August 15, 2007
Vol. 41, Iss. 16

• 
• 
• 
• 
• 

Vol. 41, Iss. 16
pp 5582–5588

Viewpoint

Formulating the Problems for Environmental Risk Assessment of Nanomaterials

The diversity of nanomaterials challenges us to consider how we prioritize them for environmental risk assessment.

Do manufactured nanomaterials pose risks to the environment? Should we commit large amounts of funding to detailed assessments of their potential environmental risks? At a time when several countries and international organizations are considering these issues and when vocal calls are being made for significant funding allocations in this area (1, 2), we pause to reflect on how we arrived at asking these questions. We then ask what can be done to help us decide with more confidence whether this is a priority that needs to be assessed in detail.

Our analysis suggests that, although many of us understand quantification of exposure and hazard as being important parts of environmental risk assessment, two initial components are critically important but sometimes overlooked: problem formulation and prioritization (3). These help us define the actual problem in the context of the environment as well as how it should be addressed for one or more nanomaterials. They also help us prioritize whether it is worth spending time and money on a more detailed quantification of environmental risk for any given material. This process, which is informed by numerous factors, including scientific evidence, public opinion, and perception of risks and benefits, forms the basis for an international approach to tiered risk assessment. It is designed to help us make better decisions about where we should invest limited resources.

We argue that the most urgent need is to formulate the problems correctly and understand the wider context in which they are framed. Establishing a harmonized framework with guiding principles for this process can help. Such a framework can help us make better, more defensible decisions about environmental risk assessment, in terms of both the level and direction of funding allocations, and develop the evidence base required to inform appropriate controls. Continuous innovation suggests that the process needs to be iterative, inclusive, and supported by enabling tools to ensure that robust and consistent data feed into it. And these issues go beyond nanomaterials themselves. They challenge us to ask how we can consistently and transparently formulate the problems for environmental risk assessment of emerging technologies in a far wider sense, and perhaps just as importantly, who should have a role in this process.

Shaping the agenda: from gray goo to nanoparticles

Our analysis starts with a short retrospective of how the agenda for environmental risk assessment of nanomaterials has been shaped during the past few years. It is important at the outset to emphasize (and welcome) the prospective approach to environmental risk assessment of nanomaterials. This contrasts with some notable cases in which quantitative risk assessment was prompted after adverse effects (e.g., endocrine disruption) were observed in the environment. This has not been the case for nanotechnologies. In fact, several years ago, concerns over nanotechnologies were not focused on the environment at all but rather on autonomous "nano robots" or "gray goo" self-replicating in an uncontrolled manner. Although later dismissed as highly unlikely and a distraction from the important issues (4), these fears did serve to kindle the public imagination and raise the profile of the issue.

So, what are considered the important environmental issues for nanomaterials? By way of illustration, it is useful to consider what has happened in the U.K. In 2003, its government commissioned two of its leading scientific bodies, the Royal Society and the Royal Academy of Engineering, to undertake an independent study investigating the opportunities and uncertainties associated with nanotechnologies, including environment, health, and safety aspects. Good consultation and an understanding of the need for public engagement and stakeholder dialogue were important features of this study. Their seminal report, published in July 2004 (4), established some important preliminary thinking for occupational and environmental risk assessment of nanomaterials. It highlighted something that has been confirmed subsequently: a lot is at stake. Nanotechnologies are expected to be a trillion-dollar industry within decades, bringing significant potential socioeconomic, health, and environmental benefits. The number of nano-based consumer products already on the market is considerable (5). The report concluded that realizing those benefits would require substantial investment in risk assessment and risk communication. Importantly, it also concluded that “many applications of nanotechnologies pose no new health or safety risks. . . . Currently we see the health, safety and environmental hazards of nanotechnologies as being restricted to discrete manufactured nanoparticles and nanotubes in a free rather than embedded form” (4).

The focus had moved away from what seemed to be rather far-fetched concerns about gray goo. The report, through careful examination of the underpinning science, showed that particles <0.1 µm could have changed properties (and, by implication, differing potential risks) when compared with the same material in bulk form. Whereas strict adherence to a 100-nm size threshold has subsequently been debated, good evidence existed for altered behavior and toxicity in the nanorange, largely from the literature on ultrafine nanoparticles and pulmonary toxicology. A key recommendation was further investigation of one specific, and arguably less emotive, sector of nanotechnologies—free manufactured nanoparticles.

This key conclusion was an important catalyzing statement of problem formulation and prioritization. It has been important in shaping much of the subsequent detailed discussions and research agendas on nanotechnologies risk assessment since the summer of 2004 (6).

Extrapolating to the wider environment: does one size fit all?

“A key unifying principle within problem formulation and throughout risk assessment is the connection between the source [of the hazard], the pathway, the receptor and the impact. It is important that connectivity or potential connectivity between these four components can be shown. If any of these components is missing then the risk assessment need go no further” (3).

Toxicological studies involving ultrafine particles, for example in atmospheric pollution, coal dust, and asbestos, provide substantial evidence to suggest a potential issue with free manufactured nanoparticles when exposure occurs via atmospheric pathways (4, 7). These studies point to important factors that might enhance or mitigate hazard, such as particle size, shape (e.g., length and diameter), biosolubility, and durability. We understand a fair amount about what happens when incidental nanoparticles deposit in the mammalian airways and lungs, including oxidative stress, inflammation, and pathology (4, 7, 8). Studies at a cellular (i.e., in vitro) level, for example with human cell lines, also point to oxidative stress as being mechanistically important (9). This understanding has helped focus similar hazard studies for manufactured nanoparticles, some of which have now been published (8, 10, 11). In some cases, manufactured nanoparticles might also be released into the atmosphere, for example from nanoparticulate fuel additives or from the use of nanomaterial aerosol products indoors. Of course, this by no means implies that significant risks are associated with any given manufactured nanoparticulate substance: sufficient exposure and hazard are necessary. But at least an understanding of incidentally formed atmospheric ultrafine nanoparticles highlights the potential connectivity. Therefore, human exposure to manufactured nanoparticles via the atmosphere is one area we might wish to consider for prioritization and quantitative environmental risk assessment.

But what if we go beyond this specific exposure–effects scenario and into the wider environment of aquatic and soil ecosystems and the microbial, invertebrate, and plant realms? In some cases, potential connectivity could be developed. Risks to microbial communities might be one such area (12). We could, by way of example, start to establish connectivity between discharges of antimicrobial nanoparticles (e.g., silver or even other nanomaterials; 13, 14) and possible impacts on microbial communities (e.g., in sewage treatment plants and surface waters). However, when nanomaterials and the environment are generally considered, in most cases the potential connectivity is currently unclear, and the supporting evidence is indirect or simply absent. An understanding that nanomaterials can have enhanced properties when compared with the bulk form justifies an exploration of this connectivity, which needs to be considered for each nanomaterial as part of an informed, objective problem formulation. This important process is the phase we are currently in when it comes to nanotechnologies and their potential risks to the environment. What should follow are prioritized decisions about detailed exposure and hazard assessment needs and, when necessary, risk management. This is the way environmental regulators formulate, prioritize, quantify, and manage a range of environmental risks, and nanotechnologies should not be treated any differently.

It is important that we get this phase right because it will be critical in terms of defining the nature and extent of any eventual regulation and will form the basis for adopting or rejecting a precautionary approach for one or more nanomaterials.

The importance of investing time and effort in effective problem formulation and prioritization is highlighted by a defining feature of nanotechnologies: their diversity. The landscape of nanomaterials, products, and uses is already vast (5, 15), and it is likely to change continuously as innovation progresses (1). More types of substances may be manufactured as nanoparticles and new formulations (e.g., coatings; 12) of existing nanoparticles (e.g., nanoscale zerovalent iron coated to reduce sticking coefficients and thereby increase transport distances in aquifers as part of remediation applications; 16). Quantifying the hazard (e.g., through ecotoxicological testing) of every nanomaterial with varying size-distribution ranges will be both difficult and costly, notably if they incorporate varying levels of trace components (e.g., trace metals for carbon nanotubes) and differ in terms of surface functionalities, all of which may contribute to varying degrees of toxicity (12). Also, detecting and quantifying manufactured nanoparticles in complex environmental and biological media will be a significant analytical challenge, both for environmental monitoring and in terms of dosimetry within hazard studies.

This highlights the importance of prioritizing nanomaterials for quantitative environmental risk assessment on a case-by-case basis (12), particularly when no current legislative requirement exists (for example, if they are defined as existing substances). Life-cycle analysis may play an important role in this regard, for example in identifying those instances in which previously “fixed” manufactured nanoparticles are released into the environment through degradation and wear (4). Understanding sources to the environment, physicochemical properties, and the fundamental environmental behavior (and perhaps even the development of structure–activity models) may play important supporting roles in assessing the potential for environmental exposure to diverse nanomaterials as part of a risk-screening approach. If nothing else, this may help to define where uncertainties lie, targeting more detailed evidence needs.

One key requirement for confident environmental risk prioritization is an early consensus on nomenclature, not least because it is important in defining when legislative risk assessment will be required. In November 2006, the U.S. EPA ruled that nanosilver particles used in washing machines were subject to approval under the Federal Insecticide, Fungicide, and Rodenticide Act. Whether silver ions qualify as being nanomaterials at all is questionable, even though they may have been marketed as such. Whether or not nanosilver is a nanomaterial, this example further illustrates that when the necessary connectivity between sources, pathways, and receptors can be established as part of informed problem formulation, the need for detailed quantitative risk assessment is better justified. A second key requirement is therefore a framework with guiding principles to allow us to better formulate the problems for environmental risk assessment of diverse nanomaterials, one that reflects the unique challenges that such new technologies and novel materials pose.

In addition to problem formulation, further guiding principles are needed for the more detailed aspects of exposure and hazard assessment of nanomaterials. Such guidance is informed by evaluation and amendment where necessary of current (eco)toxicological methods and models for detailed exposure assessment, notably those that support important pieces of legislation (e.g., the Registration, Evaluation and Authorisation of Chemicals [REACH] legislation, which has defined risk assessment requirements for chemicals in the EU since June 2007). These appraisals need to be validated through case studies with model nanomaterials in production today and the data compared with those for well-characterized reference materials, the results being made freely and publicly available. Effective engagement with industry is critical to ensure the supply of high-quality characterization information on products. These are issues of harmonized guidance, quality assurance, and knowledge management and transfer that help build the foundations for consistent, comparable, and prioritized risk assessment.

Building the foundations for environmental risk assessment of nanomaterials

Many countries and international organizations have recently begun to commission significant research programs and cooperative initiatives in the broad area of environmental risk assessment of nanomaterials. Maynard (1) and Weisner et al. (12) provide useful analyses of funding in this area in the U.S. These programs are intended, over the next few years, to provide evidence that policy makers can use to draw preliminary conclusions about the significance of environmental risks of nanomaterials and formulate appropriate controls.

The year 2006 will be remembered as pivotal, and the initiatives and programs that emerged during the year reflect a three-pronged approach. The first component is the development of voluntary reporting schemes (or stewardship programs) for industry. The aim is to gather information from manufacturers, importers, and users of nanomaterials about the nature, properties, and uses of the materials, and, when provided, exposure and ecotoxicological information. In September 2006, the U.K. launched its voluntary reporting scheme (17), which also extends to research institutions and will run until 2008. Similar schemes have been developed or proposed by the U.S., Canada, and Australia (18).

The second component is various reviews of regulatory gaps to consider many aspects of horizontal and sector-specific legislation that might apply to nanomaterials. Of note, work is being done by the European Commission (EC), which is examining the issue of nanomaterials under REACH and the technical guidance document that supports it within two working groups (e.g., under the EC’s Scientific Committee for Emerging and Newly Identified Health Risks [SCENIHR]; 19). An evaluation of the technical guidance document was put out for consultation in April 2007.

The final component of the approach is the commissioning of basic programs of environmental nanosciences research to investigate generic aspects of life-cycle analysis, environmental fate and behavior, and ecotoxicology. Capacity building and encouraging interdisciplinary work are important objectives of these programs. To date, only a few perspectives have been heard, because the number of experienced ecotoxicologists and environmental fate and behavior scientists working on manufactured nanoparticles is limited. Their viewpoints are important for developing a balanced view of the significance of the issue. It is encouraging that work has begun in this area under the U.S. EPA National Center for Environmental Research (NCER) program (20), in the U.K. (21), and in Europe under the EC Sixth and Seventh Framework Programmes (22). In December 2006, the EC Seventh Framework Programme announced calls for new proposals under its nanosciences, nanotechnologies, materials, and new production technologies theme. In all of these cases, fundamental research into the environmental risks of nanomaterials is identified as a significant work area.

We are encouraged that the wider environmental scientific communities are now beginning to engage with the issue of manufactured nanoparticles, although it is only the beginning of this endeavor. This was a key feature of the first dedicated European conference on Environmental Effects of Nanomaterials and Nanoparticles, held in London in September 2006 (23). This conference brought scientists and policy makers from across the world together to share early data on fate, behavior, and ecotoxicology. The preliminary data presented showed the need for both acute and chronic sublethal effects to be examined.

An overall perspective from the conference can be summed up by a comment made by one highly experienced bench ecotoxicologist who has recently started to work in the field: “We want to do the studies, but we need the tools to do them properly.” The issue of dose–response, a central element to the assessment of hazard, is just one example of how important it is to get such enabling tools in place. From a regulatory viewpoint, important questions are whether dose–response relationships are influenced by particle size, shape, and number, and whether hazard assessments for bulk materials are sufficient for the same substance presented as a nanomaterial. For ecotoxicologists accustomed to working with chemicals (often in the dissolved phase for aquatic exposure), measuring and characterizing manufactured nanoparticles in complex environmental matrices to establish dose is proving to be problematic. Although techniques are available for measuring nanoparticles in test systems (e.g., field-flow fractionation), these need to be optimized and tested with manufactured nanoparticles. In addition, considerable work must be done to optimize them for use in the real environment (e.g., in aquatic systems) and to make them cost-effective.

Questions of dosimetry go beyond techniques to measure nanoparticles. They include what to measure in the first place (i.e., the dose metrics: should these be particle number, size, surface area, material concentration?), how to prepare and deliver nanomaterials for hazard studies, and how to assess the relative toxicity of the substance itself and when it is presented as a formulated product or preparation (e.g., understanding interaction of coatings with particle surfaces and how this influences toxicity). Preparation techniques already have been recognized as one factor influencing the ecotoxicity of manufactured nanoparticles (24, 25). Researchers need to understand which factors (e.g., pH, ionic strength, organic matter concentration, and microbiology) influence the behavior of nanoparticles in the environment (e.g., aggregation), so that these can be measured and controlled for in dose-response studies (13). For instance, a rich seam of information about the behavior of nonmanufactured aquatic nanoparticles could be mined. Unless these factors—and, where relevant, their measurement and reporting—are identified, data may be incomparable. This emphasizes the need for an interdisciplinary approach, bringing (e.g., in the aquatic realm) material scientists, colloid chemists, and ecotoxicologists together. It also emphasizes the need for editorial rigor in managing the peer review of submitted manuscripts, and even minimum acceptance criteria for publication—issues that have been faced before in the context of other new technologies, such as genomic microarrays.

Many countries, international organizations, and businesses active in the area of environment, health, and safety of nanomaterials are beginning to understand the importance of developing these enabling tools, which build solid foundations for robust and consistent risk assessment. Common themes addressing requirements to enable development are beginning to emerge from research strategies published during the past year in the U.S., U.K., and Europe (1, 2, 6, 26). This bodes well for international collaboration on the issue. These themes include the need for standardization and nomenclature. The International Organization for Standardization (ISO) and the European Committee for Standardization (CEN) are taking leading roles in the ISO TC229 and CEN TC352 technical committees (27) on these issues. These are supported by important work within the British Standards Institute (BSI) and the American Society for Testing and Materials (ASTM International), which have developed a publicly available vocabulary for nanoparticles (BSI PAS 71) and standard terminology for nanotechnology (ASTM E2456). Awareness is growing for the need to develop the infrastructure for characterization and metrology of nanomaterials in complex matrices and for reference materials—both for calibration of instruments used for assessing exposure and dosimetry—and for benchmarking ecotoxicological tests. The importance of evaluating the suitability of test methods is reflected by ongoing work in this area (18).

The Organisation for Economic Co-operation and Development (OECD) is playing an increasingly important role in the area of risk assessment of nanomaterials. In October 2006, the first meeting of the OECD working party on manufactured nanomaterials was convened to discuss potential study areas associated with environmental, health, and safety issues (28). The objective of the program, which will run until 2008, is to ensure that the approach to hazard, exposure, and risk assessment is of a high, science-based, and internationally harmonized standard and that international cooperation is promoted to meet this objective. Specific projects focus on information sharing, cooperation, and dissemination (including the need for a central database for research outputs); test methods; and risk assessment approaches. Safety testing of a representative set of nanomaterials will also be conducted as part of the program.

Problem formulation: a shared responsibility

The next 2 years will be important in terms of establishing the significance of environmental risks of manufactured nanoparticles and thereby the magnitude and nature of controls. The earliest view will be obtained through coordination of ongoing and forthcoming global programs and by ensuring that reliable, robust, and quality-assured data are collated and translated into information that can help researchers and policy makers formulate the problems and prioritize more confidently. Such research will not only support the prioritization of issues for further investigation but also give us insights about what effects we should look out for in the environment. Our understanding of ecotoxicological issues in the past highlights the importance of identifying the types of impacts we might foresee to target environmental surveillance that can act as a safety net for a tiered, risk-based approach.

Central to the success of this will be the establishment of a framework for problem formulation and prioritization with guiding principles, an activity that is best developed through an international, collaborative mechanism. We argue that it is worth investing time in this to focus where quantification of environmental risks should be undertaken. It needs to be supported by development of tools to ensure that robust and comparable data feed into this process. And of course, nanotechnologies are far more than just nanoparticles. Continuous innovation, with successive generations of nanotechnology development (1, 29), suggests the need to embed iterative problem formulation within risk assessment. This must be supported by horizon-scanning activities, technological road mapping, and effective mechanisms that provide important social and ethical contexts to frame the questions (30).

Indeed, continuous innovation of diverse nanomaterials is set within the broader context of ongoing innovation of technologies as a whole and the constant potential for emerging environmental risks of varying significance and with varying degrees of public concern. These potential risks are themselves set within the wider context of the many potential risks to the environment and the health of those who live in it. Should we be investing in problem formulation for environmental risk assessment of nanomaterials? Of course, understanding that the supporting research needs to be integrated and defined in its widest sense from natural to social and economic sciences. A framework with guiding principles can facilitate this and help us to prioritize more effectively and transparently. Should we be throwing extremely large sums of money at undertaking detailed environmental risk quantification for nanomaterials (or indeed any emerging technology), knowing that this will mean other potential risks may receive proportionately less consideration? Perhaps, but such arguments are surely better made if we can formulate the problems well and prioritize accordingly. Nanotechnologies are challenging us to define the key elements of this process, who needs to have a role, what their roles should be, and how governance is overseen (29). Clarity in this should help us manage the wider emergence of technologies in a democratic, inclusive, and evidence-based way. It should help those who wish to make a contribution to this process understand how they can be more effectively engaged (30). It might also serve to highlight a defining feature of problem formulation and risk prioritization. It is not a job for one institution, one business, one sector of the public, or one government. It is a shared responsibility.

Richard Owen is head of the Environment and Human Health Programme at the U.K. Environment Agency and a research fellow at the University of Plymouth (U.K.). He chairs the U.K. government’s Nanotechnologies Environmental Risk Assessment Task Force and is the program manager for the U.K. Environmental Nanoscience Initiative (www.nerc.ac.uk/research/programmes/nanoscience). Richard Handy is a reader in ecotoxicology and physiology at the University of Plymouth. He is a member of the U.K. government’s Nanotechnologies Environmental Risk Assessment Task Force and is the current president of the U.K. branch of the Society of Environmental Toxicology and Chemistry (SETAC). He was the organizer of the 2006 SETAC/Society for Experimental Biology Environmental Effects of Nanomaterials conference. Address correspondence about this article to Owen at richard.owen@environment-agency.gov.uk.

Disclaimer

The views expressed are those of the authors and do not necessarily represent those of the U.K. Environment Agency, nor are they a statement of U.K. government policy.

References

1. Maynard, A. D. Nanotechnology: A Research Strategy for Addressing Risk; Report PEN 3; Woodrow Wilson International Center for Scholars Project on Emerging Nanotechnologies: Washington, DC, July 2006; http://nanotechproject.org/reports.
2. Maynard, A. D.; et al. Safe Handling of Nanotechnology. Nature 2006, 444, 267–270.
3. U.K. Government. Guidelines for Environmental Risk Assessment and Management; www.defra.gov.uk/environment/risk/eramguide/index.htm.
4. Royal Society and Royal Academy of Engineering. Nanoscience and Nanotechnologies: Opportunities and Uncertainties, 2004; www.nanotec.org.uk/finalReport.htm.
5. Woodrow Wilson Institute Inventory on Consumer Nanoproducts; www.nanotechproject.org/inventories.

(6) U.K. Government. Characterizing the Potential Risks Posed by Engineered Nanoparticles: A First U.K. Government Research Report, 2005; www.defra.gov.uk/environment/nanotech/research/pdf/nanoparticles-riskreport.pdf.

6. Oberdorster, G.; Oberdorster, E.; Oberdorster, J. Nanotoxicology: An Emerging Discipline Evolving from Studies of Ultrafine Particles. Environ. Health Perspect. 2005, 113, 823–839.
7. Handy, R. D.; Shaw, B. J. Toxic Effects of Nanoparticles and Nanomaterials: Implications for Public Health, Risk Assessment, and the Public Perception of Nanotechnology. Health Risk Soc. 2007, 9, 125–144.
8. Sayes, C. M.; et al. The Differential Cytotoxicity of Water-Soluble Fullerenes. Nano Lett. 2004, 4, 1881–1887.
9. Lam, C.-W.; et al. Pulmonary Toxicity of Single-Wall Carbon Nanotubules in Mice 7 and 90 Days after Intratracheal Instillation. Toxicol. Sci. 2004, 77, 126–134.
10. Warheit, D. B.; et al. Comparative Pulmonary Toxicity Assessment of Single-Wall Carbon Nanotubes in Rats. Toxicol. Sci. 2004, 77, 117–125.
11. Weisner, M. R.; et al. Assessing the Risks of Manufactured Nanomaterials. Environ. Sci. Technol. 2006, 40, 4336–4345.
12. Fortner, J. D.; et al. C₆₀ in Water: Nanocrystal Formation and Microbial Response. Environ. Sci. Technol. 2005, 39, 4307–4316.
13. Lyon, D. Y.; et al. Bacterial Cell Association and Antimicrobial Activity of a C₆₀ Water Suspension. Environ. Toxicol. Chem. 2005, 24, 2757–2762.
14. Aitken, R. J.; et al. Manufacture and Use of Nanomaterials: Current Status in the U.K. and Global Trends. Occup. Med. 2006, 56, 300–306.
15. Tratnyek, P. G.; Johnson, R. L. Nanotechnologies for Environmental Cleanup. NanoToday 2006, 1 (2), 44–48.
16. U.K. Voluntary Reporting Scheme for Engineered Nanoscale Materials, 2006; www.defra.gov.uk/environment/nanotech/policy/index.htm.
17. Australian Department of Health and Aging, www.nicnas.gov.au/publications/chemical_gazette/pdf/2006feb.pdf.
18. SCENIHR. Opinion on the Appropriateness of Existing Methodologies To Assess the Potential Risks Associated with Engineered and Adventitious Products of Nanotechnologies. European Commission: Brussels, 2005; http://europa.eu.int/comm/health/ph_risk/committees/04_scenihr/scenihr_cons_01_en.htm.
19. NCER, http://es.epa.gov/ncer/nano/research/index.html.
20. U.K. Environmental Nanoscience Initiative, www.nerc.ac.uk/research/programmes/nanoscience.
21. EC, http://cordis.europa.eu/nanotechnology/home.html.
22. SETAC-UK, www.setac-uk.org.uk/nano.pdf.
23. Lovern, S. B.; Klaper, R. Daphnia magna Mortality When Exposed to Titanium Dioxide and Fullerene (C₆₀) Nanoparticles. Environ. Toxicol. Chem. 2006, 25, 1132–1137.
24. Fernandes, T. F.; et al. Ecotoxicology of Nanoparticles (NPs). SETAC Globe, September 2006.
25. U.S. National Science and Technology Council. Environment, Health and Safety Research Needs for Engineered Nanoscale Materials, 2006; www.nano.gov/NNI_EHS_research_needs.pdf.
26. ISO, www.iso.org.
27. OECD, www.oecd.org/env/nanosafety.
28. International Risk Governance Council. White Paper on Nanotechnology Risk Governance, 2006; www.irgc.org.
29. U.K. Environment Agency. A People’s Inquiry on Nanotechnology and the Environment 2006; www.environment-agency.gov.uk/science/.