Towards linking environmental law and science

We use the term “actionable science” to refer to tools, data, or analyses that can inform or guide decisions in the management of natural resources (Palmer 2012; Beier et al. 2017). We note that the term “actionable” in legal circles is used to describe something that can lead to legal action, such as slander leading to a lawsuit—this is not how we use the term.

Actionable science often targets a specific knowledge gap in a specific environmental decision-making context (i.e., policy window (Rose et al. 2017)). For example, during the environmental assessment of a large energy industrial development in the Skeena River estuary (British Columbia, Canada), researchers discovered that the project was proposed for a location that was particularly important for young migratory salmon and that risks of the project were under-assessed (Carr-Harris et al. 2015; Moore et al. 2015, 2016). This science was submitted as evidence by First Nations’ legal teams during environmental assessment comment periods and consultative processes and likely contributed to dialogue, negotiations, and project design modifications. Nevertheless, the Canadian Environmental Assessment Agency (CEAA) proceeded to grant the project approval, highlighting that while actionable science can play a role in environmental decision-making, the extent to which it will do so also depends on the relevant legislative and political context. As further discussed below, very few Canadian environment-related laws set out a scientific standard to guide government decision-making.

Actionable research may be best enabled through co-development of research programs with diverse communities and collaborators with close attention to emerging policy windows (Adams et al. 2014; Moon et al. 2014; Rose 2014; Beier et al. 2017; Keeler et al. 2017; Rose et al. 2017) rather than the classic approach of testing scientific theory that is often taught in university (Palmer et al. 2005; Keeler et al. 2017). Collaboration among scientists and lawyers, government agencies, Indigenous peoples, and stakeholders can enable the early identification of specific data gaps or uncertainties that will be relevant to environmental policy windows (Sutherland et al. 2012; Adams et al. 2014; Cook et al. 2014). For example, science and Haida laws brought attention to flaws in the federal government’s management of herring. This resulted in an injunction to prevent re-opening the fishery to continue the process of rebuilding fish stocks in Haida Gwaii (Council of the Haida Nation et al. v Minister of Fisheries and Oceans 2015).

Governments also rely on science to inform the formulation of standards and guidelines that are often incorporated in environmental laws and policies, such as fish stock assessment or air and water quality guidelines (Tollefson et al. 2009). In contrast to the actionable science described above, the timing windows for such processes are generally longer and consequently more in line with traditional research programs. As a recent example, Environment and Climate Change Canada announced its intention to draft regulations for coal mining effluent under the federal Fisheries Act, with a stakeholder consultation phase lasting well over a year (Environment and Climate Change Canada 2017).

Establishing the scientific foundation for guidelines or standards can be challenging for several reasons. “In a perfect world regulatory thresholds would correspond to clear ecological thresholds, but in practice, this is difficult to achieve because ecosystems are highly variable” (Hunter et al. 2009, p. 3). In fact, ecological thresholds may not actually exist for some processes or species (Huggett 2005; Reckhow 1994). Further, guidelines or standards are often set through an uneasy mixture of science and values. For example, the regulation of toxic substances is often based on “dose–response” relationships that must be translated to regulatory thresholds based on societal value judgments with respect to acceptable levels of risk. Although it may be challenging, scientists can help lawyers, legislators, and regulators to understand the limitations of science with respect to relevant ecological thresholds and objectives, while still emphasizing the strength of existing evidence (Tear et al. 2005; Hunter et al. 2009).

Commentary pieces written for widely read media outlets can identify issues with environmental law or decision-making and provide constructive solutions. These opinion editorials (op-eds) can communicate to a broad audience, generate further media attention, and bring lawmakers’ and regulators’ attention to an issue and the potential solution(s). For example, Otto et al. (2013) highlighted the lack of action for species at risk in Canada. Op-eds such as this can be published rapidly and be strategically aligned with key moments in the legislative cycle (Fig. 1).

Scientists and law professors in Canada have used open letters signed by large groups of experts to show consensus from the scientific and environmental law community on evidence or law (where it exists). Similar to op-eds, these letters can target different portions of the policy and (or) legislative cycle; open letters have critiqued the scientific basis for the regulatory approval of particular projects such as the Site C dam (Schindler et al. 2015) and the Enbridge Northern Gateway pipeline (Chan et al. 2014), criticized pending amendments to core environmental laws such as the Fisheries Act (Schindler et al. 2012), and offered suggestions for environmental policy such as the Canadian Environmental Assessment Act (Jacob et al. 2016). These letters can garner substantial media attention and exert political pressure on decision-makers as well as enable future engagement.

Scientists can play key roles in quantifying the environmental consequences of management decisions. Although some monitoring may occur internally by industry proponents as part of regulatory compliance, this may be vulnerable to conflicts of interest (Casselman 2015) and government agencies with mandates to conduct monitoring may be limited by resources, capacity, or political pressure. Monitoring by independent scientists from academia or NGOs can provide independent validation of this regulatory monitoring and help fill knowledge gaps or oversights. For example, independent science has repeatedly found deficiencies in the environmental performance of oil sands developments (Kelly et al. 2010; Rooney et al. 2012). One study discovered that emissions of volatile organic compounds were two to four-and-a-half times higher and contained between nine and 53 more types of reportable compounds than self-reported measurements by oil sands facilities (Li et al. 2017). Although rapidly publishing the findings of monitoring is a key step, effective science communication can increase the impact of the research findings. For example, David Schindler brought several deformed whitefish to a press conference as a dramatic illustration of the risks of contamination from oil sands developments to aquatic ecosystems (Schindler 2010). Robust and independent science can evaluate the consequences of different policy and regulatory interventions. Impactful science communication can increase awareness of this science and the pressure for those insights to be acted upon.

There are opportunities for lawyers and scientists to examine the effectiveness of environmental and resource management laws and policies. For example, two of the coauthors of this paper with law and science backgrounds collaborated on an empirical assessment of DFO’s fish habitat management regime (Favaro and Olszynski 2017). They found that the regime was likely resulting in a net loss of habitat—a contravention of the department’s “no net loss” policy, a central feature of the fish habitat legal regime in Canada. Other retrospective analyses have examined the (lack of a) regulatory burden of the Fisheries Act (Favaro et al. 2012), how economics overrides listing threatened species (Schultz et al. 2013), how the composition of recovery teams affects the identification of critical habitat for species at risk species (Taylor and Pinkus 2013), and the inadequacy of habitat mitigation efforts (Harper and Quigley 2005; Quigley and Harper 2006). Retrospective analyses can be a powerful way to reveal if there are systematic weaknesses in government decision-making, implementation, or policy and can be a driver for reform.

Scientists and lawyers can collaborate to address potential empirical disconnects between science and environmental regulations, laws, and policies. The laws that govern decision-making may not effectively incorporate scientific understanding. For example, the Canadian federal Oceans Act authorizes the designation of areas of the sea for “special protection” as marine protected areas, yet that law does not reflect widespread findings in the scientific literature regarding which human activities are inconsistent with the purposes of these designated areas (Jamieson and Levings 2001; Hutchings et al. 2012; Vanderzwaag et al. 2012; Devillers et al. 2015; Bailey et al. 2016). Indigenous and environmental lawyers and scientists are currently collaborating on briefs and testimony to a Parliamentary Committee studying the Oceans Act’s Marine Protected Areas regime, making the case for amendments that would reflect scientific guidance on effective marine protected areas (Nowlan and Watson 2017). As another example, during the Parliamentary review of the federal Fisheries Act, environmental lawyers cited research on fish habitat impacts in the Skeena estuary from proposed energy development projects to bolster their arguments for the need for stronger habitat protection provisions in the law (Carr-Harris et al. 2015; Moore et al. 2016; Nowlan 2016). As these examples illustrate, legal researchers and practitioners can connect with scientists working in their area of interest for collaboration, and collectively they can communicate needs to improve and strengthen environmental law.

Before a law is revised, there are opportunities to quantify how legal changes might influence environmental protection if the law is passed. For example, the sulphur emission controls imposed in the 1980s that included a cap on sulphur dioxide (SO₂) and eventually led to the 1991 Canada–US Air Quality Agreement (Environment and Climate Change Canada 2013) were influenced by scientific research showing reduced lake acidity following large reductions in sulphur emissions from Sudbury area smelters (Keller et al. 1998). When a legal change is proposed, scientists and legal experts can provide prospective feedback. This feedback might not reach an open policy window, but it may contribute to policy when the window subsequently opens. For example, scientific studies on the likelihood of Atlantic cod stock recovery (Neubauer et al. 2013; Rose and Rowe 2015) may influence future decisions regarding re-opening the cod commercial fishery, which has been closed since the 1990s.

Multiple scientific activities across the legislative and policy cycle can build towards substantive impact in linking science and law (Fig. 1, Table 1). Initial engagement can foster future engagement and build towards policy impact. Initial efforts may also foster additional follow-up opportunities such as testifying as a witness in court or in a parliamentary hearing, providing written briefs based on research results to policy-makers, meeting with elected officials, or guiding public interest environmental law organizations, or as an informal advisor (Meyer et al. 2010).

Environmental decision-making by governments can happen more quickly than scientists may expect, and opportunities to participate often appear with little or no warning (Cook et al. 2014; Rose et al. 2017). If scientists want to contribute, they need to respond within legislated time limits. As one example, under the current Canadian Environmental Assessment Act, 2012, members of the public (including the scientific community) only have 20 d to comment on whether an environmental assessment of a project should be conducted and may be asked to submit evidence for an assessment within relatively short time windows (e.g., 30 d). As another example, when project proponents apply for an authorization to destroy fish habitat pursuant to section 35 of the federal Fisheries Act, the applicable regulations require the relevant department (DFO) to issue authorizations within 90 d of the application being deemed complete. A further difficulty is that the government is not required to give public notice of these applications, and there is presently no public registry for such authorizations.

Such time windows are much shorter than those typical in the scientific process. Scientific research can take years or even decades, particularly for ecological studies conducted at large geographic scales. Data analyses can take weeks to months to years. The peer-review process may take months to years. Thus, it may be years from the time that a scientist seeks to answer a question to the time when they are “ready” to share results. By this time, it is likely that the window for contributing to a particular environmental decision point has closed (though the results can be shared with environmental lawyers to bolster future law reform efforts).

We offer scientists three suggestions for closing this time disconnect:

One of the key challenges in the interface between science and law is how to understand and clearly communicate certainty and uncertainty (Sutherland et al. 2013). Scientists need to learn how to express themselves in ways that are powerful and accessible as well as honest with regards to certainty. For instance, effective science communication with decision-makers entails leading with what is known rather than leading what is not known. This is counter to many scientists’ inclinations and training.

The legal standard of proof varies according to both the type of law (criminal vs. tort vs. judicial review of government decisions vs. aboriginal law) and the arena of law (courts vs. legislatures vs. government agencies and departments), but it is fundamentally different than the threshold that generally underpins scientific hypothesis testing (e.g., p < 0.05). In negligence cases, for example, the legal standard of proof is a “balance of probabilities” or about 51%. In the criminal law context, the government must prove its case “beyond a reasonable doubt”, which is widely understood as a higher standard (S.C.R. 2001). While there does not appear to be any uniform standard of proof in the regulatory context (e.g., in hearings before a CEAA panel or in decisions by DFO), when challenging such decisions the impugned analysis needs to be considered “unreasonable” for the decision to be set aside. This more qualitative standard could be considered higher than a balance of probabilities but further precision is difficult and can even be misleading. Canadian courts have also stated that it is not their role in this context to arbitrate conflicting scientific predictions, an approach criticized in the environmental law scholarship (Olszynski 2015). For the present, it appears that alleged errors or deficiencies that may appear minor to a court will not be sufficient to successfully challenge the decision or assessment that relies on (potentially flawed) science.

Scientists can solicit guidance from legal experts to become familiar with how different legal settings demand different levels of certainty. Scientists should also become familiar with the role of science and its certainty in Crown consultation and accommodation of Indigenous peoples’ concerns in decisions or assessments.

To help translate scientific certainty to decision-makers, advisory bodies such as the Intergovernmental Panel on Climate Change have used scales that link Bayesian-based probabilities to legally equivalent phrases such as “beyond a reasonable doubt” or “reasonable grounds for belief” (Weiss 2003). Traditional frequentist statistics present the probability that a null hypothesis occurs due to chance, which is a somewhat convoluted output that will demand careful and clear explanation (e.g., “There is an X% change that this pattern occurred by chance alone.”). The use of 95% confidence intervals of effect sizes will likely be more effective than using p-values alone (e.g., “We found that this chemical will increase cancer rates by X- to Y-fold.”). Bayesian analytical approaches may provide scientists with estimates of probability (e.g., “There is X% chance that this is true given the data.”) that are more easily communicated within law and policy realms (Ellison 1996). Further, emerging scientific approaches whose potential weaknesses have not yet been fully quantified by the scientific community may require careful assessment regarding the weight of evidence that these new technologies or analyses can provide. Regardless of their specific statistical approach, being comfortable with translating statistical outputs into understandable and compelling language that accurately captures scientific certainty and uncertainty is a key aspect of effectively linking science to law. Often the deciding factor for legal decision-makers (particularly judges) is the professional judgment of a credible scientific expert.

The law–science interface demands effective scientific communication, as many have noted in papers and guidebooks (Owens et al. 2006; Baron 2010; Cornell et al. 2013; Smith et al. 2013; Cooke et al. 2017; Dunn and Laing 2017), though progress on improving scientific communication and engagement with nonscientists arguably remains mixed. With a few exceptions, the lawyers, judges, and elected officials who make and interpret environmental laws are not scientists and may not work to stay abreast of relevant scientific literature and concepts. Thus, the take-home messages from results and conclusions of scientific research need to be clear, strong, and free of jargon. Scientific communication may be strengthened if integrated with a personal story that resonates. While scientific training encourages scientists to focus on what is not known, communications training can emphasize on what is known, and where the weight of evidence lies. Furthermore, impactful scientific communication will frame results with the appropriate policy window or law (Owens et al. 2006; Rose 2014). This proper framing necessitates a nuanced understanding of how scientific evidence relates specifically to legal issues. Further, a fundamental question in science communication is identifying the research audience: who is the target of the communication, what do they already know, and what do they care about?

Canadian environmental law is complex and involves multiple jurisdictional levels. It encompasses the common law (essentially judge-made law inherited from Britain and modified over time, e.g., negligence law) as well as legislation passed by federal, provincial, territorial, and municipal governments. Indigenous peoples have been governing their territories using their own laws since time immemorial. Indigenous laws are independent of the Canadian state and also govern and regulate land, water, and marine use (Borrows 2002; Clogg et al. 2016). Aboriginal law refers to the law of the Canadian state relating to Indigenous peoples, whereas Indigenous law refers to the legal traditions of Indigenous nations. International laws may also be involved.

If scientists engaged in actionable science or other law–science activities want to see their research applied to legal reforms (Table 1), it can be helpful to be aware of which level of government regulates the subject of their research and which legal processes are involved. Without knowledge and understanding of the multiple points of entry that may occur at different levels of government, the science may stay in the “ivory tower” (Baron 2010). While it is likely not feasible for scientists to become experts in all areas of law, talking to environmental law experts and NGOs about research can elevate their understanding of the legal context and opportunities for influence. In addition, if institutions want to empower the next generation of scientists to contribute to society, then they could offer classes in environmental policy and law (Keeler et al. 2017), including Indigenous laws (Finch 2012).

Indigenous laws often inform and parallel modern scientific and sustainability principles such as interconnectedness, balance, stewardship, respect, reciprocity, and responsibility. Many scholars have advocated for not only the reinvigoration and incorporation of Indigenous laws into decision-making and Canada’s legal system (Borrows 2002, 2010; Christie 2007; Napoleon 2015) but also the reconciliation of Canada’s legal system to pre-existing Indigenous laws and a deeper understanding of “the colonial enterprise and injustices it has so often created” (Finch 2012, p. 2.1.8). This reconciliation requires an understanding of Indigenous laws, which in turn requires a duty to learn on the part of all law practitioners (Finch 2012) and scientists. To do so will provide an opportunity to learn from Indigenous peoples’ knowledge and laws governing millennia of successful land, water, and marine use and governance.

Lawyers are bound by professional ethics to a number of parties—to the state, to courts and tribunals, to other lawyers, and to themselves. Foremost, however, lawyers are ethically bound to their clients to obtain the benefit of any and every remedy and defense that is authorized by law (National Judicial Institute 2013). In other words, they are advocates. Scientists, on the other hand, are supposed to assess evidence as objectively as possible. Thus, there may be an ethical disconnect between scientists and lawyers. Scientists risk their reputations and careers if they are perceived as lacking objectivity or being a “hired gun”. Lawyers should recognize that scientists must meet their own expectations of scientific integrity in all evidence that they produce (Doremus 2008), and scientists should similarly be aware of lawyers’ ethical obligations. Scientists and environmental lawyers should also be aware of moral and ethical obligations with respect to Indigenous peoples and Indigenous legal rights and title such as around the potential confidentiality of traditional knowledge.

Many environmental law concepts and principles, such as the precautionary principle or adaptive management, have both scientific and normative, or value-laden, aspects. In such instances it may be preferable for scientists to restrict themselves to the scientific aspects and allow lawyers (or others) to address the more value-laden aspects (e.g., what level of risk is society willing to tolerate?).

Scientists may be called to testify in environmental law disputes that end up in court. To be “court ready”, lawyers advise scientists to keep organized and consistent paper trails and save relevant emails. Scientists should also keep in mind that written correspondence is often subject to disclosure rules (e.g., in the context of actual litigation) and access to information legislation and could be used in court. While these differences may seem challenging, it is important to note that they have successfully been addressed in other sectors, such as linkages between forensic sciences and law enforcement.