AECOM's Nigel Pilkington and Jane Knowles discuss the probability and its role in determining impact significance
A key purpose of environmental impact assessment (EIA) is to assess the “likely significant effects” of a development on the environment, however, there is little explanation as to what this actually means.
The term implies, for example, there are also “unlikely significant effects”, as well as “likely and unlikely non-significant effects” and it could be interpreted as asking us to reflect our confidence in the prediction of significance.
This article focuses on what “likely” means and how it should, or should not, affect our assessment of the significance of effects.
Two aspects of likelihood are:
- The probability of an effect occurring – for example, an accident at a power plant could be catastrophic for the environment, but the probability of it occurring may be very small.
- Our confidence in judgments made of the sensitivity or value of the receptor, or whether there is a receptor at all – for example, excavation of a field is certain, but the potential for important buried archaeology to be present is low, therefore significant effects are unlikely.
Previous articles from EIA Quality Mark participants have presented useful discussions about taking a risk-based approach, mostly where there is uncertainty about the receptor. This article focuses on situations where the probability of a source or pathway existing in the first place is considered unlikely.
Schedule 3 of the EIA Regulations gives the first insight. When considering the potential for significant effects of a schedule 2 development the Regulations require the character of the development to be considered, including the “risk of accidents, having regard in particular to substances or technologies used” and the characteristics of the impact, including the “probability of the impact”.
Once it has been determined that a development does have the potential to cause significant effects and is an “EIA development”, schedule 4 of the Regulations deals with the required content of an environmental statement.
Point 1.3 requires a description of the “aspects of the environment likely to be significantly affected by the development”, and point 1.4 a description of the “likely significant effects of the development on the environment”.
Meanwhile, part 2.3 requires data “to identify and assess the main effects which the development is likely to have on the environment”.
Note that there is no longer any reference to probability; we now have the term “likely” included before the term “significant” and, importantly, separated from the main effects in point 2.3.
This could be interpreted as meaning that probability should not be a factor in deciding significance, but should be applied after significance is defined.
Given that the term “probability” only occurs in schedule 3, it may be that consideration of the probability of an effect occurring is deemed to be of greater importance at the screening and scoping stage.
This is highlighted again in part 4.13(6c), which states that in developing its scoping opinion the relevant planning authority must take into account “the environmental features likely to be affected by the development”.
If the probability of the effect occurring is very small, can we really say that a feature is likely to be affected?
The implication seems to be that once we are undertaking an EIA and presenting the environmental statement, we are dealing with those effects that are likely to happen.
Tests of the term “likely significant effects” in case law are limited. In R v Rochdale Metropolitan Borough Council, ex parte Milne, Sullivan J appears to define “likely” as a factor of probability.
“One is identifying and mitigating the ‘likely significant effects’, not every conceivable effect, however minor or unlikely,” he ruled, which is interesting, in that he again appears to separate probability from significance:
It could be argued that while the EIA Directive is intended to deal with the likely effects of a development, other directives, such as SEVESO II (2003/105/EC), are intended to deal with the potential environmental consequences of major accidents. However an AC-IMPEL publication of 2001 concluded: “The scope of the information required under the EIA procedure is the widest. It should therefore be the starting point for information for all (and frequently subsequent) procedures.”
Uncertainty surrounding how probability should influence screening, scoping and EIA reporting is evident when reviewing environmental statements.
There are a wide range of responses to dealing with the consequences of low-probability effects. Some completely avoid the issue of accidents/non-standard operating conditions, while others consider them only in terms of human safety effects and few take a full environmental risk-assessment approach.
Perhaps the best option may be to separate out effects that could result from an accident, or non-standard operating conditions, into a stand-alone chapter of the environmental statement, or to reference other documents where such information can be found.
Consideration also needs to be given to keeping source and pathway probability separate from significance. For example an effect could be described as significant, as the magnitude is large and the sensitivity of the receptor is high, but it is not likely, as the probability of occurrence is extremely low.
Given the wide range of approaches being adopted, perhaps it is time for further debate on the role of probability in EIA, and whether it can, and should, help determine significance, or should be applied separately.
This article was written as a contribution to the EIA Quality Mark’s commitment to improving EIA practice.
Nigel Pilkington is a regional director, environment and Jane Knowles is a principal consultant at AECOM.
