Assessing the Effectiveness of Interventions on Air Quality

P. S. Monks, James D. Allan, David Carruthers, D.C. Carslaw, Gary W Fuller, Roy M. Harrison, Mathew R Heal, A. C. Lewis, Eiko Nemitz, Claire Reeves, Martin Williams, David Fowler, Ben B. Marner, Andrew Williams, Sarah Moller, Richard Maggs, Tim Murrells, Paul Quincey, Paul Willis

Research output: Book/ReportOther report

Abstract / Description of output

In the air quality context, interventions cover a wide range of actions from ‘deliberate’ measures to reduce air pollution to those primarily aimed at other outcomes, but which can indirectly affect air pollution. Air quality interventions span a wide range of situations, spatial scales from (e.g. a single road through to continental scales) and temporal scales from the short-term closure of a road to decadal (or longer) changes. The principal focus of this report is on local scale interventions, such as those that might reasonably be considered by local authorities.
Understanding the impact interventions have on air quality is highly desirable because of the need to quantify the outcome on air quality and health i.e. relate a policy aimed at improving air quality to a robust understanding of the outcome.
The assessment of interventions can be challenging for several reasons. These challenges include the common situation where interventions rarely occur in isolation from other changes that affect air quality, and the difficulty in detecting and quantifying changes if the interventions are small. Indeed, not every intervention is detectable in terms of quantifying changes in pollutant concentrations or health outcomes, even using sophisticated analysis techniques.
The work in the report suggests the ‘accountability chain’ model provides a useful way to assess the impact of an intervention from a change in activity through to potential beneficial health effects i.e. activity -> emissions -> concentrations -> health outcomes.
From an analysis of a wide range of case studies the following needs have been identified:
1. The availability of appropriate activity data that would be expected to change because of an intervention.
2. The availability of well-sited ambient air quality measurement data increases the potential for data analysis and the application of statistical methods to quantify changes to the concentrations of pollutants.
3. Air quality models can have an important role to play. Although most intervention analyses considered in this report do not use air quality models, they provide the means of assessing the likely impact that interventions have on air quality and allow scenarios to be considered.
4. When interventions are assessed based on changes in concentrations of pollutants, it is important to take account of meteorological variation. Meteorological variation can easily mask or emphasise changes in concentrations resulting from changes in emissions. Methods exist to account for or ‘remove’ the influence of meteorology.
The report gives clear recommendations that detail the need for pragmatic advice for practitioners, the assessment of causality against confounding factors and the counter-factual, enhanced statistical approaches and future siting criteria for air quality monitoring.
The following recommendations have been identified.
- The design of the assessment of an intervention should be considered at the planning stage, with the practitioners receiving pragmatic advice on the process. To date, the information relating to the analysis of interventions often resides in academic journals and is not easily accessible by most practitioners.
- Many areas of policy should seek to incorporate the approaches and thinking of intervention analysis, e.g. in determining the efficacy or otherwise of technologies that offer potential reductions to emissions of different pollutants.
- An analysis of the intermediate steps which the intervention is designed to influence should be carried out, to assess the causality of the relationship between the intervention and any measured concentration changes. For example, if the intervention is to change a pattern of fuel use, has fuel consumption changed as intended? Furthermore, there is also a need to take account of other changes that may also have affected the assessment of an intervention and to develop an understanding of whether any changes in concentrations specifically resulting from an intervention can reasonably be quantified.
- Results from the implementation of local plans to mitigate air pollution should be pooled to derive a statistically more robust overall assessment of local measures.
- Where new air quality monitoring sites are planned, their relevance for assessing interventions should be considered.
- Intervention analysis is an active area of research that continues to evolve and be refined. Defra should retain a watching brief in this area to understand new developments and promote good practice.

Original languageEnglish
Place of PublicationLondon
PublisherAir Quality Expert Group
Number of pages70
Publication statusPublished - 24 Jun 2020

Keywords / Materials (for Non-textual outputs)

  • Air quality assessment
  • Air quality policy
  • air quality measurement
  • Air quality modelling

Fingerprint

Dive into the research topics of 'Assessing the Effectiveness of Interventions on Air Quality'. Together they form a unique fingerprint.

Cite this