Environmental Science and Pollution Research

, Volume 21, Issue 1, pp 17–27

Sediment quality guidelines: challenges and opportunities for improving sediment management

Authors

    • Nicholas School of the EnvironmentDuke University
  • Graeme E. Batley
    • Centre for Environmental Contaminants ResearchCSIRO Land and Water
  • Richard J. Wenning
    • ENVIRON
  • Lingyan Zhu
    • College of Environmental Science and Engineering, Key Laboratory of Pollution Processes and Environmental Criteria, Ministry of EducationNankai University
  • Marnix Vangheluwe
    • ARCHE Consulting
  • Shirley Lee
    • Hong Kong Institute of Environmental Impact Assessment
Environmental Quality Benchmarks for Protecting Aquatic Ecosystems

DOI: 10.1007/s11356-013-1778-7

Cite this article as:
Kwok, K.W.H., Batley, G.E., Wenning, R.J. et al. Environ Sci Pollut Res (2014) 21: 17. doi:10.1007/s11356-013-1778-7

Abstract

During the International Conference on Deriving Environmental Quality Standards for the Protection of Aquatic Ecosystems held in Hong Kong in December 2011, an expert group, comprising scientists, government officials, and consultants from four continents, was formed to discuss the important scientific and regulatory challenges with developing sediment quality guidelines (SQGs). We identified the problems associated with SQG development and made a series of recommendations to ensure that the methods being applied were scientifically defensible and internationally applicable. This document summarizes the key findings from the expert group. To enable evaluation of current SQG derivation and application systems, a feedback mechanism is required to communicate confounding factors and effects in differing environments, while field validation is necessary to gauge the effectiveness of SQG values in sediment quality assessments. International collaboration is instrumental to knowledge exchange and method advancement, as well as promotion of ‘best practices’. Since the paucity of sediment toxicity data poses the largest obstacle to improving current SQGs and deriving new SQGs, a standardized international database should be established as an information resource for sediment toxicity testing and monitoring data. We also identify several areas of scientific research that are needed to improve sediment quality assessment, including determining the importance of dietary exposure in sediment toxicity, mixture toxicity studies, toxicity screening of emerging chemicals, how climate change influence sediments and its biota, and possible use of new toxicity study approaches such as high throughput omic-based toxicity screenings.

Keywords

Environmental quality standardsSediment quality guidelinesAquatic ecosystem healthEcotoxicity

Introduction

Sediments are generally recognized as a sink for many substances in aquatic systems, and also a potential source of dissolved and particulate-bound contaminants to overlying waters (Batley et al. 2005). Consequently, sediment quality has emerged as an important and critical consideration for protection of benthic ecosystem health, fisheries conservation, and protection of surface water quality in both marine and freshwater environments (Babut et al. 2005; Wenning et al. 2005). The careful management of contaminated sediments is required for a range of activities including dredging projects, waterway restoration programs, recreational and commercial fisheries management, water quality protection, and natural resource restoration (Wenning et al. 2005).

Use of sediment quality goals

Since the early 1980s, one of the most important regulatory tools for sediment management has been the use of sediment quality guidelines (SQGs). The hypothesis that a concentration threshold could be identified for chemicals in sediments below which aquatic life was not harmed was first proposed as part of defining criteria to assess dredged material for disposal at sea by the USEPA and US Army Corps of Engineers in the early 1970s (USEPA/USACE 1973; Engler et al. 2005). One of the first applications of this hypothesis occurred in 1989 when the US National Oceanic and Atmospheric Administration (NOAA) proposed screening values for 190 chemicals in sediment as a means of gauging the quality of sediment in marine and freshwater environments (Engler et al. 2005). The primary intent of the NOAA effects-range-median and -low values (commonly referred to as ER-M and ER-L, respectively) has remained unchanged to the present—to establish a scientific rationale for taking corrective actions to protect aquatic ecosystems from exposure to chemicals in sediment that have the potential to harm aquatic life. NOAA’s SQGs continue to evolve as new knowledge becomes available, and have been either adopted directly or used as the foundation for deriving sediment quality values in several countries around the world (Batley et al. 2005).

It is increasingly apparent that an essential first step for any regulatory body considering the adoption of an environmental quality paradigm involving the use of SQGs is to identify the specific regulatory objectives and desired environmental management and environmental quality goals. The construct of the environmental quality program can directly influence decisions regarding either the adoption of SQG values from other governments or the derivation of different values that may or may not be more ecologically relevant to the environmental conditions that the regulatory agency is charged to protect. The regulatory objectives and environmental quality goals should balance the needs of different private and public sector stakeholders, social and cultural values concerning the environment, and public safety in the context of a water body’s value as a resource for food and recreation.

Therein lies one of the most difficult scientific challenges that must be resolved for current and future uses of SQGs in a regulatory context. A growing body of scientific evidence suggests that differences in persistence, bioaccumulation, speciation and toxicity of many organic compounds and metals in surface waters and sediments in freshwater and marine systems, and aquatic organisms in arctic, temperate, and tropical ecosystems respond differently to chemical exposure (Maltby et al. 2005; Kwok et al. 2007; Rombke et al. 2008; Daam and Van den Brink 2010; Sanchez-Bayo and Hynet 2011; Wenning et al. 2011). For example, Fig. 1 illustrates how environmental factors differ in tropical and temperate regions and the likely biological and chemical differences that have been observed in the aquatic environment. The dose–response characteristics of sediment-bound chemicals in tropical aquatic environments have not been well studied, yet regulatory agencies in many countries have adopted SQGs—particularly those from North America—without adequate consideration of environmental differences that might make such guidelines more or less relevant (Babut et al. 2005).
https://static-content.springer.com/image/art%3A10.1007%2Fs11356-013-1778-7/MediaObjects/11356_2013_1778_Fig1_HTML.gif
Fig. 1

Environmental factors in tropical and temperate environments and the associated changes in biological and chemical characteristics (from Wenning et al. 2011)

An issue of perspective

From a scientific perspective, environmental quality objectives for sediments are primarily driven by the aquatic life and ecosystem services associated with a body that are valued either inherently or by society. The SQGs that evolve from setting environmental quality objectives could vary in different aquatic environments depending on natural physical and chemical characteristics such as those that sharply distinguish lentic and lotic aquatic ecosystems. Similarly, SQGs may be focused differently to reflect the environmental characteristics in different types of estuarine ecosystems and those associated with coastal and deep water marine environments. Another important factor influencing decisions on environmental quality (and resulting SQGs) is the biological resource itself; the protection of sensitive or threatened and endangered species, or to minimize the bioaccumulation of certain chemicals in aquatic species important to human consumption, is often the basis for establishing objectives and deriving chemical-specific SQGs.

From a regulatory perspective, environmental quality objectives are typically set to reflect policy decisions that, more often than not, closely reflect economic considerations and government goals to foster improvements in the quality of life. While pressures from non-governmental organizations (NGOs), large financial institutions, and other national governments may have some influence on the breadth and tone of environmental quality objectives, there are rarely any external influences on the SQGs and other environmental quality criteria selected for regulation and enforcement. More often than not, the SQGs currently found in the regulatory programs of many countries reflect the scientific knowledge available at that time about chemical and biological interactions and awareness of similar approaches used in other countries. It is only in more recent years that many of these same countries are undertaking a re-examination of their environmental quality benchmarks with the aim to be more environmentally relevant to the watersheds and aquatic biota found within their national borders. For countries and regulatory jurisdictions with extensive coastal areas, some bold and tough decisions are beginning to be made regarding an acceptable balance between ecological risk and the economic losses and benefits associated with setting chemical-specific sediment and water quality threshold limits (Borja et al. 2006; Iovannaa and Griffiths 2006).

From an economic perspective, sediment management is often an integral element of large coastal and river waterfront infrastructure projects. Project managers are often under considerable financial pressures to deliver completed work on a specific schedule and within a prescribed budget; the opportunity to generate site-specific information addressing the significance of contaminants in sediment and the ecological impact of remediation activities is rare or, at the very least, inadequate to properly inform decision making. For example, waterfront development projects in Hong Kong’s Victoria Harbor typically involve decisions on sediment dredging permits within 30 days of the permit application. Under these time constraints and the often large financial investments involved, SQGs must have a rational and scientifically sound basis because the Hong Kong Special Administrative Region Government considers the use of SQGs as a legally defensible decision criterion for approving or rejecting waterfront development projects (ETWB 2002; Shirley Lee, personal communication).

Since the granting of permits is crucial for the timely delivery of many major projects in most countries, the validity and effectiveness of SQG values in protecting the environment often generate intense debates and conflicting views amongst stakeholders with different interests. NGOs typically argue for a precautionary approach that favors more stringent SQGs to preserve biodiversity and safeguard ecological resources, whereas commercial interests often urge a more pragmatic approach on economic grounds. Both stakeholder groups generally agree that it is imperative to derive environmentally relevant SQGs that are appropriate to use for a range of reasonable environmental conditions. Setting aside the new and somewhat unpredictable future challenges posed by global climate change, it remains difficult to define a reasonable range of environmental conditions. Hence, scientists increasingly urge consideration of site-specific SQGs and the use of quantitative models that can be quickly applied to derive chemical concentration limits in sediment when chemical and biological conditions change abruptly (Wenning et al. 2005).

An issue of risk

Increasingly in many countries sediment management is a risk-based process, wherein SQGs are used to identify sediments that are clearly harmful to aquatic life, may be harmful to some degree to certain aquatic species, or pose no threat. SQGs are also used to examine in-water locations where dredged materials can be disposed safely (i.e., the chemical concentrations at the surface in the deposited sediment will not harm aquatic life). This approach is embodied in several sediment risk assessment paradigms involving the use of SQGs as an initial screening tool to evaluate sediment chemistry information before undertaking more detailed exposure modeling and ecotoxicology evaluations (e.g., ANZECC/ARMCANZ 2000; ETWB 2002). The U.S. Environmental Protection Agency (USEPA) guidance on managing contaminated sediments (USEPA 2005a) and U.S. Army Corps of Engineers program for dredged material assessment (US Army Corps of Engineers 2004) are the two most often cited regulatory programs where risk assessment is used to inform decision making.

The management actions that follow from a risk-based approach depend on the likely significance of the action, the resources available, and stakeholder concerns. If the source of the contamination can be identified, then a mandate to remediate with legal enforcement may be possible depending on the regulatory jurisdiction. Sediment remediation, however, is often expensive and time-consuming. Risk assessment is increasingly used to guide the prioritization of sediment areas identified through SQG screening evaluations as contaminated, and to identify whether exposure and risk can be effectively mitigated by dredging, capping, or enhancement of natural recovery processes (USEPA 2005a). For governments facing financial constraints or unable to identify the private parties responsible for the contamination, the possibility to eliminate the source of the contamination or control the contaminated sediment deposit through isolation of the affected area or capping can be an effective short-term solution (USEPA 2005b).

The language of SQGs

The meaning and words used by different regulatory jurisdictions around the world when referring to sediment quality are different, and often subtle to be compatible with policy intent or legal authority. These differences create a different type of language barrier that has often generated confusion and miscommunication among decision makers and between regulatory authorities. If the current practice among countries for setting environmental quality goals independently continues, then achieving a common language addressing sediment management, risk assessment, and SQGs will remain difficult.

In the USA, Canada, Australia, New Zealand, and Hong Kong among others, sediment quality guideline is commonly used to describe the chemical-specific values used to assess—often as part of a preliminary screening exercise—the significance of chemicals found in sediment. Other countries addressing sediments directly such as Norway (Bakke et al. 2010) refer to sediment quality criteria, while others refer to sediment quality standards; both terms, by comparison, denote the chemical-specific values specified in a government regulation as legally enforceable limits. The quality of sediment, to some degree, is determined by the number of chemicals that occur above their specific SQG values, and the degree to which those values are exceeded.

For example, Australian and New Zealand (ANZECC/ARMCANZ 2000) water quality guidelines explicitly state that “… guidelines should not be used as mandatory standards because there is significant uncertainty associated with the derivation and application of water quality guidelines…. The user should be aware of this uncertainty when determining if an environmental value has been supported or not” (ANZECC/ARMCANZ 2000). Their guidance points to the lack of data on biological effects for many species indigenous to Australia and New Zealand; uncertainty about chemical behavior in the environments under their jurisdiction; and the absence of sufficient monitoring data and understanding of ambient water quality. Similar caution applies to the use of SQGs in many other parts of the world.

Derivation of SQGs

A number of SQGs have been developed for relating chemical concentrations in sediment to their potential for biological effects, and several are used in different countries. An understanding of the environmental quality objectives and the science underpinning the derivation methods used to set SQG values in different regulatory context is useful because the uncertainties and limitations associated with each of the different derivation methods can be important (Batley et al. 2002; Wenning et al. 2005). The derivation of SQGs is largely based on either (1) an empirical correlation of effects data for co-occurring sediment contaminants, (2) mechanistic predictions using equilibrium partitioning combined with water quality guidelines, or (3) a combination of both empirical and mechanistic approaches to arrive at a consensus-based guideline (Batley et al. 2005). According to Wenning et al. (2005), none of the three approaches are fundamentally flawed, but each has unique uncertainties and limitations. Empirically derived SQGs are widely used and typically developed to identify an upper and lower value for specific contaminants from statistical associations using matched sediment chemistry and biological effects data. The upper value represents a mean or median concentration in sediment associated with either a specific biological response or any adverse effect. The lower value represents a concentration in sediment where the biological response or any adverse effect has been found in a small portion of available studies, typically 10 % of the available studies. Adverse effects on benthic organisms are generally considered rare if sediment concentrations are below the lower SQG value. Numerous studies have indicated that empirically derived SQGs are sufficiently predictive of adverse effects and a good basis for making conclusions about the likely impairments attributable to contaminated sediments in aquatic ecosystems, although many of these predictions have been with the upper guideline values (e.g., ERM), which logically will be more predictive than the lower value. There remains a need to focus on protection of the sediment ecosystem so the lower guideline remains the most important (Batley et al. 2005). Multiple collections of several empirically derived approaches based on different statistical approaches have been developed for marine and freshwater environments (Long et al. 1995; MacDonald et al. 1996; Barrick et al. 1988; Fairey et al. 2001; Field et al. 2002). It is also important to note that most empirical SQGs are based on acute toxicity test data, which is quite different from water quality guidelines that are often based on chronic data or acute data converted with an acute-to-chronic ratio. Nonetheless, Simpson and Spadaro (2011) have recently showed that when considering dilute acid-extractable metal concentrations, the upper guidelines are probably predictive of the onset of acute lethality effects (and plenty of chronic effects) and the lower guidelines (trigger values) are predictive of the onset of chronic/sublethal effects.

Consensus-based SQGs are an evolution of the empirical approach and aggregate several different SQGs having a similar narrative intent (e.g., median effect). Marine consensus SQGs have been developed for some constituents, including metals, polychlorinated biphenyls, and polycyclic aromatic hydrocarbons (Swartz 1999; MacDonald et al. 2000; Vidal and Bay 2005).

The mechanistic approach models the chemical and biological processes that affect contaminant bioavailability. Current mechanistic SQGs are based on equilibrium partitioning theory and apply to selected classes of contaminants, primarily divalent metals and several types of non-ionic organics in the USA (USEPA 2003, 2005b, 2008). While these models are useful for describing contaminant bioavailability, mechanistic SQGs address the question of what may be causing sediment toxicity, not whether or not sediment will be toxic (Burgess et al. 2013). Environmental parameters needed to apply these guidelines to metals include sediment acid volatile sulfides and simultaneously extracted metals.

It is unclear which empirical SQG approach is the most effective for describing the potential for biological effects associated with chemical contamination. Numerous studies have shown that each SQG approach has a somewhat similar predictive ability with respect to biological effects, but most studies have generally been limited to examination of just one or two approaches and often use variable methods to measure performance (Wenning et al. 2005). Long et al. (2000) applied ERMs and PELs to several data sets and observed different patterns in predictive ability. Vidal and Bay (2005) compared five SQG approaches using a common dataset and found large differences in predictive ability among some approaches; however, their study did not include the logistic regression modeling approach. A later study (Bay et al. 2012) included this approach and indicated that they all had similar predictive ability. Vidal and Bay (2005) also observed that comparisons of SQG performance can be strongly influenced by the selection of thresholds used to classify the results. Existing studies are inadequate for comparing the performance of empirical SQGs because of their limited scope, lack of comparability in methods, and lack of thresholds derived using a consistent methodology.

It is also unclear whether performance of SQGs is improved when they are calibrated to local conditions. The predictive ability of SQGs to biological effects has been shown to vary when the same guidelines are applied to data from different regions (Long et al. 1998; 2006; O’Connor et al. 1998; Fairey et al. 2001; Vidal and Bay 2005). These variations in performance may be due to differences in the nature of the chemical mixtures between sites or regions, variations in bioavailability due to geochemical factors, or differences in the sensitivity of methods used to measure biological effects. Variation in SQG performance among studies creates uncertainty in determining when the SQG threshold associated with adverse impacts is exceeded. The use of SQGs and interpretation thresholds that are derived or calibrated relative to site-specific conditions has been recommended as a way to reduce the uncertainty of SQG interpretation (Fairey et al. 2001; Vidal and Bay 2005; Long et al. 2006).

It is generally recognized that a higher degree of certainty regarding the significance of contaminants in sediments is possible when SQGs are coupled with the examination of several different environmental conditions in the aquatic environment. The use of a weight-of-evidence approach melds the consideration of information derived from the use of SGQs with information derived from other considerations such as benthic community surveys, bioaccumulation studies, and field-based ecological assessments (Linkov et al. 2009; Suter and Cormier 2011). A weight of evidence framework for decision making has been proposed in several countries such as the USA (USEPA 2005a, b; Bay and Weisbery 2012) and in Australia and New Zealand (Batley and Simpson 2009); however, it has not been widely adopted in sediment risk assessment practices in other parts of the world.

More recent refinements to empirical, mechanistic, and consensus-based approaches address the environmental factors that have the potential to modify chemical and metal bioavailability in sediments (Simpson and Batley 2007; Strom et al. 2011), notably sediment grain size and particulate organic carbon. The sediment biotic ligand model addresses the mechanistic behavior of metals in sediment (Di Toro et al. 2005). While historically there has generally been poor discrimination between freshwater and marine sediment studies, current approaches recognize the environmental significance of fresh and salt water on chemical and metal behavior and clearly distinguish the information derived from either aquatic environment (Smith et al. 1996).

Where the management need is to define differing levels of protection for specific applications similar to the approach used to define water quality guidelines, the use of species sensitivity distribution (SSD) curves can contribute to the derivation of SQGs. For example, Simpson et al. (2011) applied a SSD of laboratory-derived toxicity data for copper in marine sediments to derive a SQG for copper that varied in a predictable way with sediment grain size and organic carbon content. Copper SQGs that vary with sediment properties are a significant improvement over reliance on a single SQG value for all aquatic environment conditions. However, the absence of a broad array of test species and associated toxicity data relevant to different aquatic environments is an important limitation in the use of SSD curves to support different SQG derivation methods. Further, such an approach would need to be undertaken for each metal of interest.

While field ecological measurements are an important line of evidence in sediment quality assessments, ecological data have contributed minimally to the underpinnings of effects-based empirical SQGs. Recently, Kwok et al. (2008) used a field-based SSD (f-SSD) of biological community response to mixtures of contaminants to derive site-specific guidelines that were surprisingly similar to some of the effects-base guidelines derived from predominantly ecotoxicity data for single species. The f-SSD approach is site-specific, but does address natural mixtures of contaminants and local sediment properties, and overcomes the issues with laboratory to field comparisons and extrapolating from single to multiple contaminants. It does, though, consider only two lines of evidence, chemistry and ecology, and so might miss more subtle toxic effects.

Although SQGs have been developed and used for more than two decades, field validation of their effectiveness is largely underutilized. Sediment quality and benthic community surveys are routinely carried out in many countries. Such surveys can also be useful for compliance checking and to form adaptive monitoring strategies. Yet current sampling designs for most these surveys did not collect the type of data that are useful for these purposes. For instance, lack of measurements of acid volatile sulfides making the normalization to bioavailable metal concentrations difficult, and missing the opportunity to validate bioavailability models for metals developed in laboratory settings (Martello et al. 2007). Moreover, as proposed by a number of scientists, data from these surveys can be used for SQG evaluation or derivation (Leung et al. 2005; Kwok et al. 2007; Bjorgaester and Gray 2008). To maximize the utility of such surveys, sample design can be optimized to provide necessary types of data to validate and improve SQGs.

Recent developments in other countries

Work underway or recently completed in several countries highlights the nature of the research and regulatory introspection that could be expected in developing countries when undertaking work aimed at making decisions regarding the choice and applicability of SQGs. A historical perspective on the evolution of different approaches to developing SQGs in different countries is available elsewhere (Burton 2002; Den Besten et al. 2003; Babut et al. 2005; Engler et al. 2005). A recent review conducted in South Africa (DEA 2012) to inform on that country’s deliberations regarding SQGs may be particularly useful; the review addresses action levels and guideline values in different countries are used for making decisions on the suitability of dredged material for both confined and unconfined, open water disposal.

In Norway, SQGs have evolved from reliance on statistical distributions of contaminant concentrations in sediments to include the consideration of concentration intervals above background concentrations (Molvær et al. 1997). Other considerations include SQG values incorporating effects data for both water and sediment (SFT 2007; Bakke et al. 2010). Initially, five levels of contamination were derived (background, good, moderate, bad, and very bad) from frequency distributions of selected contaminants in the field. These boundaries are now defined as the upper limit of background concentrations, the predicted no effects concentration (PNEC) for chronic exposure, the PNEC for acute exposure, and two to ten times above the acute PNEC.

After a prolonged consideration, the European Union (EU) as part of its Water Framework Directive has recently provided technical guidance for the derivation of SQGs for EU member states (EU WFD 2010). Historically, sediment quality was assessed using chemical quality criteria (Ahlf et al. 2002; Environment Agency 2002; Den Besten et al. 2003). Over the past decade, biological effects-based assessment approaches have gained more interest to assess in situ risks in ports and waterways where sediment quality is either suspected to deteriorate or to determine management options for dredged material disposal. At present, however, research and regulation of contaminated sediments within the EU lacks coherence; EU member states continue to develop SQGs and monitoring strategies independently (Crane 2003; Babut et al. 2005). The absence of uniformity in sampling methods, analytical techniques, and applied sediment quality standards or guideline values within the EU (Brils 2008) hampers an overarching reliable assessment of sediment quality in Europe.

A new impetus for the derivation of environmental quality standards in Europe is the introduction of the Water Framework Directive (WFD 2000/60/EC). The WFD is the most important piece of water legislation developed in Europe over the last years and aims to reach good chemical and ecological status in inland and coastal waters by 2015. Achieving a good ecological status of surface waters as required by the WFD is expected to require improvements in sediment management (De Deckere 2011). Article 16 (7) in the WFD requires the European Commission to identify priority substances presenting a significant risk to aquatic environment and to set environmental quality standards for those substances in water, sediment, and/or biota. According to the Scientific Committee on Health and Environmental Risks (SCHER 2010), the current state of technical and scientific knowledge is mature enough to support the development of legally binding standards for sediments and/or biota.

Technical guidance relevant to the WFD was published in 2011 (WFD 2011), and suggests a tiered assessment framework for sediments. The derivation process is based on that used for effects assessment under REACH (ECHA 2006), but with an additional consideration of field or mesocosm data. This enables the use of different lines of evidence (e.g., sediment toxicity tests and aquatic toxicity tests in conjunction with partition equilibrium and field/ mesocosm studies) to generate standards.

In Australia and New Zealand, SQGs established in 2000 are being updated with revised values for a number of contaminants plus the introduction of a numerical weight of evidence assessment scheme (Simpson et al. 2010). The intent is to provide an intuitive approach that minimizes the use of best professional judgement in sediment quality assessment work. The revisions in Australia and New Zealand reflect new scientific findings. New societal and scientific concerns such as those associated with the effects of global warming and ocean acidification also need to be considered. This is similar to the regulatory approach evident in the USA, Canada, and elsewhere where freshwater and marine SQGs are routinely reviewed and improved as new sediment toxicity test data using relevant indigenous species become available, thereby reducing reliance on species extrapolation and the use of freshwater species data to derive marine SQGs.

Current challenges

Without question, there is increasing recognition of the need to develop site-specific environmental quality guidelines, including SQGs. For many countries, this is a costly undertaking, which often necessitates the use of guidelines from other countries as an initial step in the process of developing national regulations. In many parts of the world today, there is insufficient matching sediment chemistry, toxicity, and benthic invertebrate community data in coastal and riverine waters to support the derivation of numeric guidelines. To date, it has been more cost-effective to adopt screening values; if screening values are exceeded, site-specific studies toxicity and ecology are prescribed to address the data gaps and determine the risk to aquatic life. Depending on the management objective, the use of a screening approach could be readily modified by adopting more site-specific guidelines to meet the desired level of ecosystem protection. Most SQG approaches provide median and 90th percentile values (unlike the Norwegian approach, which specifies guidelines for five different conditions). For large sites, the application of the f-SSD approach of Kwok et al. (2008) may be an option, noting that this approach examines the effects on the abundance of benthic fauna only.

The ultimate goal in most regulatory jurisdictions is to modify the SQGs available elsewhere based on toxicity testing methods relevant to local environmental conditions and using indigenous species. There are uncertainties in adopting guidelines from other jurisdictions. A survey of guidelines demonstrates the variability (Burton 2002). It is important that lower “protective” guidelines are used rather than upper values that are generally predictive of effects and hence underprotective. This can be achieved if guidelines are used as screening values that if exceeded require further lines of evidence to be investigated, in any sediment quality assessment. Such screening frameworks need to consider contaminant bioavailability and background concentrations. Biological responses to contaminant exposures may differ significantly in different geographic locations, although some indications suggest the difference is likely to be small (Maltby et al. 2005; Kwok et al. 2007). These differences could have imposed significant constraints on economic development, regulations, and environmental consequences if unaddressed. The differences can be attributed to several environmental factors that contribute to altering the behavior of substances, as well as the presence or absence of species unique to a particular environment (Kwok et al. 2007; Wenning et al. 2011).

According to Nascimento (2007), several countries in South America are working to develop SQGs that are meaningful to the contaminant dynamics and unique biological conditions evident in the tropical and high-altitude environments found throughout the continent. The more significant issue might be related to the background concentrations at a particular location, where, in highly mineralized sediments, for example, the guideline for some metals might be exceeded. For this reason, an improved understanding of background conditions, modifying factors for bioavailability and toxicity, and subsequently how these impact under different management-specific conditions (e.g., dredging and dredged sediment disposal), is required.

The paucity of sediment toxicity data, and most of them derived from acute lethality tests, presents the largest difficulty in SQG development. This is in part a consequence of the lack of universally accepted sediment toxicity testing protocols. The lack of sediment toxicity studies has also hampered the ability to define cause and effect relationships between contamination and toxicity. In the field, the effects of sediment contamination are also difficult to define due to presence of confounding factors such as the presence of nutrients, hypoxic conditions, and ammonia in sediment. This is further complicated by a lack of general understanding of background concentrations of naturally occurring substances in sediments. A lack of technical expertise and experience in industry and government also impedes the development of SQGs. There is, therefore, a general lack of SQGs for many contaminants. For instance, there is a need for guidelines for endocrine-disrupting chemicals and many new pesticides.

Most of the current single-chemical SQGs do not account for mixtures effects, dietary exposure, and trophic transfer. The use of mean SQG quotients (ratios of measured concentrations to guideline values) to estimate overall mixture effects based on single-chemical guidelines (Long et al. 2006) goes some way to this, but the mean value may hide the exceedance of one contaminant by a number of non-exceeding contaminants.

Sediments are a multi-dimensional problem, yet the adoption of common SQGs for trans-boundary pollution has been difficult to achieve. This could be across different regions of the same nation. For example, Guangdong Province and Hong Kong have different environmental quality standards despite sharing a common interest in the preservation of the Pearl River. A similar situation is evident with between ASEAN countries sharing borders with the Mekong River (Mekong River Commission 2013). A flexible framework is needed to accommodate regulatory differences between regulatory partners, and ideally harmonization of standards between coastal countries and regions should be the objective.

Recommendations

To overcome many of these and other challenges, studies that involve the application of current available SQG derivation approaches should include a feedback mechanism to communicate confounding factors, effects in differing environments, and other findings that may cause SQG values to be reevaluated. Field validation will improve understanding of the effectiveness of SQG values in sediment quality assessments. This can be achieved through modifications of existing sediment quality and benthic community surveys conducted regularly by many countries.

Promoting international society and meetings is instrumental to review progress and to promote new methods and new discoveries, and setup new regional groups of existing professional organizations (e.g., SEDNET in Europe, SETAC Sediment Advisory Groups). Special sessions and workshops during international conferences can target to identify priority chemicals or metals in different regions or countries for focused knowledge exchange and to harmonize assessment and testing methods. Such platforms can also be used for education to new environmental managers and scientists, and knowledge exchange on specific topics including new development in sediment physical testing methods (e.g., grain size), chemical testing methods (e.g., TOC), and toxicity testing methods. Practical manuals documenting such information are also needed (e.g., Simpson et al. 2005).

Given that the paucity of sediment toxicity data poses the largest hurdle to deriving new SQGs, standardized international databases could be useful and should be established as an information resource for toxicity testing and monitoring data. This would involve compiling local databases, translation of data into English, and quality assurance via an inter-laboratory calibration network or international societies. To achieve these goals, collaborations and funding must be secured. The first step would be to setup a working group forming strategy to setup the sediment database, including decisions of quality control, content, review process, funding, computer requirements, and relationship to other existing databases in other countries.

Governments are encouraged to support startup funding to source and culture test organisms relevant to local environments for sediment toxicity tests. This can convince local industries to also play a role. Meanwhile, local aquatic biologists should provide information and basic research about culturing and rearing species to support water and sediment toxicity evaluations.

There are also several areas of scientific research that are needed to improve sediment quality assessment, including determining the importance of dietary exposure in sediment toxicity, mixture toxicity studies, toxicity screening of emerging chemicals, how climate change influence sediments and its biota, possible use of new toxicity study approaches (e.g., biomarkers and high throughput omic-based toxicity screening). Methods to better measure the dose received by the test organism need to be incorporated into field studies if we are to improve the protection afforded by SQGs, by accurately predicting an absence of toxic impacts. Recent research on the application of eco-genomics in the ecological assessment of sediments has the potential to significantly enhance our ability to see changes in benthic communities resulting from contaminants (Chariton et al. 2010). Rather than the common practice of identifying some 40 or so species using optical microscopy, the DNA signals from thousands of species can be collected and assigned identified in one hit, from a benthic organism homogenate. Such eco-genomic data can be easily integrated with the existing f-SSD approach for deriving SQGs.

More education of policy and decision makers may be needed to communicate the function of benthic species in aquatic ecosystems and to highlight the importance of proper management of sediments. Coupled with a weight of evidence approach, chemicals in sediment that are above SQG values do not necessarily imply that the sediment is harmful to benthic communities, aquatic life, or human health. Education is needed on the proper consideration of uncertainties and the constraints imposed by the current state of knowledge. It is evident that SQGs have an important role in sediment risk assessment, and can inform decision making for management of the aquatic environment.

Acknowledgments

The authors wish to thank the Hong Kong Special Administrative Region Government for supporting the International Conference on Deriving Environmental Quality Standards for the Protection of Aquatic Ecosystems (EQSPAE-2011) held at the University of Hong Kong during 3–7 December 2011. Specially, the authors would like to thank Vincent Chan, Tae Seob Choi, Jing Ding, Juan Du, Jie Fu, Jinbo Gao, In Ae Huh, Hojeong Kim, Jong-Hyeon Lee, Kenneth M. Y. Leung, Hong-Bo Li, Huizhen Li, Wenhua Liu, and Daniel Yang for participating the discussion syndicate on the development of sediment quality guidelines during EQSPAE-2011 conference. The authors are grateful to Kenneth M. Y. Leung (the Chairman of EQSPAE-2011 conference) and Stuart Simpson (CSIRO) for their constructive comments on early drafts of this paper.

Disclaimer

The views and opinions contained in this work are solely those of the authors and do not necessarily reflect the views and opinions of their supporting organizations.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013