Environmental assessment

Dr. Lise Fortin is an aquatic ecologist who has spent much of her career studying fish around power station cooling water intakes. At Meridian Nuclear Engineering she leads environmental assessment work, and her office is full of maps. "Engineers think the environment is what is left over after the plant is designed," she tells Hana, spreading out a chart of the shoreline near Darlington. "For an assessment it is the other way round. You start with what is there, the fish, the wetlands, the people who fish and hunt here, and you ask what the project will change."

Purpose of environmental assessment

Environmental assessment is a process for predicting the effects of a proposed project on the environment before decisions are made, so that harmful effects can be avoided, reduced or offset and decision-makers and the public know what they are accepting. It looks at the whole project: site preparation, construction, operation, decommissioning and accidents and malfunctions. It considers the biophysical environment, such as air, water, soil, fish and wildlife, and the human environment, including health, the economy, land use, culture and heritage. Modern assessments also consider effects on the rights of Indigenous peoples, on climate and on sustainability.

For a nuclear project, some effects are similar to those of any large industrial project: land clearing, noise, traffic, construction workforce, water intake and thermal discharge. Others are specific to nuclear: routine releases of radionuclides to air and water, doses to the public and to non-human biota, radioactive waste, and the consequences of accidents. Assessments for nuclear projects therefore draw on radiation protection and safety analysis as well as on ecology and social science.

The federal framework

In Canada, the Impact Assessment Act came into force in 2019, replacing the Canadian Environmental Assessment Act, 2012. It applies to designated projects listed in the Physical Activities Regulations, which include most new nuclear fission or fusion reactors, although some smaller reactors on existing licensed nuclear sites are not designated. The Impact Assessment Agency of Canada leads the assessment, and for nuclear projects it works with the CNSC through an integrated review panel so that the impact assessment and the licensing review proceed together. In 2023 the Supreme Court of Canada found parts of the Act to be beyond federal jurisdiction, and Parliament amended it in 2024 to focus federal assessment more closely on effects within federal jurisdiction.

Projects that are not designated under the Impact Assessment Act are still subject to environmental review by the CNSC under the Nuclear Safety and Control Act. Every licence application must show that the applicant will make adequate provision for the protection of the environment, and the CNSC carries out an environmental protection review proportionate to the risk. REGDOC-2.9.1 sets out the CNSC's requirements for environmental protection, including environmental risk assessment, effluent and environmental monitoring and environmental management systems.

The steps of an assessment

Although the legal details differ, assessments in both countries follow similar steps. Scoping decides which components of the environment and which effects the assessment will cover, based on the project description and on public and Indigenous input. Baseline studies describe the existing environment, often over several years to capture natural variation. Effects analysis predicts how the project will change each valued component, using models of dispersion, dose, habitat loss, noise and so on. Mitigation measures are proposed for adverse effects. Residual effects, those remaining after mitigation, are assessed for significance. Cumulative effects, the combined effects of the project with other past, present and foreseeable projects, are considered. Finally, a follow-up programme is designed to check whether the predictions were right and whether mitigation works.

Judging significance

The central judgement in most assessments is whether residual effects are significant. Assessments usually describe each effect by several criteria: its direction, adverse or beneficial; its magnitude, from negligible to high; its geographic extent, from the site to the region; its duration, from short-term to permanent; its frequency; its reversibility; and the ecological or social context, such as whether the affected species is at risk. These criteria are combined, sometimes in a matrix, to reach a conclusion. The combination always involves judgement, and good assessments explain their reasoning rather than hiding it in a formula.

# Toy significance matrix for residual effects (teaching scheme, scores 1-3 per criterion).
CRIT = ("magnitude", "extent", "duration", "reversibility", "context")
effects = {
    "construction noise on residents": (2, 1, 1, 1, 1),
    "loss of shoreline wetland":       (2, 1, 3, 2, 2),
    "thermal plume on fish":           (3, 2, 3, 2, 3),
    "routine radiological dose":       (1, 2, 3, 1, 1),
}
THRESHOLD = 12
for name, s in sorted(effects.items(), key=lambda kv: -sum(kv[1])):
    total = sum(s)
    flag = "significant" if total >= THRESHOLD or (s[0] == 3 and s[4] == 3) else "not significant"
    detail = ", ".join(f"{c} {v}" for c, v in zip(CRIT, s))
    print(f"{name:34s} score {total:2d}  {flag:16s} ({detail})")
print("scores frame the discussion; the conclusion must be explained in words")

thermal plume on fish              score 13  significant      (magnitude 3, extent 2, duration 3, reversibility 2, context 3)
loss of shoreline wetland          score 10  not significant  (magnitude 2, extent 1, duration 3, reversibility 2, context 2)
routine radiological dose          score  8  not significant  (magnitude 1, extent 2, duration 3, reversibility 1, context 1)
construction noise on residents    score  6  not significant  (magnitude 2, extent 1, duration 1, reversibility 1, context 1)
scores frame the discussion; the conclusion must be explained in words

The output ranks the thermal plume's effect on fish as the only effect that crosses the significance threshold in this toy scheme, because it is high in magnitude, long-lasting and affects a sensitive species. Lise points out to Hana that the score is a starting point for discussion. In a real assessment the team would examine whether a different intake design, such as an offshore intake with low velocity, could reduce the effect, and would set up a monitoring programme to check the prediction.

Radiological effects on people and biota

For routine operation, the radiological part of an assessment estimates doses to the most exposed members of the public, usually a hypothetical person or group living near the site and eating local food and fish. The calculation follows releases through air and water pathways to the dose received. For a well-run reactor these doses are a small fraction of the public dose limit of 1 millisievert per year, and much smaller than doses from natural background radiation. Assessments also estimate dose rates to non-human biota, such as fish, birds and small mammals, and compare them with screening benchmarks. The assessment of accidents uses the safety analysis to estimate doses for a range of accident scenarios and to show that emergency plans can protect the public.

Assessment and the licensing decision

An impact assessment and a licence are separate decisions. Under the Impact Assessment Act, the Minister of the Environment decides, based on the assessment report, whether the adverse effects within federal jurisdiction are in the public interest, and may set conditions. Only then can the CNSC issue a licence. The Commission's licensing decision draws on the assessment but also on its own review of safety. For the Darlington new nuclear project, a joint review panel under the 1992 Act reported in 2011 and the Commission issued the site preparation licence in 2012; the assessment's bounding approach later allowed a small modular reactor to be licensed on the same site.

Cooling water: the largest ecological question

For most nuclear stations on lakes, rivers and coasts, the largest ecological effect is not radiological but the use of cooling water. A large once-through cooled station draws tens of cubic metres of water per second for each unit. Fish and other organisms can be pinned against the intake screens, called impingement, or drawn through the screens and the condenser, called entrainment, where eggs and larvae are killed by heat, pressure changes and mechanical stress. The heated water returned to the lake forms a thermal plume that can change the habitat for fish that are sensitive to temperature, particularly cold-water species and their eggs.

Canadian stations have reduced these effects in several ways. At Pickering, a barrier net in the lake in front of the intake keeps many fish away from the screens. Assessments for new units consider intake designs that draw water from deeper offshore locations at low velocity, diffusers that spread the warm discharge, and offsetting measures under the Fisheries Act, such as creating or restoring habitat elsewhere when some loss cannot be avoided. Fisheries and Oceans Canada reviews these measures and may issue an authorization with conditions.

Lise has a habit of bringing Hana back to specifics. "Ask which fish, which life stage and which month," she says. "A thermal plume that is harmless in July can be harmful in February if it draws fish to warm water and then the unit trips, dropping the temperature suddenly. That kind of cold shock has killed fish at plants before. You only find these effects if you know the biology."

Indigenous knowledge in assessment

Modern assessments draw on Indigenous knowledge as well as scientific studies. Indigenous Nations and communities often hold detailed knowledge of the land, water and species around a proposed site, built up over many generations, including where species spawn, how their numbers have changed and which places are culturally important. This knowledge can change what an assessment looks at and what it concludes. It is shared on the community's terms, often under agreements that protect confidential information such as the locations of sacred sites. Lesson 6 discusses engagement in more detail.

Accidents and malfunctions

An assessment must consider the effects of accidents and malfunctions, not only of routine operation. For a nuclear project this means using the results of the safety analysis to estimate the consequences of a range of accident scenarios, from minor spills to severe accidents with off-site releases, and describing how emergency plans would protect the public. The assessment does not repeat the safety analysis, but it translates it into effects on the environment and people. For a severe accident, the assessment may consider effects on land use, agriculture and fisheries, and the need for long-term protective actions, even though the probability is very low.

Public concern about nuclear accidents is often greater than concern about routine effects, and assessments that treat accidents too briefly tend to lose public trust. Good practice is to describe the accident scenarios clearly, explain how their frequencies were estimated, show the range of consequences, and be open about uncertainties.

Follow-up and adaptive management

Predictions in an assessment are uncertain, and a follow-up programme checks them. The programme specifies what will be monitored, where, how often and for how long, what results would show that a prediction was wrong, and what action would follow. Monitoring results are reported to the regulator and usually to the public. If effects turn out larger than predicted, mitigation is adjusted, a practice called adaptive management. Follow-up also builds knowledge for future assessments: the measured effects of one project help predict the effects of the next.

CNSC licences include conditions requiring environmental monitoring programmes, and the CNSC runs its own Independent Environmental Monitoring Program, in which staff collect samples of air, water, soil, vegetation and food around nuclear facilities and publish the results. The results to date have generally shown radiation levels near facilities well below levels that would affect health.

At the end of the week Lise gives Hana a task: read the follow-up reports for a station's first ten years and compare the measured effects with the original predictions. Hana finds that most predictions were close, that the thermal plume was somewhat larger than predicted in some winters, and that the monitoring had led to a change in how the station managed discharges during cold periods. "That is what follow-up is for," Lise says.