	dibenzodioxins
			(PCDDs)/polychlorinated
			dibenzofurans
			(PCDFs)
	other anti-fouling
	agents

　

In deciding which individual organic contaminants to determine, reference should be made to existing priority substance lists, such as those prepared by OSPAR and the EU³.

Normalisation

It is recommended that normalised values of contaminants should be used to enable a more reliable comparison of contaminant concentrations in dredged material with those in sediments at disposal or reference sites, as well as with action levels. The normalisation procedure (see Technical Annex II) used within a regulatory authority should be consistent to ensure effective comparisons.

　

Analytical Techniques

Reference should be made to the Technical Annexes of the JAMP monitoring guidelines (cf. reference OSPAR, 1997) for recommended analytical techniques.

　

Tier III: BIOLOGICAL PROPERTIES AND EFFECTS

　

In a significant number of cases the physical and chemical properties described above do not provide a direct measure of the biological impact. Moreover, they do not adequately identify all physical disturbances and all sediment-associated constituents present in the dredged material. If the potential impacts of the dredged material to be dumped cannot be adequately assessed on the basis of the chemical and physical characterisation, biological measurements should be carried out.

The selection of an appropriate suite of biological test methods will depend on the particular questions addressed, the level of contamination at the dredging site and the degree to which the available methods have been standardised and validated.

To enable the assessment of the test results, an assessment strategy should be developed with regard to granting a permit authorising disposal at sea. The extrapolation of test results on individual species to a higher level of biological organisation (population, community) is still very difficult and requires good knowledge of assemblages that typically occur at the sites of interest.

　

　

The primary purpose of toxicity bioassays is to provide direct measures of the effects of all sediment constituents acting together, taking into account their bioavailability. For ranking and classifying the acute toxicity of harbour sediment prior to maintenance dredging, short-term bioassays may often suffice as screening tools.

The outcome of sediment bioassays can be unduly influenced by factors other than sediment-associated chemicals. Confounding factors like ammonia, hydrogen sulphide, grain size, oxygen concentration and pH should therefore be determined during the bioassay.

Guidance on the selection of appropriate test organisms, use and interpretation of sediment bioassays is given by eg. EPA/CE (1991/1994) and IADC/CEDA (1997) while guidance on sampling of sediments for toxicological testing is given by e.g. ASTM (1994).

　

　

Biomarkers may provide early warning of more subtle (biochemical) effects at low and sustained levels of contamination. Most biomarkers are still under development but some are already applicable for routine application on dredged material (e.g. one which measures the presence of dioxin-like compounds - Murk et al., 1997) or organisms collected in the field (e.g. DNA strand/breaks in flat fish).

　

　

There are short-term microcosm tests available to measure the toxicant tolerance of the community e.g. Pollution Induced Community Tolerance (PICT) (Gustavson and Wangberg, 1995)

　

　

ln order to investigate long-term effects, experiments with dredged material in mesocosms can be performed, for instance to study the effects of PAHs in flatfish pathology. Because of the costs and time involved these experiments are not applicable in the process of authorising permits but are useful in cases where the extrapolation of laboratory testing to field condition is complicated r environmental conditions are very variable and hinder the identification of toxic effects as such. The results of these experiments would be then available for future permitting decisions.

　

　

Monitoring in the surrounding of the disposal site of benthic communities e.g. in situ (fish, benthic invertebrates) can give important clues to the condition of marine sediments and are relevant as a feed-back or refinement process for authorising permits. Field observations give insight into the combined impact of physical disturbance and chemical contamination. Guidelines on the monitoring of benthic communities are provided by e.g. OSPAR, ICES, HELCOM.

　

　

Where appropriate, other biological measurements can be applied in order to determine e.g. the potential for bioaccumulation and for tainting.

　

SUPPLEMENTARY INFORMATION

　

The need for further information will be determined by local circumstance and may form an essential part of the management decision. Appropriate data might include: redox potential, sediment oxygen demand, total nitrogen, total phophorus, iron, manganese, mineralogical information or parameters for normalising contaminant data (e.g. aluminium, lithium, scandium- cf. Technical Annex II). Consideration should also be given to chemical or biochemical changes that contaminants may undergo when disposed of at sea.

　

Literature References related to Technical Annex I

　

ASTM, 1994. Standard guide for collection, storage, characterisation and manipulation of sediment for toxicological testing. American Society for Testing and Material, AnnualBook of Standards. Vol. 11.04, E1391-96.

EPA/CE, 1991. Evaluation of Dredged Material Proposed for Ocean Disposal: Testing Manual EPA-503/8-91/001. US-EPA Office of Water (WH-556F). EPA/CE, 1994. Evaluation of Dredged Material Proposed for discharge in Waters of the US. Testing Manual (Draft): Inland Testing Manual EPA- 823-B-94-002. (will be replaced by "coast of waters manual").

International Association of Dredging Companies (IADC)/Central Dredging Association (CEDA), 1997. Environmental Aspects of Dredging. Guide 3 (Investigation, Interpretation and Impact). ISBN 90-75254-08-3.

Gustavson, K. and Wangberg, S.A., 1995. Tolerance induction and succession in microalgae communities exposed to copper and atrazine. Aquatic Toxicology. 32: 283-302.

HELCOM, 1997. Draft Manual for Marine Monitoring in the Cooperative Monitoring in the Baltic Marine Environment (COMBINE Programme) of HELCOM. Part C - Programme for Monitoring of Eutrophication and its Effects. Annex C-4: Directives for sampling and analysis of hydrographic, chemical and biological variables. Annex C-8 : Soft bottom macrozoobenthos.

Murk et al., 1996. Chemical-activated luciferase gene expression (CALUX): a novel in vitro bioassay for Ah receptor active compounds in sediments and pore water. Fund. & Appolied Tox. 33: 149-160.

OSPAR, 1997 (available from the OSPAR Secretariat)

JAMP Eutrophication Monitoring Guidelines: Benthos - Technical Annex 1 (Hard- bottom macrophytobenthos and hard-bottom macrozoobenthos) - Technical Annex 2 (Soft-bottom macrozoobenthos)

JAMP Guidelines for Monitoring Contaminants in Sediments

Rees, H.L., C. Heip, M. Vincx and M.M. Parker, 1991. Benthic communities: use in monitoring point-source discharges. ICES Techniques in Marine Environmental Sciences No. 16.

Rumohr, H., 1990. Soft-bottom macrofauna: collection and treatment of samples. ICES Techniques in Marine Environmental Sciences No. 8.

　

Normalisation Techniques for Studies on the Spatial Distribution of Contaminants^*

　

　

Normalisation in this discussion is defined as a procedure to compensate for the influence of natural processes on the measured variability of the concentration of contaminants in sediments. Most contaminants (metals, pesticides, hydrocarbons) show high affinity to particulate matter and are, consequently, enriched in bottom sediments of estuaries and coastal areas. In practice, natural and anthropogenic substances entering the marine system are subjected to a variety of biogeochemical processes. As a result, they become associated with fine-grained suspended solids and colloidal organic and inorganic particles. The ultimate fate of these substances is determined, to a large extent, by particulate dynamics. They therefore tend to accumulate in areas of low hydrodynamic energy, where fine material is preferentially deposited. In areas of higher energy, these substances are "diluted" by coarser sediments of natural origin and low contaminant content.

It is obvious that the grain size is one of the most important factors controlling the distribution of natural and anthropogenic components in the sediments. It is, therefore, essential to normalise for the effects of grain size in order to provide a basis for meaningful comparisons of the occurrence of substances in sediments of various granulometry and texture within individual areas or among areas. Excess levels, above normalised background values, could then be used to establish sediment quality.

For any study of sediments, a basic amount of information on their physical and chemical characteristics is required before an assessment can be made on the presence or absence of anomalous contaminant concentrations. The concentration at which contamination can be detected depends on the sampling strategy and the number of physical and chemical variables that are determined in individual samples.

The various granulometric and geochemical approaches used for the normalisation of trace elements data as well as the identification of contaminated sediments in estuarine and coastal sediments has been extensively reviewed by Loring (1988). Two normalisation approaches widely used in oceanography and in atmospheric sciences have been selected here. The first is purely physical and consists of characterising the sediment by measuring its content of fire material. The second approach is chemical in nature and is based on the fact that the small size fraction is usually rich in clay minerals, iron and manganese ox-ihydroxides and organic matter. Furthermore, these components often exhibit a high affinity for organic and inorganic contaminants and are responsible for their enrichment in the fine fraction. Chemical parameters (e.g., Al, Sc. Li) representative of these components may thus be used to characterise the small size fraction under natural conditions.

It is strongly suggested that several parameters be used in the evaluation of the quality of sediments. The types of information that can be gained by the utilisation of these various parameters are often complementary and extremely useful considering the complexity and diversity of situations encountered in the sedimentary environment. Furthermore, measurements of the normalising parameters selected here are rather simple and inexpensive.

This report presents general guidelines for sample preparation, analytical procedures, and interpretation of physical and chemical parameters used for the normalisation of geochemical data. Its purpose is to demonstrate how to collect sufficient data to normalise for the grain-size effect and to allow detection, at various levels, of anomalous concentrations of contaminants within estuarine and coastal sediments.

　

　

Ideally, a sampling strategy should be based on a knowledge of the source of contaminants, the transport pathways of suspended matter and the rates of accumulation of sediments in the region of interest. However, existing data are often too limited to define the ideal sampling scheme. Since contaminants concentrate mainly in the fine fraction, sampling priority should be given to areas contining fine material that usually correspond to zones of deposition.

The high variability in the physical, chemical and biological properties of sediments implies that an evaluation of sediment quality in a given area must be based on a sufficient number of samples. This number can be evaluated by an appropriate statistical analysis of the variance within and between samples. To test the representativity of a single sediment specimen at a given locality, several samples at one or two stations should be taken.

The methodology of sampling and analysis should follow the recommendations outlined in the "Guidelines for the Use of Sediments as a Monitoring Tool for Contaminants in the Marine Environment" (ICES 1987). In most cases, the uppermost layer of sediments collected with a tightly closing grab sampler (Level in the Guidelines) is sufficient to provide the information concerning the contamination of the sediments of a given area compared to sediments of uncontaminated locations or other reference material.

Another significant advantage of using sediments as monitoring devices is that they have recorded the historical evolution of the composition of the suspended matter deposited in the area of interest. Under favourable conditions, the degree of contamination may be estimated by comparison of surface sediments with deeper samples, taken below the biological mixing zone. The concentrations of trace elements in the deeper sediment may represent the natural background level in the area in question and can be defined as baseline values. This approach requires sampling with a box-corer or a gravity corer (Levels II and III in the Guidelines).

　

　

Typical analytical procedures to be followed are outlined in Table 1. The number of steps that are selected will depend on the nature and extent of the investigation.

　

3.1 Grain size fractionation

It is recommended that at least the amount of material <63μm, corresponding to the sand/silt classification limit, be determined. The sieving of the sampleat 63μm is, however, often not sufficient, especially when sediments are predominantly fine grained. In such cases, it is better to normalise with lower size thresholds since the contaminants are mainly concentrated in the fraction <20μm, and even more specifically in the clay fraction (<2μm). It is thus proposed that a determination be made, on a sub-sample, of the weight fraction <20μm and that <2μm with the aid of a sedimentation pipette or by elutriation. Several laboratories are already reporting their results relative to the content of fine fractions of various sizes and these results may be useful for comparison among areas.

　

3.2 Analysis of contaminants

It is essential to analyse the total content of contaminants in sediments if quality assessment is the goal of the study, and it is thus recommended that the unfractionated sample (<2mm) be analysed in its entirety. The total content of elements can be determined either by nondestructive methods, such as X-ray fluorescence or neutron activation, or by a complete digestion of the sediments (involving the use of hydrofluoric acid (HF)) followed by methods such as atomic absorption spectrophotomety or emission spectroscopy. In the same way, organic contaminants should be extracted with the appropriate organic solvent from the total sediment.

An individual size fraction of the total sediment may be used for subsequent analysis, if required, to determine the absolute concentrations of contaminants in that fraction, Providing that its contribution to the total is kept in perspective when interpreting the data. Such size fraction information might be useful in tracing the regional dispersal of metals associated with specific grain-size fractions, when the provenance of the material remains the same. However, sample fractionation is a tedious procedure that introduces considerable risk of contamination and potential losses of contamination to leaching. The applicability of this approach is thus limited.

　

　

4.1 Granulometric normalisation

Since contaminants tend to concentrate in the fine fraction of sediments, correlations between total concentrations of contaminants and the weight percent of the fine fraction, determined separately on a sub-sample of the sediment by sieving or gravity settling constitute a simple but powerful method of normalisation. Linear relationships between the concentration and the weight percentage of the fine fraction are often found and it is then possible to extrapolate the relationships to 100% of the fraction studied, or to characterise the size dependence by the slope of the regression line.

　

4.2 Geochemical normalisation

Granulometric normalisation alone is inadequate to explain all the natural trace variability in the sediments. In order tointerpret better the compositional variability of sediments, it is also necessary to attempt to distinguish the sedimentary components with which the contaminants are associated throughout the grain-size spectrum. Since effective separation and analysis of individual components of sediments is extremely difficult, such associations must rest on indirect evidence of these relationships.

Since contaminants are mainly associated with the clay minerals, iron and manganese oxi-hydroxides and organic matter abundant in the fine fraction of the sediments, more information can be obtained by measuring the concentrations of elements representative of these components in the samples.

An inert element such as aluminium, a major constituent of clay minerals, may be selected as an indicator of that fraction. Normalised concentrations of trace elements with respect to aluminium are commonly used to characterise various sedimentary particulate materials (see below). It may be considered as a conservative major element, that is not affected significantly by, for instance, early diagenetic processes and strong redox effects observed in sediments.

In the case of sediments derived from the glacial erosion of igneous rocks, it has been found that contaminant/Al ratios are not suitable for normalising for granular variability (Loring, 1988). Lithium, however, appears to be an ideal element to normalise for the grain size effect in this case and has the additional advantage of being equally applicable to non-glacial sediments.

In addition to the clay minerals, Mn and Fe compounds are often present in the fine fraction, where they exhibit adsorption properties strongly favouring the incorporation of various contaminants. Mn and Fe are easily analysed by flame atomic absorption spectrometry and their measurement may provide insight into the behaviour of contaminants.

Organic matter also plays an important role as scavenger of contaminants and controls, to a major degree, the redox characteristics of the sedimentary environment.

Finally, the carbonate content of sediments is easy to determine and provides additional information on the origin and the geochemical characteristics of the sediments. Carbonates usually contain insignificant amounts of trace metals and act mainly as a diluent. Under certain circumstances, however, carbonates can fix contaminants such as cadmium and copper. A summary of the normalisation factors is given in Table 2.

　

4.3 Interpretation of the data

The simplest approach in the geochemical normalisation of substances in sediments is to express the ratio of the concentration of a given substance to that of the normalising factor.

Normalisation of the concentration of trace elements with respect to aluminium (or scandium) has been used widely and reference values on a global scale have been established for trace elements in various compartments: crustal rocks, soils, atmospheric particles, river-borne material, marine clays and marine suspended matter (cf., e.g., Martin and Whitfield, 1983; Buat-Menard and Chesselet, 1979).

This normalisation also allows the definition of an enrichment factor for a given element with respect to a given compartment. The most commonly used reference level of composition is the mean global normalised abundance of the element in crustal rock (Clarke value).

The enrichment factor EF is given by:

　

EF crust = (X/Al) sed/ (X/Al) crust

　

where X/Al refers to the ratio of the concentration of element X to that of Al in the given compartment.

However, estimates of the degree of contamination and time trends of contamination at each sampling location can be improved upon by making a comparison with metal levels in sediments equivalent in origin and texture.

　

These values can be compared to the normalised values obtained for the sediments of a given area. Large departures from these mean values indicate either contamination of the sediment or local mineralisation anomalies.

　

When other variables (Fe, Mn, organic matter and carbonates) are used to characterise the sediment, regression analysis of the contaminant concentrations with these parameters often yields useful information on the source of contamination and on the mineralogical phase associated with the contaminant.

　

A linear relationship between the concentration of trace constituents and that of the normalisation factor has often been observed (Windom et al., 1989). In this case and if the natural geochemical population of a given element in relation to the normalising factor can be defined, samples with anomalous normalised concentrations are easily detected and may indicate anthropogenic inputs.

　

According to this method, the slope of the linear regression equation can be used to distinguish the degree of contamination of the sediments in a given area. This method can also be used to show the change of contaminant load in an area if the method is used on samples taken over intervals of some years (Cato, 1986).

　

A multi-element/component study in which the major and trace metals, along with grain size and organic carbon contents, have been measured alows the interrelationships between the variables to be established in the form of a correlation matrix. From such a matrix, the most significant ratio between trace metal and relevant parameter(s) can be determined and used for identification of metal carriers, normalisation and detection of anomalous trace metal values. Factor analyses can sort all the variables into groups (factors) that are associations of highly correlated variables, so that specific and/or non-specific textural, mineralogical, and chemical factors controlling the trace metal variability may be inferred from the data set.

　

Natural background levels can also be evaluated on a local scale by examining the vertical distribution of the components of interest in the sedimentary column. This approach requires, however, that several favourable conditions are met: steady composition of the natural uncontaminated sediments; knowledge of the physical and biological mixing processes within the sediments; absence of diagenetic processes affecting the vertical distribution of the component of interest. In such cases, grain size and geochemical normalisation permits compensation for the local and temporal variability of the sedimentation processes.

　

　

The use of the granulometric measurements and of component/reference element ratios are useful approaches towards complete normalisation of granular and mineralogical variations, and identification of anomalous concentrations of contaminants in sediments. Their use requires that a large amount of good analytical data be collected and specific geochemical conditions be met before all the natural variability is accounted for, and the anomalous contaminant levels can be detected. Anomalous metal levels, however, may not always be attributed to contamination, but rather could easily be a reflection of differences in sediment provenance.

Geochemical studies that involve the determination of the major and trace metals, organic contaminants, grain size parameters, organic matter, carbonate, and mineralogical composition in the sediments are more suitable for determining the factors that control the contaminant distribution than the measurement of absolute concentrations in specific size fractions or the use of potential contaminant/reference metal ratios alone. They are thus more suitable for distinguishing between uncontaminated and contaminated sediments. This is because such studies can identify the factors that control the variability of the concentration of contaminants in the sediments.

　

References

　

Buat-Menard, P. and Chesselet, R. 1979. Variable influence of atmospheric flux on the trace metal chemistry of oceanic suspended matter. Earth Planet. Sc. Lett. 42: 399-411.

Cato, I., Mattsson, J. and Lindskog, A. 1986. Tungmetaller och petrogena kolvateni Brofjordens bottensediment 1984, samt forandringar efter 1972. /Heavy metals and petrogenic hydrocarbons in the sediments of Brofjorden in 1984, and changes after 1972. /University of Goteborg. Dep. of Marine Geology, Report No. 3, 95 pp. (English summary).

ICES, 1987. Report of the ICES Advisory Committee on Marine Pollution, 1986.

ICES Coop. Res. Report No. 142, pp 72-75.

Loring, D.H. 1988. Normalization of trace metal data. Report of the ICES Working Group on Marine Sediments in Relation to Pollution. ICES, Doc. C.M. 1988/E:25, Annex 3.

Martin, J.M. and Whitfield, M. 1983. River input of chemical elements to the ocean. In: Trace Metals in Sea-Water. C.S. Wong, E. Boyle, K.W. Bruland. J.D. Burton and E.D. Goldberg, Eds. Plenum Press, New York and London. pp 265-296.

Windom, H.L., Schropp, S.T., Calder. F.D., Ryan. J.D., Smith Jr., R.G., Burney, L.C., Lewis, F.G. and Rawlinson, C.H. 1989. Natural trace metal concentrations in estuarine and coastal marine sediments of the southeastern United States. Erviron. Sci. Tech. 23: 314-320.