Air quality expert group - Fine particulate matter (PM2.5) in ... - Defra

Chapter 5
Modelling PM2.5 and the future
5.1 Introduction
1. This chapter discusses modelling of PM2.5 in the UK, covering what models can
do and their limitations, the different model scales considered, how different
components of PM2.5 are modelled, how models are evaluated and what models
can tell us about future PM2.5 trends. A number of models are used in the UK
which may differ in basic methodology, model domain size, inputs and output
(i.e. level of chemical speciation, averaging times and spatial resolution). A nonexhaustive
survey of the approaches used to model PM2.5 in the UK is presented
in Table 5.1. A description of these models and some examples of how they
have been used is given in Annex A2. Outputs from these models are used in
the discussions throughout this chapter.
2. An important application of models is the synthesis of data from emission
inventories and observations. However, such modelling of PM2.5 is challenging
because of uncertainties in the measurement data (Chapter 2), limited trends
data and uncertainties in the different particulate matter (PM) components
(Chapter 3), coupled with uncertainties in the emission data and their
projections (Chapter 4).
5.2 What is modelling? What can models do?
Modelling PM2.5 and the future
3. Models for predicting concentrations of PM2.5 have a number of important
roles, some of which are complementary to measurement. These roles include
assessing concentrations at locations without monitors and answering questions
such as how will PM levels change in the future, what are the most important
emission sources to control to reach acceptable air quality and what balance
should be struck between policy actions within the UK and abroad. The models
encapsulate our current scientific understanding of the different physical and
chemical processes which determine the generation, transport and fate of small
particles in the atmosphere and therefore help understand how these different
processes interact and which of them are most important in different situations.
4. PM2.5 concentrations at a particular location are determined as a
superimposition of the processes taking place at a large range of spatial
scales from continental (e.g. long-range transport) to regional (e.g. secondary
particulate production), as well as urban and local scales (close to sources of
primary emissions). Because of this diversity of spatial ranges no single model
currently represents the full range of scales, and a hierarchy of models is used
for predictions of PM2.5 in the UK. These are typically regional (e.g. Europeanscale),
national (UK-scale) or local (e.g. urban-scale) models. These models
all have particular strengths and weaknesses and are suitable for different
applications.
123