Embed Size (px)
Citation preview
Published Project Report PPR601
A transport emissions database for Hillingdon
K Turpin, A Savage and P G Boulter
© Transport Research Laboratory 2012
Transport Research Laboratory Creating the future of transport
PUBLISHED PROJECT REPORT PPR601
A transport emissions database for Hillingdon
K Turpin, A Savage, P G Boulter
Prepared for: London Borough of Hillingdon,
Project Ref: BCVB/ENV/1
Quality approved:
Anna Savage
(Project Manager)
Alaric Lester
(Technical Referee)
Disclaimer
This report has been produced by the Transport Research Laboratory under a contract
with London Borough of Hillingdon. Any views expressed in this report are not
necessarily those of London Borough of Hillingdon.
The information contained herein is the property of TRL Limited and does not necessarily
reflect the views or policies of the customer for whom this report was prepared. Whilst
every effort has been made to ensure that the matter presented in this report is
relevant, accurate and up-to-date, TRL Limited cannot accept any liability for any error
or omission, or reliance on part or all of the content in another context.
When purchased in hard copy, this publication is printed on paper that is FSC (Forest
Stewardship Council) and TCF (Totally Chlorine Free) registered.
Contents amendment record
This report has been amended and issued as follows:
Version Date Description Editor Technical Referee
1 05/9/2011 First draft AS/KT AL
2 27/1/2012 Second draft including results from scenarios AS/KT
3 23/2/2012 Final draft incorporating client comments AS/KT AL
4 19/3/2012 Final version of report AS/KT
5 31/4/2012 Publishable version of report
1 PPR601
Contents
Executive summary 3
1 Introduction 4
1.1 Background 4
1.2 Context 4
1.2.1 Air quality: the Hillingdon AQMA and AQAP 4
1.2.2 Transport Plan 6
1.2.3 Indicators 7
1.2.4 Greenhouse gas emissions 7
1.3 The need for the EDB 7
1.4 The scope of the EDB 8
2 Methodology 9
2.1 Overview 9
2.2 Task 1: Definition of road and rail networks 9
2.2.1 Road network 9
2.2.2 Rail networks 11
2.3 Task 2: Emission model development 12
2.3.1 Road transport 12
2.3.2 Rail transport 15
2.4 Task 3: Population of the networks with baseline activity data 16
2.4.1 Road transport 16
2.4.2 Rail transport 18
2.4.3 Heathrow airport 19
2.5 Task 4: Definition of scenarios and effects on traffic 19
2.6 Task 5: Calculation of emissions and compilation of EDB and Task 6:
Data analysis and reporting 25
3 Results 26
3.1 All estimated road and rail emissions 2010 26
3.2 In-borough baseline emissions 2010 27
3.2.1 Road traffic 28
3.2.2 Diesel rail 29
3.3 Scenarios 30
3.3.1 Modal shift 30
3.3.2 Dedicated bus service (Uxbridge to Heathrow) 32
3.3.3 Transport interchange at Uxbridge 33
2 PPR601
3.3.4 Reduction in traffic due to Crossrail 34
4 Discussion and further work 36
4.1 Headline results 36
4.2 Limitations of approach and further work. 37
4.2.1 TEEM refinements 37
4.2.2 Improved assessment methodology 38
Acknowledgements 40
References 40
Appendix A Integrating emissions from Heathrow Airport 42
Appendix B Traffic survey data 45
Appendix C Traffic data scaling note 46
Appendix D ARTEMIS rail model 57
Appendix E Scenarios considered for emission assessment 59
3 PPR601
Executive summary
The aim of this project was to produce an emissions database (EDB) for use by the
London Borough of Hillingdon. The Council had a need for this database to assess
changes in emissions and fuel consumption of individual and combinations of measures
in the Council‟s Transport Local Implementation Plan (LIP) and Air Quality Action Plan.
The EDB can also be used to provide evidence to justify the implementation of measures
and to provide data to assess progress with measures against performance indicators.
The outcome of the project has been to produce a working EDB including link-based road
transport emissions and a rail module that can be used by the Council. The EDB has
been used to determine the contribution of emission sources in the borough and to
assess the emissions impact of selected measures from the LIP against a 2010 baseline.
Emissions from electric trains (those operating on the London Underground, Paddington
to Heathrow airport rail link and Chiltern north national rail) are assumed to be from
power stations outside the borough, so have been treated separately. Nitrogen oxides
(NOx) and particulate matter (PM) emissions from these sources will not have any impact
on air pollution levels. CO2 emissions are of interest, however, since CO2 is a global
pollutant.
The headline findings were that, of sources in the EDB that are within the borough, road
traffic contributes to over 77 percent NOx emissions and 69 percent of PM emissions. Of
the road traffic sources, vehicles travelling on the motorways (M4 and M25) contributed
to over 60 percent of each pollutant. For CO2, road transport contributed to 95 percent
of the total emissions. Emissions from the diesel railway network contributed 23 percent
of overall NOx emissions, 31 percent to PM emissions and 5 percent of CO2 emissions in
the borough.
Emissions of NOX from electric trains were approximately half that of diesel trains, PM
emissions were very similar and for CO2, emissions were more than double.
Four measures from the LIP were assessed for their impact on emissions in the entire
borough or in a given local area.
Achieving modal shift targets in the LIP were found to reduce emissions of NOx,
PM and CO2.
Introducing a dedicated Uxbridge to Heathrow bus service may increase
emissions in the short term (due to a large increase in the numbers of buses),
but if cleaner buses were used, there would be longer-term emissions reductions.
On a per-passenger basis, emissions reductions were found due to improvements
to a bus and train transport interchange at Uxbridge. For example, for all
pollutants, emissions due to increased bus patronage buses were predicted to
reduce more than 25 percent and by more than 60 percent for increased train
use.
The introduction of Crossrail in 2017 was predicted to reduce borough-wide
emissions by more than two percent.
4 PPR601
1 Introduction
1.1 Background
The London Borough of Hillingdon (LBH) is currently estimating emissions of air
pollutants from all sectors of activity in the borough, one of the main objectives being to
establish the „carbon footprint‟ of the borough and meet emission targets. In addition,
implementation of a wide range of measures is being pursued by the Council in order to
reduce emissions and improve air quality.
TRL was commissioned to develop an emissions database (EDB) which will provide the
transport input to the borough‟s calculations, and will enable the effects of the measures
to be determined. This report outlines the methodology to develop the EDB and results
from an emissions assessment of selected measures.
1.2 Context
Local authorities in the United Kingdom are faced with a number of challenges and
obligations in relation to air pollution, including compliance with objectives for ambient
air pollutants in the UK Air Quality Strategy (Defra, 2007), progressing of measures to
improve air quality through Air Quality Action Plans (AQAP) and transport plans. Local
authorities are also required to report progress with respect to national indicators, and
meet targets for the reduction of greenhouse gas emissions. The relevance of these to
LBH is briefly summarised in the following sections.
1.2.1 Air quality: the Hillingdon AQMA and AQAP
The UK Air Quality Strategy (Defra, 2007) specifies objectives, as given in the Air Quality
(England) Regulations 2000 and Air Quality (England) (Amendment) Regulations 2002,
which apply to local authorities in their Local Air Quality Management (LAQM) duties. The
strategy sets out objectives for seven pollutants which are based on their health effects
over the short term and/or long term. The long term (annual mean) objective for
nitrogen dioxide (NO2) is exceeded in many parts of Hillingdon and, as a result, LBH
declared an Air Quality Management Area (AQMA) in 2001, covering the southern half of
the borough. Following a review in 2003 the AQMA boundary was increased to
encompass a wider area north of the A40, as shown in Figure 1.
5 PPR601
Figure 1: Map of LBH’s AQMA boundary (LBH, 2011).
The Council has had an Air Quality Action Plan (AQAP) in place for a number of years
(LBH, 2004). This includes a range of measures to improve air quality in the borough.
The majority of these measures are ongoing, and involve the co-operation of several
Council departments as well as neighbouring authorities. They include:
Funding - through the LIP - of a series of projects aimed at improving non-car
transport in the borough.
A „Walk on Wednesdays‟ initiative, in which Hillingdon has the highest uptake
amongst schoolchildren in London.
Launch of the Walkit system to help people to plan walking journeys along the least
polluted routes in West London.
A Driver Training initiative, which has covered all council drivers.
Several projects to improve freight handling at the South Ruislip and Uxbridge
Industrial Business Areas.
Various actions to raise awareness amongst the general public of air quality issues
and climate change, such as improved marketing of the AirTEXT1 system and
participation at various events in and around Hillingdon.
1 The airTEXT service provides air pollution alerts and forecast by text, email or voicemail,
http://www.airtext.info/
6 PPR601
A best-practice taxi review, providing recommendations on the opportunities for
the taxi market to reduce pollutant emissions (Latham et al, 2008).
An air quality network audit to ensure that the monitoring network in West London
is adequate to support future work carried out by the local authorities as part of
Local Air Quality Management (LAQM).
Commissioning of new air quality monitoring stations, including the Hillingdon
Sipson monitoring station in 2006 and Hayes station in 2008.
The development of a Road Network Monitoring Strategy to gain a better
understanding of traffic flows in the borough.
However, in its AQAP the Council recognised that there were limits to what it could do to
improve air quality, due to the following:
The Council does not control capacity at Heathrow. Whilst it is now current
Government policy for no third runway at Heathrow, LBH will continue to ensure
this is a preserved position due to the negative environmental impacts associated
with expansion at the airport. This is a particular threat following the Government‟s
decision to end the Cranford Agreement2 in 2009 as part of its expansion plans.
Major road links are the responsibility of the Highways Agency and Transport for
London (TfL).
Large-scale industry is regulated by the Environment Agency, although smaller
sites are subject to Council control.
Sources of pollution external to Hillingdon are outside the Council‟s remit.
1.2.2 Transport Plan
In London, transport plans are known as Local Implementation Plan (LIPs) and these are
required to set out how each London local authority will deliver the Mayor‟s Transport
Strategy (MTS) (TfL, 2011). LBH adopted its 2nd LIP (LIP2) in October 2011 for the
period 2011-2014.3 Objective LIP2 34 is relevant to improving air quality as it aims to
“Ensure that transport infrastructure improvements deliver better quality of life with
corresponding reductions in air pollution emissions”. Relevant actions within this
objective include improving the public realm (e.g. better pedestrian facilities in town
centres), enhancing villages near Heathrow (e.g. through parking schemes, HGV re-
routing, streetscape improvements), planting street trees and introducing parking and
traffic management schemes. For example, the traffic calming scheme is due to be
extended in Sipson Road and a 20 mph zone introduced in Hatch Lane. The Council‟s
AQAP is considered to be instrumental in helping deliver this objective.
The Sub Regional Transport Plan for West London5 outlines the framework for the LIP in
translating the MTS objectives to this region. The plan sets challenges including
2 The Cranford Agreement was a verbal deal struck in 1952 between the British Government and the residents of Cranford in London that prohibited take-off on the northern runway (except when absolutely necessary) due to noise concerns.
3 http://www.hillingdon.gov.uk/media/pdf/5/r/lip2_draft_consultation_jan2011.pdf
4 See Table 2.3 (List of objectives/Goals of relevant strategies/plans).
5
http://www.westlondonalliance.org/wlaextranet.nsf/0/36b4ac6b8b4d0c39802578300073132d/$FILE/110204%
20west-london-sub-regional-transport-plan.pdf
7 PPR601
improving north-south connectivity, improving access from and within key locations,
enhancing efficiency of freight movements and improving air quality.
1.2.3 Indicators
The Local Government White Paper Strong and Prosperous Communities, published in
October 2006 (DfCLG 2006), proposed a simplification of the performance framework for
local authorities. A single set of 198 national indicators was announced in 2007. These
indicators are now the only measures by which central government can set targets to
monitor local government performance.
Hillingdon worked with the Government Office for London6 to deliver its Local Area
Agreement (LAA) for the period 2008-20117. All targets in the Council LAA are based on
existing strategic plans and the priorities for action that communities have identified
over a number of years. There are 19 priority areas within the LAA. Some of the priority
areas are complementary to existing strategies, including the Council Plan and the
Sustainable Community Strategy. The priority areas are represented through 39 national
indicators, which include a number of measurable targets.
1.2.4 Greenhouse gas emissions
It is the intention of LBH to undertake projects and monitoring as part of the Council's
commitment to reduce the carbon footprint of its operations by 40 percent by 2015
relative to 2008 (LBH, 2009). LBH is keen to develop the necessary software to monitor
the effect of dedicated projects and to maximise their effects on emissions. Such
projects include the promotion of good practice and the establishment of links with area-
wide travel plan initiatives, as indicated in the borough‟s Travel Plan. This document is
still in draft format and has been embedded within the Council‟s Climate Change
Strategy (LBH, 2009).
The Council is exploring funding opportunities with TfL and the West London Partnership
with a view to further developing the West London Traffic model. The aim of this traffic
model is to provide a means of estimating the impacts of developments on transport
provision, plus any proposed intervention measures on CO2 and other pollutants (e.g.
NOx and PM).
1.3 The need for the EDB
LBH has identified a need to understand the impacts of measures for reducing emissions
in greater depth, with typical requirements including the following:
The likely reductions in emissions and fuel consumption from individual measures
and combinations of measures.
The evidence on which decisions for the practical implementation of measures can
be based.
The effectiveness of measures with respect to key performance indicators.
The effects of measures on the temporal and spatial distribution of air pollution.
6 The Government Office for London was abolished following the Government‟s spending review in 2010.
7 http://www.hillingdon.gov.uk/index.jsp?articleid=7578
8 PPR601
The EDB forms the evidence base for the evaluation of the impacts of emissions
reduction measures. It also supplements the existing greenhouse gas inventories for
LBH‟s own estate (NI185), and allows the Council to refine its estimates of greenhouse
gas emissions, as currently calculated using Defra‟s „top-down‟ approach8.
1.4 The scope of the EDB
The EDB covers the following emission sources:
Road transport (on public roads) including two wheelers, motorcycles and mopeds,
taxis, cars, buses and coaches, light goods vehicles (LGVs), rigid heavy goods
vehicles (HGVs) and articulated HGVs. Note that the data do not currently allow for
the disaggregation of Heathrow-related traffic on the surrounding road network.
Rail transport (diesel freight and passenger trains). Emissions from electric trains
on the national network and London Underground have been included in the overall
calculations and are presented separately.
Land-side and air-side ground operations (road vehicles) within the boundary of
Heathrow Airport.
Emissions from aircraft at Heathrow Airport are not being considered.
Spatially, these sources are considered within the whole of the borough of Hillingdon, an
area of approximately 115 km2 for the base year of 2010.
8 NI 185: Percentage CO2 reduction from local authority operations:
http://www.decc.gov.uk/en/content/cms/statistics/indicators/ni185/ni185.aspx
9 PPR601
2 Methodology
2.1 Overview
The development of the EDB has been broken down into the following six tasks:
Task 1: Definition of road and rail networks
Task 2: Emission model development
Task 3: Population of the networks with baseline activity data
Task 4: Definition of scenarios and effects on traffic
Task 5: Calculation of emissions and compilation of EDB
Task 6: Data analysis and reporting
These Tasks are described in more detail in the following sections.
2.2 Task 1: Definition of road and rail networks
This Task involved the spatial definition of the transport networks that are covered by
the EDB. To do this, the networks were divided into „links‟ which were effectively straight
lines between two node points. The grid references for the start and end points of each
link and characteristics of each link such as gradient, length, speed limit have been
defined where information is available.
2.2.1 Road network
The public road network in Hillingdon has been taken from the West London Traffic
Model (WLTM) provided by Transport for London (TfL)9. Originally it was hoped that the
WLTM model would be geo-referenced and based on a real-world network10 which would
mean that it could be spatially linked with the GLA‟s London Atmospheric Emissions
Inventory (LAEI) (GLA, 2010). The LAEI provides a database of emissions from all
sources (not just road traffic) in Greater London. The latest version contains emissions
for a base year of 2008 with forecasts to 2011 and 2015. Attribute information taken
from the LAEI is already represented as a project file for use in the TEEM tool (Transport
Enhanced Emissions Model) - this tool is used to generate emissions in this project11.
It was subsequently found that the links in the WLTM were not geo-referenced. Further
research found that it was not practicable to enable georeferencing of the WLTM within
this project. TfL has identified a potential solution12, but this is unlikely to be achieved in
the near future.
The benefits of a geo-referenced network include providing correct spatial representation
of the road network in the borough in a Geographic Information System (GIS). A geo-
9 Tim Cooper. Programme Manager and Policy Advice. Transport for London.
10 Personal communication, Tim Cooper (TfL). December 2010.
11 TEEM currently contains more than 12,000 road links in West London.
12 Personal communication, Tim Cooper (TfL). December 2010a.
10 PPR601
referenced network would also allow emissions to be better visualised and could be used
for dispersion modelling. It would therefore offer a more accurate analysis and
manipulation of transport and emissions, and allow a spatial link with the LAEI, as
identified above.
Figure 2 illustrates the extent of the WLTM road network in Hillingdon and Figure 3
shows an example of how the modelled network differs to the actual road layout. There
are 1,116 modelled links in the WLTM network with a total road link length of 604
kilometres.
Figure 2: WLTM road network in Hillingdon.
©Crown copyright. All rights
reserved London Borough of
Hillingdon 100019283 2009
11 PPR601
Figure 3: Example of WLTM network (in red) in relation to the actual road
layout.
The internal road network of Heathrow Airport has not been represented spatially in the
EDB. Road traffic on the surrounding road network associated with the airport (e.g.
passengers, staff and freight accessing the airport) could not be disaggregated, as this
information is not available within the WLTM.
The emission estimates for combined road operations (land-side and air-side) within the
Heathrow airport boundary have been taken en bloc from the LAEI as area emissions.
Further details outlining the method for integrating Heathrow in the EDB are given in
Appendix A.
2.2.2 Rail networks
The rail network has been obtained from digital maps from LBH and includes the
following:
Lines on the national network from Paddington and Marylebone (freight and
passenger trains);
London Underground; and
The rail link from Paddington to Heathrow airport (known as the Heathrow
Express).
Stations have also been mapped based on OS grid references using GIS.
Figure 4 provides a network map showing the location of the national rail network and
Paddington to Heathrow rail link (green lines), and Piccadilly (blue) and Metropolitan
(purple) London Underground lines travelling through Hillingdon. The stations within the
borough have been labelled.
Each railway line has been separated into 1 km links to improve the spatial resolution of
the emission estimates. The direction of flow on each link is denoted by a segment
©Crown copyright. All rights
reserved London Borough of
Hillingdon 100019283 2009
12 PPR601
name with the first three characters being an abbreviation of the departure station, a
sequential 2 digit number and three characters representing the arrival station. If the
first segment leaves or arrives at a station the name is tagged with either „LS‟ or „AS‟.
For example “HounW02HounC_AS” refers to the train from Hounslow West to Hounslow
Central.
Figure 4: Rail and underground network operating in Hillingdon.
2.3 Task 2: Emission model development
Task 2 in the development of the EDB involves the identification and/or definition of the
models used to generate the emission data. The following sections describe the approach
used.
2.3.1 Road transport
TRL, in conjunction with the West London Alliance Air Quality Cluster Group
(WLAAQCG)13 continue to develop the emission model; TEEM14. Being readily available
13 Hillingdon is a member of the WLAAQCG.
14 Current project entitled „TEEM 2012 Update‟ is due to be completed in June 2012. Lead representative of the
West London Alliance Air quality Cluster Group is Davide Pascarella of the London Borough of Brent.
©Crown copyright. All rights
reserved London Borough of
Hillingdon 100019283 2009
13 PPR601
and ideally suited to be used in the context of West London, TEEM has been used in this
project to calculate emissions from road transport.
In the EDB the activity data for road transport were taken from the West London Traffic
Model (see Section 2.4.1).
2.3.1.1 Overview of TEEM
The basic functional structure of TEEM is shown in Figure 5. It is a system comprising a
range of modules. TEEM itself is described by the blue box, and external sources of data
or software packages are shown in different colours.
Fundamental to TEEM is a representative model of the road network for which emissions
are to be estimated. The road network takes the form of a database of links, for which
the start points and end points are geo-referenced (i.e. they are stated as grid co-
ordinates). A road network of any size can be defined by the user.
Figure 5: Generalised TEEM structure.
For each road link in TEEM the characteristics of the road (e.g. length, gradient, speed
limit) and information on the flow, composition and operation of the traffic are required
so that emissions can be estimated.
14 PPR601
The purple boxes in Figure 5 show the traffic data which are currently used in TEEM15.
The sources of the traffic data used in TEEM are as follows:
(i) For most of the roads on the defined network – including all main roads - the
primary source of traffic data is a traffic assignment model. For West London, the
recent evolution of traffic model data originated from the Highways Agency (HA)
model; NAOMI16 which was developed in 1994 to assess the strategic effects of
introducing additional capacity to the M25. Naomi was based on the SATURN17
suite. The main simulation area covered by NAOMI includes the entire south east
of England. NAOMI formed the cordon extents for more detailed HA models
including M25AM (M25 Assignment Model) which in turn formed the basis of a
more detailed variant called the M25SWQM (M25 South West Quadrant Model)
focusing on West London and the Heathrow area. It was this variant which TfL
used to develop WeLHAM (West London Highway Assignment Model) the output
from which is applied for this particular study and termed the West London Traffic
Model (WLTM). Prior to this study TEEM applied the M25SWQM output.
(ii) The LAEI. The current version based on year of 2008.
(iii) Traffic surveys. Data for a small number of roads not covered by either the WLTM
or the LAEI, but considered strategically important by West London Boroughs
were extracted from existing traffic surveys. Survey data can also be used to
refine the data obtained from any traffic model or the LAEI.
The traffic data (and any other relevant attributes) are matched to the specific road links
on the modelled network. This is conducted using GIS. In GIS, each road link is
referenced to the M25SWQM traffic data and then to the LAEI data. This step is required
because the links in M25SWQM and the LAEI are often not ideal for modelling emissions.
For example, they may not provide enough spatial resolution, or may give only a poor
representation of road curvature. In each case, a unique identifier is allocated to each
road link, and then files containing the adjusted data are exported. Any available traffic
survey data can also be treated in this manner, but must initially be converted to the
LAEI format.
The adjusted M25SWQM and LAEI data files are then imported into TEEM via the user
interface. Within TEEM, the two files are combined (based on the unique link identifier)
using a „look-up‟ table. The look-up table is also developed in GIS. The output from the
look-up table effectively forms an internal database of „program data‟ for the required
network, as shown in Figure 5. These data must then be further adjusted to allow
emissions to be modelled.
To model emissions TEEM uses a particular set of definitions for vehicle categories, and
requires emission estimates for each link and each hour of the year. As the vehicle
categories and time periods are different from those used in the M25SWQM and the
LAEI, a series of traffic scaling factors is required. A pre-processor module within TEEM
applies scaling factors to each link separately. The overall average speed of the traffic on
each link – which is taken to be constant during the day – is entered separately via the
15 The TEEM software can be adapted to allow different types of input from other traffic models, but this would
require further development work.
16 NAOMI = New Assessment of Motorway Improvements
17 SATURN = Simulation and Assignment of Traffic to Urban Road Networks.
15 PPR601
user interface. The final traffic data are then sent to the emission model for processing.
This process is then repeated for all links on the modelled network, and the emission
calculations are stored within TEEM.
Finally, TEEM produces an output file which contains the link attributes and emission
values (in grammes per kilometre). The links in the TEEM road network are geo-
referenced, and the attributes can be viewed using GIS. The emission estimates from
TEEM can be used as inputs to emission maps, emission inventories and dispersion
modelling software. In such cases, GIS also aids visualisation of links.
2.3.1.2 Modification of TEEM for use in the Hillingdon EDB
For this project the following modifications to the TEEM software were made:
(i) The activity data have been taken from the WLTM rather than M25SWQM (see
explanation in 2.3.1.1), and the requirements for converting the data were
different.
(ii) The WLTM includes the effects of queuing within its modelling approach. The
implications of this for emission modelling were therefore considered and the
representativeness of the data was checked.
(iii) A constant speed for all vehicles over 24 hours is currently used in TEEM. The
limitations of this approach are recognised, so a greater temporal resolution
was used in the EDB for this study. This was possible by generating a speed
profile function which coincides with the flow profile. This allowed hourly
emissions to be estimated using both parameters. The model also limits HGV
speeds to a maximum speed of 90 km/h.
2.3.2 Rail transport
For the rail element of the EDB, emissions from trains were calculated using a modified
version of the model produced during the ARTEMIS project (see Appendix D for more
details of the ARTEMIS model). The model calculates energy consumption and emissions
for any train and any type of operation, based upon a train operation matrix entered by
the user. On a per vehicle basis this approach is slightly more complicated than that
used for road transport, but it is one of the few rail models which allows train speed to
be included in detail, and hence provides the required spatial dimension to the emission
estimates.
The classification of trains in emission models is generally simpler than the classification
of road vehicles. For each rail link, train movements were defined according to the five
train categories shown in Table 1. This classification was based on the requirements of
the ARTEMIS model. Consequently, a separate fleet model was not required.
16 PPR601
Table 1: Train categories.
Passenger/
Freight Train type Power source
Passenger High-speed train (HST) Diesel
Electric
Inter City Diesel
Electric
Regional Diesel
Electric
Urban Diesel
Electric
Freight Freight Diesel
Electric
Examples of the emissions factors applied in the prototype are shown in Table 2. The
emission factors for electrified rail take into account the mix of electricity generation
from power stations in the UK.
Table 2: Rail emissions factors (taken from the ARTEMIS model).
Traction
Type
Units CO2 CO NOx SOx NMHC PM
Electric g/GJ 167,800 27.4 631.8 1,446 20.2 69.9
Diesel g/GJ 74,440 250 1,320 90 70 80
2.4 Task 3: Population of the networks with baseline activity data
The EDB was populated with road traffic activity data, and then extended through the
inclusion of activity data for rail operations and Heathrow Airport.
2.4.1 Road transport
The activity data for road transport were based primarily on the outputs from the West
WLTM, complemented by surveys (given in Appendix B). The WLTM is a traffic
assignment model which, like M25SWQM is based on SATURN18. The base year of the
WLTM is 2009. These data were forecast to the base year of 2010 using a local scaling
factor based on the DfT TEMPRO 6.219 software. The activity data for road transport were
defined using the steps described below.
Step 1: Matching the WLTM and LAEI networks
18 SATURN = Simulation and Assignment of Traffic to Urban Road Networks
http://www.saturnsoftware.co.uk/index.html
19 http://www.dft.gov.uk/tempro/
17 PPR601
Spatial and attribute road link information could not be cross-referenced with the LAEI or
any other modelled road network. This was not possible for the following reasons. The
first was a mismatch of the node points (or the geo-referenced start/end points of line
objects representing road links). The second was a mismatch of the actual GIS line
objects (or links) between the two data sets. Spatial analysis was undertaken to
determine whether the datasets could be cross-referenced but it was found to not be
possible. TfL is currently looking into longer term solutions to resolve this network issue,
for example by mapping the nodes to an existing map in GIS (such as the Ordnance
Survey Meridian layer). For this study, the existing WLTM network had to be applied.
Step 2: Conversion of WLTM data
In the second step the outputs from the WLTM were converted into a form suitable for
emission modelling. This method was similar to that used in TEEM (as described by
Boulter et al. (2008)).
The Council provided the WLTM output data to TRL according to the information given in
Table 3. For a given link it was assumed in the WLTM that all types of vehicle travel at
the same speed.
Table 3: Parameters available from WLTM.
Parameter Units
Link ID -
„A‟ node (6 figure ref) -
„B‟ node (6 figure ref) -
Annual average daily traffic (AADT) -
Annual average weekday traffic (AAWT) -
Traffic speed km/h
Car and private hire vehicle (PHV) %
LGV %
OGV %
Taxi %
As emission estimates were determined on an hourly basis, the AADT (or AAWT) values
from the WLTM had to be distributed across the day. To do this, traffic flow scaling
factors (daily profiles) were applied for different roads and different days of the week
(including bank holidays). These have already been compiled for West London (Boulter
et al., 2008), but were updated in this study. Three outputs from the WLTM were scaled
- the AADT, traffic composition and speed. Appendix C provides details on how the
scaling factors were applied in this study.
The coarse vehicle composition was represented according to road type. This allowed for
a composition to be assumed where information is limited and simpler representation of
scenarios. In this way LBH will be able to assign a composition to a road if no other data
are available.
The EDB requires a detailed and representative vehicle fleet profile for the base year, in
this case 2010, and future years based on the scenarios. These could be extrapolated
from the current 2007 fleet profile developed in TEEM using video analysis in conjunction
18 PPR601
with estimates provided by the National Atmospheric Emission Inventory20. Using video
analysis in conjunction with automatic number plate recognition (ANPR) cameras offers
an opportunity to profile the traffic in terms of vehicle fuel use, size and age distribution.
These data allow emission standards of vehicles to be determined for a given road type.
By feeding these data into emission models, the contribution of emissions by vehicle
type can be derived. This provides useful information for development of appropriate
emissions mitigation measures.
Step 3: Incorporation of traffic survey data
Road traffic survey data were used to derive 24-hour vehicle speed and traffic flow
profiles. These profiles are used by TEEM to estimate emissions for each hour of the day.
Additional traffic survey data included are listed in Appendix B.
Step 4: Incorporation of data on traffic queuing
The different ways in which traffic queues are taken into account in emission models
were reviewed. The WLTM incorporates queuing/delay effects to some extent for each
link. However, to incorporate traffic queues into the EDB better, it may be most practical
to develop a separate road network where queue lengths are determined for each hour
in which they occur. This could be incorporated as a separate database for TEEM to
process, which would allow adjustments to be made more easily and impacts to be more
readily assessed. It should also be noted that emissions from queuing traffic are
determined from emission models by applying very low average speeds. So-called
„average speed‟ emission functions with emissions expressed in grammes per kilometre,
such as those includes in TEEM and most other emissions models, are not appropriate for
speeds below 5 km/h, particularly when nearing zero speed.
2.4.2 Rail transport
For the national rail network the baseline activity data were derived from a variety of
sources, including rail timetables for Chiltern Railways (operating passenger trains from
London Marylebone to Aylesbury and trains on the Chiltern Main Line from Marylebone to
Birmingham Snow Hill), First Great Western (trains from London Paddington towards
Wales) and Heathrow Express trains (rail link from Paddington to Heathrow airport).
Information from timetables included train routes and numbers of trains (passenger and
freight) that arrives at each station on the link during the day. The most recent version
of the National Rail trends yearbook, containing data for 2010, was consulted to
determine activity information for these lines, such as class of train, number of
passengers, train kilometres travelled, number of carriages and class of train.
Information on train occupancy could not be obtained, so a constant loading was
assumed. The class of train was used to determine emission rates according to the
factors given in Section 2.3.2.
20 http://naei.defra.gov.uk/report_link.php?report_id=626
19 PPR601
Activity data such as numbers of passengers, train kilometres, services per day for the
two London Underground lines operating in LBH were taken from the TfL website21.
Emission factors were taken from TfL‟s annual Environment Report22.
Table 4 compares some of this activity information obtained for the train lines. The table
shows train kilometres in LBH are highest for the Metropolitan line and lowest for the
Paddington to Heathrow rail link. However, in terms of passenger kilometres travelled
per year, services from Paddington (excluding those operated by Heathrow Express)
have the highest number followed by the Piccadilly line.
Table 4: Numbers of passengers and kilometres travelling on train lines each
year in Hillingdon.
Train service Number of train
kilometres
accessible to LBH
combined
directions (km)
Approximate
number of
passengers per
train kilometre*
Number
of
services
per day
Million
passenger
kilometres
travelled per
year.
a*b*c*365 (a) (b) (c)
Long distance and
commuter services from
Marylebone
40 103 47 70.7
Long distance and
commuter services from
Paddington
30 124 116 157.0
Paddington – Heathrow
airport rail link
15 101 158 87.3
Piccadilly line 35 36 190 87.4
Metropolitan line 46 15 150 37.7
2.4.3 Heathrow airport
As stated earlier, the gridded emission estimates for Heathrow Airport were taken en
bloc from the LAEI due to a lack of available disaggregated data. The emissions for 2008
in the LAEI were assumed to be apply in 2010.
2.5 Task 4: Definition of scenarios and effects on traffic
This Task involved developing scenarios to test the impact of measures in the AQAP and
LIP2 on emissions. A range of scenarios were considered and discussed with LBH to
21 http://www.tfl.gov.uk/
22 http://www.tfl.gov.uk/assets/downloads/corporate/environment-report-2008.pdf
20 PPR601
prioritise those to test within TEEM. The scenarios considered are given below and in
more detail in Appendix E.
1. Change in traffic flows from 2010 to 2013 without any intervention.
2. Meeting LIP targets for modal shift from 2010 to 2013/4 compared to a
baseline.
3. Reduction of cars on the Uxbridge Road.
4. Dedicated bus service on the Uxbridge to Heathrow airport route.
5. Improving transport interchanges to increase public transport use.
6. Reduction in road traffic due to introduction of Crossrail.
7. Implementation of business travel plans that lead to modal shift.
8. Re-investigating the impact of major planning developments.
9. Introduction of a north-south bus service and corresponding reduction in car
travel.
10. Relieving traffic congestion from Heathrow villages.
11. Overall impacts of generalised policies from the baseline (for example
reducing all traffic in the borough by one percent or increasing buses by five
percent).
12. Urban freight consolidation centre for Heathrow to reduce HGV travel.
Following consultation on these twelve scenarios with London Borough of Hillingdon, the
following four were taken forward to the modelling stage. For each scenario, annual
modal emissions (CO2 NOX, PM) were determined for the borough for the do-minimum
(baseline) compared with the do-something (with scenario). Emission results and
differences are given in Section 3.
Final scenario 1: Meeting LIP targets for modal shift from 2010 to 2013/4
compared to a baseline.
This scenario looked at the impact of meeting LIP targets to increase walking and cycling
and reduce car modal split as detailed Table 5. Note that the figures in the LIP2 did not
total 100%; figures presented in the table below have been scaled to total 100%.
To achieve this using TEEM, the following steps were taken:
Step 1: A working/travelling population in LBH was determined for 2010.
Step 2: The number of passengers per day per mode of travel was derived.
Step 3: The number of vehicles this represents (based on occupancy rate) was
derived.
Step 4: The average emissions for each type of vehicle (from available data) for
each year were obtained to calculate total emissions for the number of
vehicles (from Step 3).
21 PPR601
Step 5: Passenger emissions per kilometre driven per year were derived from Step
4.
Step 6: Emissions in the LIP target year of 2013 were compared with those from
the base year 2010.
Table 5: Percentage of people travelling in/out of Hillingdon by mode (LBH
2011).
Mode 2006 to 2008 Base
2010
2011 2012 2013
(Percentage)
Rail 1.0 1.0 1.0 2.0 2.0
Underground 3.0 3.0 4.0 4.9 4.9
Bus 10.0 10.1 11.0 11.7 12.7
Taxi 1.0 1.0 1.0 1.0 1.0
Car 60.0 59.6 57.0 53.7 51.7
Cycle 1.0 1.0 1.0 1.5 1.5
Walk 24.0 24.2 25.0 25.4 26.3
TOTAL 100 100 100 100 100
For this scenario it was assumed that public transport polices included in the LIP have
the effect of increasing modal switch year on year. The increased patronage of the public
transport system is achieved without any increase in the number and frequency of
services compared with 2010. The scenario also assumes that the car occupancy rate
reduces as a result of policies to switch to use public transport.
Final Scenario 2: Dedicated bus service on the Uxbridge to Heathrow airport
route.
This scenario considered introducing an additional dedicated bus service along a
proposed route from Uxbridge to Heathrow airport. This scenario fits in with the Sub
Regional Transport Plan for West London which outlines the need to increase bus
demand on this important bus corridor23. It was found that there are six bus services
operating between Uxbridge and Heathrow24; the A10, U3, and A40/A30/724/740 taking
three different routes, as indicated in Figure 6. The most direct route is that taken by
the A40/A30/724 and 740 (labelled as A40 on the map), which have no late evening
services. The same route was chosen for the proposed additional bus service in this
scenario.
23
http://www.westlondonalliance.org/wlaextranet.nsf/0/36b4ac6b8b4d0c39802578300073132d/$FILE/110204%
20west-london-sub-regional-transport-plan.pdf
24 http://www.tfl.gov.uk/tfl/gettingaround/maps/buses/busdiagrams.asp
22 PPR601
Figure 6: Existing bus services from Uxbridge to Heathrow.
This scenario assumed that this additional service would have the effect of transferring
one percent of car passengers onto buses. This assumption is based on Table 5 (LBH,
2011a) where the yearly target is to increase bus patronage across the borough by just
under one percent. The assumption here is that car drivers are attracted to use the new
bus service in addition to regular bus patrons. There are a number of ways to make
services more attractive including:
Effective marketing of the route;
Effective timetabling including being sensitive to modal interchange options;
Appropriate location of bus stops;
No obstruction of dedicated bus routes by other vehicles between intersections;
Parking places along the roads eliminated where conflicts might occur between
the buses and any vehicles manoeuvring into parking places;
A traffic control system with automatic bus location. These systems keep the
drivers aware of any disruptions to the timetable and allow appropriate corrective
and assistance measures to implemented, and;
Traffic signal priority. The system can be used at all intersections regulated by
traffic lights, independent of the time-table.
It was assumed that there was only one individual per car, meaning that the one percent
reduction was applied to the total car flow on each road link along the entire route.
Assuming a one percent modal shift, for an average day the total number of additional
©Crown copyright. All rights
reserved London Borough of
Hillingdon 100019283 2009
23 PPR601
passenger journeys formally made by car and now being provided by the bus service is
8,823. The passengers are assumed to use the service at various locations on the route
(and, hence, not all passengers will ride the entire route).
It is assumed that the service would operate single decker buses with a capacity of 60
(combined seated and standing). The proposed bus service was assumed to run every
half hour in both directions (i.e. a total of four buses an hour), operating 24 hours a day
– which would equate to a maximum of 5,760 passengers a day. If passengers
remained on the service for just a quarter of the route and the bus remained at full
capacity this would equate to 23,040 different passenger trips (5,760 times 4) a day.
Without knowing the trips that are likely to be made, it is therefore difficult to estimate
the percentage of passenger trips in terms of total potential passenger trips provided by
the service.
The bus flow on each link was increased to reflect the change in numbers. TEEM was run
to calculate emissions from this scenario against a 2010 baseline.
Final Scenario 3: Improving transport interchanges to increase public transport
use.
This scenario modelled the impact of increasing passengers using existing services (bus
and train) due to improvements made to the Uxbridge transport interchange. The
scenario assumed that the interchange affects commuters within a 1 km study area (see
Figure 7). The effect on the annual average daily car flow within all WLTM road links
within this radius was reduced by one percent (assuming a car occupancy of one driver).
Bus and rail services (the Metropolitan underground line) operating within this radius
were also assumed to increase by one percent as it was assumed that drivers switching
from cars travelled by a combination of bus and train through the interchange. The
average bus occupancy was taken as 15 people.25 The same occupancy figure was used
for the Metropolitan line.
25 http://philtaylor.org.uk/2007/02/mayors-buses-not-so-green-check-the-facts/
24 PPR601
Figure 7:1 km study area (travel cordon) from Uxbridge interchange.
Final Scenario 4: Reduction in road traffic due to introduction of Crossrail.
Little evidence has been provided examining the likely effects of modal switch through
the introduction of Crossrail. However, a 2010 Summary Update Report26 on the benefits
of Crossrail suggests that it could lead to a traffic reduction of two percent in London in
general with higher benefits accruing on parallel roads and in the Heathrow area in
2017.
To assess this, the traffic data was scaled forward from 2010 to 2017 using the factor
1.0530 (based on the national and local situation) from DfT TEMPRO 6.2. The fleet
composition was estimated for 2017 based on the National Atmospheric Emissions
Inventory (NAEI) fleet27. This accounts for the London Low Emission Zone and taxi
strategy (in 2017, 46 percent of London taxis are Euro 4 and 54 percent are assumed to
be Euro 5). Note that the current version of TEEM28 does not include Euro 6/VI29
26 TfL Crossrail Business case summary report (2010).
27 Murrells T, Li Y (2009). UK - Proportion of VKM by Euro Standard (All Vehicles).
rtp_fleet_projection_April09_FINAL (29-6-09).xls. ED48954003
28 The TEEM 2012 Upgrade project will include a revision of the current emissions factors.
©Crown copyright. All rights reserved London
Borough of Hillingdon 100019283 2009
25 PPR601
emission standards. As such any estimated Euro VI/6 vehicles are assumed to comply
with Euro V/5.
The effect of Crossrail was assessed by reducing the traffic flows in 2017 by two percent
globally and then a further two percent on the M4 and A4 corridors and estimating
emissions using TEEM to compare against the 2017 baseline. These assumptions were
based on the estimated benefits outlined in TfL‟s 2010 Summary Updated report30 .
2.6 Task 5: Calculation of emissions and compilation of EDB and Task 6: Data analysis and reporting
In Task 5, emissions from transport were calculated using the models and activity data
described in the sections above. Task 6 involved checking and analysing the results and
presenting the data in tables to demonstrate the impact of each scenario with reference
to the relevant baseline, following the method in Section 3.
29 Euro VI/6 emission standards will apply in January 2013 and 2014 respectively.
30 TfL Crossrail Business case summary report (2010) (ibid).
26 PPR601
3 Results
3.1 All estimated road and rail emissions 2010
This section presents a comparison of road and rail emissions sources for the base year
2010. This section includes emissions from both diesel and electric rail. The contribution
of all modelled sources to emissions of CO2, CO, NOx and PM within and outside the
borough are summarised in Table 6, Table 7 and Figure 8.
The results show that emissions from road vehicles on the motorway are the most
significant, contributing more than 50 percent of emissions of CO2 and CO and a third of
PM emissions. Overall, road traffic contributes the most to modelled emissions sources,
making up more than 85 percent of CO2 and CO emissions and two-thirds of NOx
emissions. Relatively equal proportions of PM emissions are emitted by road traffic and
rail (i.e. electric and diesel). Landside airport vehicles contribute more to CO emissions
than to other pollutants.
Table 6: Estimated emissions within borough per year, 2010.
Source Emissions (tonnes per year)
CO2 CO NOx PM
Minor "B" roads 120,349 334 243 9
Major "A" roads 144,145 426 321 12
Motorways 429,581 1,964 1,140 40
Landside airport vehicles 3,359 258 22 1
Airside airport vehicles 25,743 133 240 17
Diesel national rail trains
(Chiltern south, FGW and
Thames)
32,913 111 584 35
Total 756,091 3,225 2,549 114
Table 7 gives emissions from electric rail only. Seventy percent of the CO2 emissions
come from electric trains compared to diesel trains. Emissions from diesel trains are
more significant for CO and NOx than from electric trains and similar for PM.
Table 7: Estimated emissions from electric trains per year, 2010.
Source Emissions (tonnes per year)
CO2 CO NOx PM
Electric trains (Paddington-Heathrow
rail link) 4,746 1 18 2
Electric trains (London Underground
and Chiltern north national rail) 72,337 12 272 30
Total 77,083 13 290 32
27 PPR601
CO2 CO
NOx PMMinor "B" roads
Major "A" roads
Motorways
Landside airport vehicles
Airside airport vehicles
Diesel rail (Chiltern S/Thames/ FGW/Chiltern N)
Paddington-Heathrow rail link (electric)
London Underground and Chiltern N (electric)
Figure 8: Contribution of modelled sources to pollutant emissions, 2010.
3.2 In-borough baseline emissions 2010
This section compares emissions estimates for road and diesel rail sources in the
borough. Emissions from electric trains have been treated separately so are not reported
in this section. The road emissions include traffic on the WLTM network and landside and
airside vehicles within the Heathrow boundary. The results from the study were
compared with road and rail emission estimates reported in the LAEI forecast for the
year 2011. The LAEI does not provide estimated figures for the year 2010 (GLA, 2010).
Emissions from road transport were greater than from diesel rail, contributing to 77
percent of modelled NOx emissions and 69 percent of PM emissions in EDB.
Figure 9 provides a summary of these results. The figure shows that overall borough
emission estimates from the EDB are higher than those in the LAEI. The annual road
transport emissions of NOX in Hillingdon were 1,965 tonnes per year compared with 890
tonnes per year in the LAEI (i.e. approximately 2.2 times greater). PM10 emissions were
79 tonnes per year compared to 65 tonnes per year for the LAEI (i.e. approximately 1.2
times greater). Note that the LAEI includes fugitive emissions of PM from brake and tyre
wear whereas the current version of TEEM does not include this functionality. There may
be a combination of reasons for these differences including different methodologies and
data provided from the WLTM. The data, methodology and model setup were checked to
ensure that they had been applied correctly in TEEM.
In terms of diesel rail, NOX emissions in the EDB for the borough were 584 tonnes per
year compared with 563 tonnes in the LAEI. For PM10, the emission estimates from the
EDB were 35 tonnes per year, compared with 11 tonnes per year in the LAEI). The
methodology used to calculate rail emissions including the applied rail emission factors
28 PPR601
was checked and it was concluded that the emission factors would benefit from further
verification studies to compare modelled estimates with in-service emissions.
Road 1,965 tpa - 77%
Diesel rail584 - tpa 23%
LBH - EDB NOX
Road 79 tpa -69%
Diesel rail35 - tpa 31%
LBH - EDB PM
Diesel rail563 tpa - 39%
Road Transport890 tpa - 61%
LAEI - NOx
Diesel rail11 tpa- 15%
Road Transport65 tpa - 85%
LAEI - PM
Figure 9: Contribution of road and rail sources to pollutant emissions in
Hillingdon, 2010.
The next sections present the baseline emission data in more detail, individually for each
source.
3.2.1 Road traffic
3.2.1.1 Public roads
Table 8 presents a summary of the estimated emissions from road transport in
Hillingdon on the WLTM road network. The table shows that emissions from vehicles
travelling on motorways (i.e. the M4 and the small section of the M25 within the LBH
boundary) contribute most significantly to emissions (over 60 percent for each pollutant)
compared to emissions on the more minor roads. The total length of the motorway
network in Hillingdon (including both directions of travel) was 100 km which is lower
than the total length of all A and B roads (504 km). However, the traffic flows on the
motorway were much higher, reaching up to 100,000 vehicles per day on some stretches
of the network.
As the motorway network is controlled by the HA it is outside the control of the Council.
Although speed management measures have been introduced on the M25 by the HA,
there is little evidence as to its effectiveness in terms of reducing emissions or improving
air quality. Equally, the majority of traffic on the motorway originates outside of
Hillingdon so are not affected by interventions targeted at the local travelling population.
It is also likely that HGVs servicing businesses located in Hillingdon use the motorway
network. Therefore, when introducing policies to reduce HGV flow on the motorway,
there needs to be a balance with their impact on the local economy. The use of electric
29 PPR601
vehicles operating from freight forwarding centres located close to the motorway but
away from residential areas may offer the Council a solution to reducing local emissions
from HGVs.
Table 8: Emissions from road transport in Hillingdon in year 2010.
Emission
source
Emissions (tonnes per year)
CO2 CO NOx PM
Minor “B” Roads 120349 334 243 9
Major “A” roads 144145 426 321 12
Motorways 429581 1964 1140 40
TOTAL 694,075 2,724 1,703 60
3.2.1.2 Heathrow airport
As outlined in Appendix A, on-airport road traffic emissions associated with Heathrow
airport have been calculated based on gridded emissions from landside and airside
vehicles in the LAEI (GLA, 2010). The emissions from road traffic on the Perimeter roads
with the boundary according to the WLTM model have been included in the road traffic
part of the model rather than Heathrow emissions to avoid double counting. The
emission estimates from on-airport road traffic are summarised in Table 9. The results
show that emissions from airside vehicles (except for CO) are much greater than
emissions from landside vehicles (i.e. those vehicles travelling to offices and car parks).
Due to the lack of available data, the proportion of traffic on the surrounding road
network that is associated with the operation of the airport cannot be quantified.
Table 9: On-airport road vehicle emissions in Hillingdon.
Emission source Emissions (tonnes per year)
CO2 CO NOx PM
Landside vehicles 3,359 258 22 1.3
Airside vehicles 25,743 133 240 17
TOTAL 29,102 391 262 18
3.2.2 Diesel rail
The total emissions per year from the diesel trains are shown in
Table 10 for 2010. As indicated in Figure 9, NOx emissions from diesel rail are 584
tonnes per year and PM emissions are 35 tonnes per year.
Table 10: Emissions from diesel railway lines operating in Hillingdon.
Train line Emissions (tonnes/per year)
CO2 CO NOx PM
Diesel national rail (Chiltern
S/Thames/ FGW/Chiltern N) 32,913 111 584 35
30 PPR601
3.3 Scenarios
This section presents the emissions results from the four scenarios, against the baseline
for the relevant year or against a 2010 baseline.
3.3.1 Modal shift
This scenario followed the steps identified in Section 2.5. The percentage of people
travelling by each mode of transport (Table 5) was multiplied by the assumed travelling
population from 2008 of 200,000 people in the borough (160,000 working population
and 40,000 school population)31. The resulting figures are given in Table 11.
Table 11: Travelling population by mode per day.
Mode 2006 to 2008 Base 2010
2011 2012 2013
(vehicles per day) Rail 2000 2020 2000 3902 3902
Underground 6000 6061 8000 9756 9756
Bus 20000 20202 22000 23415 25366
Taxi 2000 2020 2000 1951 1951
Car 120000 119192 114000 107317 103415
Cycle 2000 2020 2000 2927 2927
Walk 48000 48485 50000 50732 52683
Total 200000 200000 200000 200000 200000
These figures were multiplied by assumed occupancy rates (i.e. number of people per
vehicle type) to obtain the number of vehicles per day in the borough. These estimates
are given in Table 12. Above-average occupancy rates were assumed for this scenario,
which means they were higher than for other scenarios (e.g. for the Uxbridge
Interchange). It is assumed that the number of public service vehicles per day remain
constant from 2010 except for taxis (see Section 2.5). The number of passenger cars
decreases, as a higher proportion of people are predicted to cycle, walk or take public
transport (see Table 5).
Table 12: Number of vehicles per day based on assumed occupancy rates.
Mode Occupancy rate
2006 to 2008
Base 2010 2011 2012 2013
(vehicles per day)
Rail 20 100 101 101 101 101
Underground 20 300 303 303 303 303
Bus 15 1,333 1,347 1,347 1,347 1,347
Taxi 1.532 1,333 1,347 1,333 1,301 1,301
Car 1.633 75,000 74,495 71,250 67,073 64,634
Total - 78,067 77,593 74,550 70,618 68,309
31 http://www.hillingdon.gov.uk/media/pdf/p/m/Hillingdon_profile_brochure_Hillingdon_profile_brochure.pdf
32 http://www.lti.co.uk/news/index.php?p=98
33 http://assets.dft.gov.uk/statistics/releases/national-travel-survey-2010/nts2010-01.pdf
31 PPR601
Based on these figures above, an average weighted journey length of 7.3 km for
Hillingdon34 and average emission rate for each vehicle in each year was calculated –
assuming a speed of 48 km/hour (see Table 13). Emissions for NOx, PM and CO2 were
calculated using TEEM for each year. It was assumed that emissions from buses and
trains remain constant from 2010-2013.
Table 13: Average emission rates by mode of transport.
Mode
2010 2011 2012 2013
NOx emission rate (g/km)
Rail (diesel) 0.189 0.189 0.189 0.189
Underground (electric)
0.026 0.026 0.026 0.026
Bus 3.936 3.706 3.412 3.005
Taxi 0.515 0.502 0.476 0.451
Car 0.192 0.172 0.153 0.139
PM emissions rate (g/km)
Rail 10.65 10.65 10.65 10.65
Underground 6.809 6.809 6.809 6.809
Bus 0.079 0.079 0.079 0.078
Taxi 0.060 0.059 0.057 0.055
Car 0.028 0.027 0.026 0.026
CO2 emission rate (g/km)
Rail 10.7 10.7 10.7 10.7
Underground 6.87 6.8 6.8 6.8
Bus 126.9 126.0 123.6 120.3
Taxi 53.1 53.1 52.9 52.4
Car 41.4 40.8 39.9 39.1
The resulting emissions for each year are given in Table 14. These results (in grammes
per passenger kilometre) show that emissions of NOX, PM10 and CO2 for rail,
underground, bus and taxis reduce from the base year of 2010 to 2012 and stabilise in
2013. In terms of passenger car NOX emissions, a gradual reduction from 2010 to 2013
is predicted due to the modal shift and the fact that the emission standards of all
existing vehicles will improve over time. For PM10, car passenger emissions reduce in
2011 and then stabilise whilst for CO2, emissions are variable.
34 http://www.hillingdon.gov.uk/media/pdf/p/m/Hillingdon_profile_brochure_Hillingdon_profile_brochure.pdf
32 PPR601
A focal point of the results shown in Table 14 is emissions estimated for „rail‟ – both
diesel and electric trains. The relative high values are due to two factors, the higher
emissions rates (g/km) when compared to for example electric rail and or road vehicles
and the relatively low occupancy rate of 20 passengers per train. It is very difficult to
obtain credible statistics on train occupancy rates. The results for 2012/13 do suggest
that if the LIP targets are met then modal switch will start to have a substantial effect on
train passenger emissions.
The results show that there would be emissions benefits of pursuing policies which
encourage bus patronage, particularly for PM and CO2. It is likely that once TfL buses are
fitted with NOX exhaust after-treatment devices, similar benefits will be evident for NOx
and NO2.
Table 14: Emissions by mode of transport.
Mode Base 2010 2011 2012 2013
NOX passenger emissions per kilometre (g/pkm)
Rail 69.2 69.9 35.8 35.8
Underground 9.4 7.1 5.8 5.8
Bus 1.9 1.7 1.4 1.2
Taxi 2.5 2.4 2.3 2.2
Car 0.9 0.8 0.7 0.6
TOTAL 84 82 46 46
PM10 passenger emissions per kilometre (g/pkm)
Rail 4.17 4.21 2.16 2.16
Underground 1.04 0.78 0.64 0.64
Bus 0.04 0.04 0.03 0.03
Taxi 0.29 0.29 0.28 0.27
Car 0.13 0.12 0.12 0.12
TOTAL 5.7 5.4 3.2 3.2
CO2 passenger emissions per kilometre (g/pkm)
Rail 3,898 3,937 2,018 2,018
Underground 2,492 1,888 1,548 1,548
Bus 59 56 52 47
Taxi 256 259 258 256
Car 179 187 183 179
TOTAL 6,883 6,327 4,058 4,047
3.3.2 Dedicated bus service (Uxbridge to Heathrow)
The changes in emissions calculated from this scenario are presented in Table 15. As the
number of buses along the route were increased, assuming a 24-hour bus service
running every 30 minutes in both directions, this had a greater impact on emissions than
the assumed one percent reduction in cars along the route. Therefore the emissions of
NOx, CO2 and PM10 were predicted to increase; for example, NOx emissions were
predicted to increase by four percent. However, if occupancy rates were taken into
account, the emissions per person are likely to be lower for the buses.
33 PPR601
If such a measure were to be implemented in the future then the scenario would need to
be retested to account for the inclusion of future improved emission standards of TfL
buses, particularly via the introduction of its new hybrid buses and the imminent NOX
reduction retrofit programme in order to meet its 2015 emissions target35. There would
also need to be more detailed studies to appreciate the likely patronage of any north to
south bus service.
Table 15: Change in emissions along Uxbridge to Heathrow bus route due to the
introduction of a dedicated bus service (2010).
Scenario Emissions (tonnes/year)
CO2 CO NOx PM
Baseline 25,889.7 85.0 49.2 2.0
Bus scenario 25,988.5 84.8 51.3 2.0
% change +0.4 -0.2 +4.1 +1.5
3.3.3 Transport interchange at Uxbridge
For this scenario, the changes in emissions are reported on a per passenger basis as the
assumption is that there is an increase in number of passengers using public transport
but not an increase in the numbers of buses or trains (as per the „modal shift‟ scenario) .
The reduction of one percent of cars on road links within the 1km study area was based
on Table 5 (LBH, 2011a). The year-on-year target is to reduce cars by more than one
percent but a balance is drawn with targets to increase bus and rail patronage across the
Borough by just under one percent. In essence, this may be considered as taking a
precautionary approach. A reduction of the traffic leads to a decrease in road traffic
emissions within the 1km study area. These results are shown separately for each
pollutant in Table 16.
Table 16: Change in road traffic emissions in 1km radius from Uxbridge station,
2010.
Scenario Emissions (tonnes/year)
CO2 CO NOx PM
Baseline 8858.43 18.14 13.62 0.28
1% reduction in cars 8799.71 18.03 13.57 0.28
% change -0.7 -0.6 -0.4 -0.6
35 http://www.transportxtra.com/magazines/local_transport_today/news/?id=27768
34 PPR601
This scenario assumes that the one percent of car drivers use the transport interchange
and transfer to buses and rail within the 1km study area. This change in average annual
daily traffic flow of 3,879 cars results in an increase in occupancy of buses from 15
people to 19 people (based on the numbers of buses per day) and an increase in train
occupancy by 26 people from 15 to 41 people based on 150 services a day (3,879/150).
In terms of total numbers of rail passengers a year, this would increase from 821,250 to
over 2.2 million.
Table 17 shows the changes in emissions per passenger for each mode of transport in
grammes. For cars, there is no change in emissions per passenger, as the occupancy
rate is assumed to stay at one. The increase in bus patronage is predicted to have a
reduction in emissions of all pollutants by 27 percent. The emissions from rail are much
higher than from buses and cars, and the large increase in rail passengers reduces
emissions by 63 percent. Train emissions are closely associated with weight. Emission
rates for this study do not consider varying levels of passenger numbers and hence the
effect this has on emissions. This is because an average train carriage weighs
approximately 43 tonnes and the additional weight of passengers when fully laden would
equate to approximately three tonnes. This additional weight will therefore make a
minimal impact on emissions. More substantial savings in emissions are gained by
reducing the weight of the carriage. But in doing so there are important implications in
terms of train safety and performance.
Table 17: Differences in emissions on a per-passenger basis, 2010.
Scenario Emissions per passenger (g/year)
CO2 CO NOx PM
Baseline 46.1 0.1 0.05 0.002
Reduction in car
passengers 46.1 0.1 0.05 0.002
Baseline 112.9 0.22 0.29 0.01
Increase in bus
passengers 89.1 0.17 0.23 0.004
Baseline 12155.8 2.0 45.8 5.1
Increase in rail
passengers 4462.1 0.7 16.8 1.6
3.3.4 Reduction in traffic due to Crossrail
The results of this scenario are given in Table 18. The reduction in vehicle traffic
assumed due to the introduction of Crossrail has resulted in a predicted reduction of
emissions of all pollutants by just over two percent from the baseline. The changes in
fleet composition in 2017 have also contributed to the reduction in emissions from the
2010 baseline (as given in Table 8). For example, NOX emissions are predicted to reduce
by 33 percent and PM exhaust emissions by 62 percent from 2010-2017 without
Crossrail.
35 PPR601
Table 18: Modelled road traffic emissions in Hillingdon due to the introduction
of Crossrail in 2017.
Scenario Emissions (tonnes/year)
CO2 CO NOx PM
Baseline (2017) 790,453.6 2,580.9 1,342.4 27.0
Crossrail scenario 771,939.5 2,518.1 1,311.4 26.3
% change -2.3 -2.4 -2.3 -2.4
36 PPR601
4 Discussion and further work
This section sets out the key findings of the project and highlights some points for
discussion and recommended areas of future work.
4.1 Headline results
The project has provided a usable emissions database (EDB) and rail module as a tool
for use by London Borough of Hillingdon. This tool allows the Council to determine the
key emission sources in the borough and to carry out emission modelling to identify the
impact of individual and combinations of measures in their Air Quality Action Plan or
transport plan (LIP).
The EDB is based on the road network from the West London Traffic Model (WLTM)
which contains 1,166 modelled road links within the borough. Activity data have been
compiled in TEEM to obtain emissions estimates for the base year of 2010. The EDB also
contains a rail module which includes national rail, London underground (not reported in
borough emission results) and the Paddington to Heathrow airport rail link. Emission
factors from trains were taken from the ARTEMIS project and activity data were taken
from a variety of sources including timetables, yearly figures from National Rail and data
from Transport for London. The EDB did not explicitly include the road network within
Heathrow airport. Instead emissions from landside and airside vehicles were taken
directly from the LAEI.
The headline findings were that, of sources in the EDB that are within the borough, road
traffic contributes to over 77 percent NOx emissions and 69 percent of PM emissions. Of
the road traffic sources, vehicles travelling on the motorways (M4 and M25) contributed
to over 60 percent of each pollutant. For CO2, road transport contributed to 95 percent
of the total emissions. Emissions from the diesel railway network contributed 23 percent
of overall NOx emissions, 31 percent to PM emissions and 5 percent of CO2 emissions in
the borough
Emissions of NOX from electric trains were approximately half that of diesel trains, PM
emissions were very similar and for CO2, emissions were more than double.
Of the four LIP scenarios assessed, the improvements to emissions in the borough were
mixed, with some increases in total emissions in the borough predicted in the short-term
associated with an increase in numbers of buses and trains used. In summary:
The modal shift targets included in the LIP appear to be achieving CO2, NOX and PM
emissions reductions. Actions are now required to affect these desired targets.
The dedicated bus service may increase emissions in the short term but these
emissions will reduce further if cleaner buses are deployed. The service appears very
attractive on a passenger emissions basis. However it is suggested that an optimistic
target for reductions in emissions could be achieved if the service could reach half
capacity by the third year of operation. In terms of this scenario a large increase in the
number of buses was assumed for a 24-hourly dedicated service, with only a one
percent associated reduction in car travel. This resulted in an increase in NOx emissions
by four percent and PM emissions by 1.5 percent.
37 PPR601
The results for improving the transport interchange are difficult to judge because of
the reliance on the assumptions made in terms of service provision and the derivation of
individuals likely to switch mode. On per passenger basis emissions, reductions are
apparent. When results are viewed on an emissions per passenger basis, the results
show improvements to emissions. For example, the increase the numbers of bus
passengers meant that emissions of all pollutants were predicted to reduce by 27
percent for buses and 63 percent for trains. The per passenger metric makes it easier to
discern the relative benefits of modal choice options although it is understood that
consideration should initially be drawn to absolute emissions differences. However, it is
worth noting that the use of both metrics can provide a more robust appreciation of the
benefits. However, it is suggested that more detailed surveys are required to improve
these assumptions before benefits can be fully appreciated.
The introduction of Crossrail has to be seen in the context of benefits in the opening
year. The scenario was predicted to reduce borough-wide emissions by more than two
percent in 2017. Given that this is a strategic policy measure the local emissions benefits
probably extend beyond the scope of the current scenario test. It is worth noting that
current results indicate that Crossrail will contribute to a reduction in borough emissions
and so on this basis it is a policy worth supporting. The issue perhaps is one of
understanding the spatial aspect of the NOX and PM emissions benefits (e.g. on
motorways, A-roads or general background emissions).
4.2 Limitations of approach and further work.
4.2.1 TEEM refinements
4.2.1.1 Scenario test module
The study has demonstrated the use of baseline inventories to test policy options.
However, manipulation of the emission database is currently achieved outside of the
TEEM and Rail tools. There is functionality in TEEM to save projects (i.e. the model setup
representing a scenario). However, it would be useful if a dedicated module could be
developed that could further process information on the internal road database via a
scenario interface. For example, for this study the scenario interface would allow a
scenario to be developed (i.e. percentage changes made to traffic flows by vehicle type,
or change of average speed) and then saved. The user would then assign these changes
to the appropriate road links. These changes would not affect the baseline database
unless the changes were exported to an external file. The module could be developed to
incorporate the necessary changes required to represent a range of policy measures.
Also the „scenario‟ module could allow the effects of joint policy measures to be tested.
4.2.1.2 Average-speed emission factors
TEEM uses average speed emission factors, which have their limitations. For example,
they are difficult to apply at very low speeds such as in traffic queues. Further work
could investigate the use of an instantaneous emissions model and use of drive cycles
with an instrumented vehicle driven along chosen streets in the borough. This would
more accurately represent driving styles and the varying emissions during the day.
Hence, emissions factors derived from instantaneous measurements could be integrated
for use within the TEEM tool.
38 PPR601
4.2.1.3 Traffic scaling factors
TEEM uses scaling factors to convert peak flows to AADT flows. These factors could be
improved by obtaining more information on daily classified flow profiles by road type
using automatic traffic counters.
Ultimately, the most effective methodology to increase our understanding of emissions
and provide robust evidence to target mitigation measures involves undertaking ANPR
surveys. As well as capturing detailed vehicle fleet statistics the cameras can also
provide the speed of every individual vehicle. Depending on the survey design ANPR
could also provide strategic fleet profiles geared to examining measures such as low
emissions zones or the effects of modal shift policies.
4.2.1.4 Geo-referencing the WLTM
The WLTM road links were not spatially geo-referenced (i.e. they did not follow a real
world road network). This meant that it could not be linked with the LAEI network with
the TEEM model. As explained in Section 2.4.1 this is not important when assessing
emission impacts at the strategic level but very important when examining impacts on a
local scale (i.e. for individual roads). There are various methods which can be explored
to provide a spatially-resolved modelled road network including running various
functions using a GIS. There may be scope to apply „fuzzy‟ logic in order to associate
nodes based on radius criteria. Ultimately, by obtaining the datasets used to derive the
WLTM traffic model it may be possible discern how the spatial aspects have been offset
from real world.
4.2.2 Improved assessment methodology
4.2.2.1 Assumptions to inform scenario test options
A number of assumptions had to be made to test the impacts of each scenario. For more
robust calculations these assumptions could be refined. For example, there was little or
no evidence to discern bus and rail patronage in the borough. This information can be
obtained from ticketing information (TfL for buses and rail operators) but is largely
dependent on whether these data are readily available and in what format. Another issue
for buses is to understand detailed patronage (bus stop level information) particularly
when bus passes are used and not individual tickets. The following surveys are
recommended:
Bus patronage surveys initially targeting strategic routes (7am to 7pm).
Rail patronage surveys (7am to 7pm).
Surveys at local bus and rail interchanges to understand traveller origin
destination information.
Local business surveys at key commercial centres to understand travel options of
employees.
4.2.2.2 Freight scenario test options
This study excluded the testing of freight based policies included in the LIP. Heavy
vehicles are disproportionate in terms of their emissions and substantial emissions
reductions can be achieved by measures which reduce freight traffic. It is recommended
39 PPR601
therefore that scenarios are developed to examine the outcomes of proposed measures
particularly given the growing interest in freight forwarding in town centres using electric
vehicles. Also given the proximity of Hillingdon, in the West London context, the options
to provide freight forwarding alongside strategic corridors may have greater potential
than previously thought.
4.2.2.3 Comparisons of results with other sources
There are two aspects to undertaking comparison studies. One element deals with
comparing modelling input data such as traffic flows, traffic speeds and fleet composition
and the second aspect involves comparing emissions output. For example, the former
could involve comparing traffic activity derived on a link basis between the WLTM and
the LAEI, whilst the latter could involve comparing outputs from TEEM with another
emissions model. However, comparisons of outputs are currently problematic as the
methodology used to derive the baseline EDB for Hillingdon differs from existing
methods such as that applied for the LAEI. One of the fundamental differences is the
application of 24-hour speed and traffic flow profiles. Alternatively, it may be possible to
manually set up other emissions models to represent the setup in TEEM.
Future development of TEEM could refine its validation module so that an output from
TEEM (i.e. for a single link) could be compared with a standard output from another
emissions model. This would provide a useful auditing tool for practitioners.
4.2.2.4 Disaggregated Heathrow road traffic data
To properly identify transport measures, the disaggregation of Heathrow versus non-
Heathrow traffic data on the surrounding road networks should be undertaken as a
matter of priority. It is understood that traffic modelling to understand this is currently
being undertaken by BAA Heathrow, and London Borough of Hillingdon will look to share
the data when this becomes available for the modelling of future scenarios.
It may be possible to apply factors to the existing WLTM data to indicate what level of
traffic, on a route or link basis, is influenced by Heathrow operations.
40 PPR601
Acknowledgements
The work described in this report was carried out in the Vehicle Safety Division (SSV) of
Transport Research Laboratory.
References
Boulter P G, Barlow T J, Emmerson P, Mao H and Turpin K (2008). West London
Traffic and Enhanced Emission Model (TEEM): traffic scaling factors. TRL Unpublished
Project Report UPR/IE/033/07. TRL Limited, Wokingham.
DfCLG (2006). Strong and prosperous Communities: The Local Government White
Paper. Department for Communities and Local Government
Defra (2007). The Air Quality Strategy for England, Scotland, Wales and Northern
Ireland (Volume 1). Department for Environment, Food and Rural Affairs in partnership
with the Scottish Executive, Welsh Assembly Government and Department of the
Environment Northern Ireland.
DfT (2007). Emissions Methodology for Future LHR Scenarios. Department for
Transport. AEAT/ENV/R/2323 Final.
http://www.dft.gov.uk/consultations/archive/2008/heathrowconsultation/technicalreport
s/futuremethodology.pdf
DfT (2009) Road traffic by type of vehicle and class of road: Transport Statistics of
Great Britain 2009: Department for Transport
DfT (2009a) Public Transport Statistics Bulletin GB: 2008 Edition. Department for
Transport
GLA (2010). London Atmospheric Emissions Inventory. Base Year 2008. Greater
London Authority.
Latham, S., Boulter P., McCrae I., Turpin K. (2008). A best practice guide for
reducing emissions from taxis in London. TRL Unpublished Project Report UPR/IE/48/08.
TRL Limited, Wokingham.
London Borough of Hillingdon (2004). Air Quality Action Plan. The London Borough
of Hillingdon. June 2004.
London Borough of Hillingdon (2009). Climate Change Strategy 2009 to 2012
(Final). London Borough of Hillingdon.
London Borough of Hillingdon (2011). Air Quality Action Plan Progress Report 2011.
London Borough of Hillingdon (2011a) Local Implementation Plan 2011 – 2014
Improving Transport in Hillingdon.
Lindgreen E and Sorenson S C (2005). Simulation of energy consumption and
emissions from rail traffic. Report MEK-ET-2005-4, Department of Mechanical
Engineering, Danish Technical University, February, 2005. Deliverable 7a for ARTEMIS,
DG-TREN contract 1999-RD.10429.
Spencer A (2010). Slough International Freight Exchange: Framework Travel plan: Doc
Ref 73382/R1.3 August 2010
TfL (2011). Mayor‟s Transport Strategy (MTS) Accessibility Implementation Plan.
Transport for London.
41 PPR601
Underwood C, Walker T and Pierce M J (2004). Heathrow Emission inventory 2003:
Part 1A report produced for BAA Heathrow Netcen/AEAT/ENV/R/1657/Issue 4; August
2004.
42 PPR601
Appendix A Integrating emissions from Heathrow
Airport
Following discussions with BAA and representatives of LBH it was evident that detailed
information on the traffic activity on individual roads at Heathrow Airport would be
difficult to obtain. For example, the methodologies used in the 2002 Heathrow Emissions
Inventory (e.g. Underwood et al., 2004) are commercially restricted documents, for
which only the executive summaries are provided in the LAEI report (GLA, 2010).
Slightly more information is provided in the report Emissions Methodology for Future LHR
Scenarios (DfT, 2007). Another issue is how often updates of Heathrow Emissions
Inventory will be made available, and whether LBH would have access to these updates.
Undoubtedly the LBH does not want to adopt a methodology which is heavily reliant on
having to request data from any third party if the process is constrained by regulation.
The road transport emission data for Heathrow Airport were extracted from the LAEI.
The LAEI methodology sets out how emissions at Heathrow are calculated, and updates
to the LAEI are fairly regular (i.e. every 2 years, the latest being produced in 2010 with
a 2008 base year).
The following was determined to enable the LAEI data to be included in the EDB:
Whether emissions from Heathrow could actually be extracted from the LAEI.
Which emission sources have been defined in the LAEI.
Which pollutants are included in the LAEI.
The spatial extent of each source.
There are two general types of road transport sources in the LAEI for Heathrow:
Landside road vehicles. Most of the roads (and emissions from vehicles) on the
airport landside road network are included in the road traffic element of the LAEI.
However, there are some additional sources within the airport boundary, such as
public car parks, taxi operations (taxi feeder parks and forecourts) and staff car
parks. Emissions from these additional sources have been estimated for 1 km2
cells, as indicated in Figure A1. The Figure shows the underlying land features,
Heathrow car parks, etc., the 1 km2 grid cells, and the grid cell centre points
coloured to indicate where emissions have been included.
Airside road vehicles and plant36. According to the LAEI (GLA, 2010), emissions
from airside vehicles and plant are estimated based on the total fuel dispensed
(for each fuel type) from the various filling stations at the airport. Emissions are
summated over all vehicles categories and all types of fuel. Further information
can be found in the DfT (2007) report. As with landside vehicles, emissions have
been aggregated as grid sources (Figure A3).
In Figure A1 and Figure A2 the grid points shown have been extracted from the main
LAEI database, which extends across the whole of Greater London. The points in yellow
equate to where zero emissions have been estimated for land-side and air-side road
vehicles, and the pints in red indicate where emissions are greater than zero. The LAEI
provides annual emission estimates of SO2, NOx, CO, CO2, NMVOC, benzene, 1,3-
butadiene, PM10 and methane.
36 Plant - principally heating plant.
43 PPR601
Figure A1. Heathrow land-side vehicles: gridded emission cells (GLA, 2010).
Figure A2: Heathrow air-side vehicles and plant: gridded emission cells (GLA,
2009).
All roads within the airport boundary (managed or otherwise by BAA) that are included in
the WLTM model are shown in blue in Figure A3. There are a total of 65 road links within
the boundary and these are referred to as perimeter roads.
Heathrow boundary
Grid centre point with emissions >0
Grid centre point with emissions of
zero
Car parks etc.
Land lines
44 PPR601
Figure A3: WLTM road links within the Heathrow airport boundary –
perimeter roads.
In the EDB it was assumed that road vehicle emissions at Heathrow Airport are the sum
of the land-side and air-side gridded outputs, as estimated in the LAEI and as indicated
by Figure A1 and Figure A2. Note that traffic on the perimeter roads from the WLTM
road network shown in Figure A3 were included as part of the WLTM road traffic
emissions rather than those associated with Heathrow.
LHR Perimeter roads
WLTM road links
45 PPR601
Appendix B Traffic survey data
Table B1. Additional data from local traffic surveys
Location Metric Flow Speed Period Source X Y
Heathrow Spur hourly y Constant TRADS 507460 178269
A3044 Hatch Lane
hourly y N Constant up to
07/09
LBH 505926 177232
A408 Sipson Road hourly y N Constant up to
07/09
LBH 507621 177291
A437 High Street
Harlington
hourly y N Constant up to
07/09
LBH 508851 177223
A4180 Ducks Hill Road Hourly y y 1 week per year LBH 507687 190540
B465 West Drayton
Road
Hourly y y 1 week per year LBH 508268 181788
B466 Eastcote Road Hourly y y 1 week per year LBH 509652 187744
B466 High Road
Ickenham
Hourly y y 1 week per year LBH 508666 187030
B466 High Road
Eastcote
Hourly y y 1 week per year LBH 510859 189009
B466 Long Lane Hourly y y 1 week per year LBH 507821 185830
B467 Swakeleys Road Hourly y y 1 week per year LBH 506360 185903
B467 Harefield Road Hourly y y 1 week per year LBH 505963 185320
B469 Green Lane Hourly y y 1 week per year LBH 508864 191513
B470 Iver Lane Hourly y y 1 week per year LBH 505067 182284
B472 Joel Street Hourly y y 1 week per year LBH 510299 189530
Station Road West
Drayton
Hourly y y 1 week per year LBH 506291 179661
Yeading Lane Hayes Hourly y y 1 week per year LBH 511159 181946
Victoria Road Ruislip Hourly y y 1 week per year LBH 511440 185400
Harvil Road, Harefield Hourly y y 1 week per year LBH 506075 187780
Dawley Road Hayes Hourly y y 1 week per year LBH 509040 179191
Breakspear Road North Hourly y y 1 week per year LBH 505695 190275
Northwood Road
Harefield
Hourly y y 1 week per year LBH 505714 190696
Shepiston Lane Hayes Hourly y y 1 week per year LBH 508650 178442
Ladygate Lane Ruislip Hourly y y 1 week per year LBH 508157 188135
Pole Hill Road, Hayes Hourly y y 1 week per year LBH 508121 182805
46 PPR601
Appendix C Traffic data scaling note
C.1 Introduction
In the EDB the West London Traffic Model (WLTM) was used to generate activity data for
road traffic, and emissions were estimated from these activity data using TEEM Version
2.2. However, the outputs of the former could not be used as direct input to the latter
due to differences in the time periods covered and the way in which traffic flow,
composition and speed were described. Various conversion factors and scaling factors
were therefore developed to align the two models.
Figure C.1 provides a summary of the steps taken to scale the traffic data in the format
of a flow chart. Further details on this process is given below.
C2. Input requirements of TEEM
Previous versions of TEEM operated on an hourly time base. The version of TEEM which
was used in conjunction with the WLTM operated on an annual time base so the output
from the WLTM was scaled prior to input into TEEM. All input and output values were
therefore stated as annual averages. TEEM therefore required the annual average daily
traffic flow (AADT), traffic composition and speed. TEEM‟s ability to accept inputs on
different basis allows it to have a greater level of flexibility than other models.
The main categories of vehicle in TEEM are listed in Table C1. TEEM required the annual
average percentage of the total traffic flow in each category. For the purposes of the
EDB, traffic composition was defined using the „Level 2‟ structure.
Table C1. TEEM vehicle categories.
Level 1 Level 2
Light-duty vehicle (LDV) Car
LGV
Taxi
Heavy-duty vehicle (HDV) Rigid HGV
Articulated HGV
Bus
Coach
Two-wheel vehicle (2WV) Moped
Motorcycle
The annual mean speed input in TEEM was applicable to all vehicle types.
47 PPR601
Figure C1. Flow chart summarising process to scale traffic data.
48 PPR601
C3. Output from West London Traffic Model
The WLTM provided results for an „average typical weekday‟. The time periods covered
by the traffic model are shown in Table C2. For the morning and evening peak periods
the results were for the individual hours shown in the Table, whereas for the inter-peak
case the results were a one-hour average for the whole period. In other words, the same
values could be applied to each individual hour of the period 10:00-16:00. Therefore,
eight hours of the day were modelled in total. No results were provided for the periods
00:00-08:00, 09:00-10:00, 16:00-17:00 and 18:00 to 00:00.
Table C2. Traffic model time periods.
Description Modelled period
Morning peak (AM) 08:00 to 09:00
Inter-peak period (IP) 10:00 to 16:00
Evening peak period (PM) 17:00 to18:00
For each time period covered by the WLTM, various output parameters were available.
The parameters which were the most relevant to emission modelling were provided to
TRL, and these are listed in Table C3.
Table C3. Traffic model parameters (separate flow and speed
data are provided for the AM, IP and PM periods).
Parameter type Parameter Description Units Example
Link description ANode Start node ID code - 62424
BNode End node ID code - 62000
AXco x grid coordinate of A node - 509553
AYco y grid coordinate of A node - 175400
BXco x grid coordinate of B node - 509583
BYco y grid coordinate of B node - 175400
Distance Link length m 36
N_lanes Number of lanes at the
junction (might be different
from the number of lanes
along the link)
- 2
Traffic flow Total_flow Total traffic flow PCUs 195.5
Traffic composition Flow_UC1 Number of cars (personal) PCUs 107.8
Flow_UC2 Number of cars (business) PCUs 8.6
Flow_UC3 Number of LGVsa PCUs 64.9
Flow_UC4 Number of HGVsb PCUs 9.9
Flow_UC5 Number of Taxis PCUs 0.3
Traffic speed ff_speed Free-flow speed km/h 25
Link_speed Average link speed km/h 20
a LGVs are defined as all vehicles less than or equal to 5.2 m in length.
49 PPR601
b HGVs are defined as vehicles greater than 5.2 m in length, and include heavy
goods vehicles, buses and coaches.
Vehicle numbers (flows) were modelled as passenger car units (PCUs). Buses, coaches
and motorcycles were not modelled explicitly. It is also worth noting that the sum of the
flows for the individual vehicle categories was not equal to the total traffic flow
(ACTFLOW). This was because in the WLTM ACTFLOW is the sum of the individual actual
flows plus „remaining‟ traffic from the previous peak hour (termed PASSQ).
C4. Conversion of WLTM data
The traffic data for the WLTM periods in Table C2 had to be converted to annual average
values for use in TEEM.
Where possible and where appropriate, the scaling factors previously developed for TEEM
were also used in this work. When the West London Air Quality Group commissioned TRL
to develop TEEM, an important part of the work was the production of traffic scaling
factors for the West London area (Boulter et al., 2008). These scaling factors were
required to align the outputs from the traffic model being used at the time (NAOMI) with
the input requirements of the emission model module in TEEM, which at the time
operated on a one-hour time base. In the NAOMI model used originally with TEEM,
results were provided for the periods 07:00-10:00, 10:00-16:00 and 16:00-19:00. The
scaling factors were developed using an extensive database of hourly measurements
from automatic traffic counting sites at various locations in West London during 2001
and 2005. There was also a need to determine 24-hour speed profiles. Scaling factors
were not developed for this purpose. Instead, the diurnal speed profile was calculated
internally in the TEEM software using the speed-flow relationships in NAOMI in
combination with the new 24-hour traffic flow profile.
For the EDB the outputs from the WLTM were treated in three separate stages:
1. Determination of AADT flow.
2. Determination of annual average traffic composition.
3. Determination of annual average traffic speed.
The three stages are described in the following sections.
Because the outputs from the WLTM were slightly different to those of NAOMI, a slightly
different approach was required for the EDB.
It is worth noting that it would not be difficult to replace the EDB scaling factors described here with new ones based upon more recent traffic data.
C4.1 Determination of AADT flow
Rather than developing entirely new sets of traffic flow scaling factors, the WLTM outputs
were adjusted to allow the scaling factors produced by Boulter et al. (2008) to be
applied. The following steps were taken for each road link (and associated road type):
Step 1: Conversion of total traffic flow in PCUs to actual flow
50 PPR601
The WLTM flows were stated in terms of passenger car units (PCUs), and were therefore
converted to actual vehicle flows. For all light-duty vehicles the PCU factor was 1. For
heavy-duty vehicles the PCU factor was 2. In other words, the WLTM flows for HDVs
were divided by 2 to give actual vehicle flows.
Step 2: Conversion of WLTM format to NAOMI format for weekdays
Adjustments were needed to allow for the fact that the AM and PM time periods covered
by the NAOMI model (AM = 07:00-10:00, PM = 16:00-19:00) were different from those
covered by WLTM. The IP period was the same in WLTM and NAOMI.
The following conversions were therefore required:
WLTM data for 08:00-09:00 to average hourly NAOMI for 07:00-10:00.
WLTM data for 17:00-18:00 to average hourly NAOMI for 16:00-19:00.
The resulting conversion factors for different road types are shown in Table C4. The road
types in the Table were explained by Boulter et al. (2008), and the values are based on
real-world traffic data for 2001 and 2005.
Table C4. Weekday conversion factors (WLTM to NAOMI).
Conversion factor by road type
Motorwaya A_1 A_2 A_3 A_4 B_3b B_4 Minor
road WLTM 09:00-09:00 to NAOMI AM period
Mean factor 0.942 0.918 0.949 1.013 0.993 0.789 1.033 1.065
Number 1720 4787 5726 2485 363 20 43 249
Standard deviation 0.071 0.068 0.079 0.295 0.083 0.111 0.081 0.247
Confidence interval 0.003 0.002 0.002 0.012 0.009 0.048 0.024 0.031
WLTM 17:00-18:00 to NAOMI PM period
Mean factor 0.963 0.939 0.949 0.986 1.010 1.564 0.988 1.046
Number 1720 4787 5726 2485 363 20 43 249
Standard deviation 0.045 0.044 0.056 0.096 0.074 1.076 0.067 0.232
Confidence interval 0.002 0.001 0.001 0.004 0.008 0.471 0.020 0.029
a M4 only
b Small sample, large CI, therefore value for A_4 roads used.
Step 3: Application of NAOMI scaling factors
The NAOMI scaling factors were provided in Tables 24 to 33 of the report by Boulter et
al. (2008). The values for Monday-Thursday were used here. For any minor roads not
included in the WLTM, the traffic flow scaling factors were taken from Table 34 of that
report. It should be noted that for minor roads the scaling factors were normalised to the
average hourly flow for the 24-hour period, which must therefore be obtained from a
source other than WLTM (e.g. estimation or measurement).
Step 4: Summation of hourly flows
51 PPR601
The hourly flows for weekdays were summed over the 24-hour period.
Step 5: Scaling factors for Saturdays and Sundays
The WLTM did not provide any traffic data for Saturdays or Sundays. New scaling factors
were therefore derived to calculate the total daily flows on these days using the total
weekday flow from Step 4. Again, these scaling factors were calculated from the TRL
traffic database for West London, and using data for the years 2001 and 2005. The
resulting scaling factors are given in Table C5.
Table C5. Scaling factors for Saturdays and Sundays.
Road type Scaling factors by road type (total flow)
Weekday to
Saturday
Weekday to Sunday
Motorwaya 0.873 0.813
A_1 0.795 0.789
A_2 0.807 0.774
A_3 0.919 0.842
A_4 0.946 0.863
B_3b 0.946 0.863
B_4 0.947 0.807
Minor road 0.920 0.762
a M4 only
b Small sample, therefore value for A_4 roads used.
Step 6: Calculation of AADT
The AADT was calculated using the following equation:
Where:
AADT = annual average daily traffic flow (vehicles/day)
FWD = annual average weekday traffic flow (vehicles/day)
FSAT = annual average weekday traffic flow (vehicles/day)
FSUN = annual average weekday traffic flow (vehicles/day)
Public holidays were ignored.
C4.2 Determination of traffic composition
TEEM required the percentage of the total flow in each vehicle category (level 2 in Table
C1), presented as annual mean values. To do this, the TEEM inputs were mapped onto
the WLTM vehicle categories, as shown in Table C6.
52 PPR601
Table C6. Mapping of vehicle categories.
WLTM output
TEEM input
(level 2)
Car (personal) } Car Car (business)
LGV
LGV
Taxi
Taxi
HGV { Rigid HGV
Artic HGV
Bus
Coach
Moped
Motorcycle
In order to achieve this the following steps were taken for each road link (and associated
road type):
Step 1: Conversion of PCUs to actual flow
The WLTM flow for each vehicle category was converted from PCUs to actual vehicles as
described earlier for total traffic flow.
Step 2: Summation of car flows
Cars used for personal travel and cars used for business travel were summed.
Step 3: Calculation of total daily flows by WLTM vehicle category and day type
The following equation was used to determine the total daily flow for each WLTM vehicle
category (car, LGV, taxi, HGV):
Where:
N_DAYi,d = Estimated total number of vehicles per day in vehicle category i (car,
LGV, taxi, HGV) and type of day d (weekday, Saturday, Sunday).
Flowp,i = WLTM flow for period p (AM, IP or PM) and vehicle category i.
SFp,i = Scaling factor for period p and vehicle category i.
The values of the scaling factors are provided in Table C7, C8 and C9 for weekdays,
Saturdays and Sundays respectively. Again, these were derived using the TRL database
of measurements.
53 PPR601
Step 4: Calculation of total weekly flow
It was assumed that the weekly average proportions of vehicles in the traffic would be
representative of the annual average. For each WLTM vehicle category (cars, LGVs, taxis
and HGVs), the weekly flow was calculated as follows:
Where i represents the vehicle category.
The overall total weekly flow was then calculated using the following equation:
Step 5: Calculation of WLTM vehicle percentages
The proportions of cars, LGVs, taxis and HGVs in the traffic flow were obtained using the
following equation:
Where i represents the vehicle category.
54 PPR601
Table C7. Scaling factors for WLTM vehicle categories - weekdays.
Road type Car LGV Taxia HGV
SFAM SFPM SFAM SFPM SFAM SFPM SFAM SFPM
Motorway 4.498 5.352 5.851 5.240 4.498 5.352 6.047 6.130
A_1 3.695 4.665 5.425 4.619 3.695 4.665 5.523 5.725
A_2 3.531 4.347 4.245 4.531 3.531 4.347 4.715 5.317
A_3 4.374 6.515 5.485 6.638 4.374 6.515 5.547 7.350
A_4 4.264 5.599 4.895 5.474 4.264 5.599 4.679 5.575
B_3b 3.910 5.950 4.651 5.042 3.910 5.950 5.247 7.052
B_4 3.910 5.950 4.651 5.042 3.910 5.950 5.247 7.052
Minor road 3.943 5.601 4.324 4.809 3.943 5.601 5.309 6.638
a Assumed to be same as cars.
b B_3 assumed to be same as B_4.
Table C8. Scaling factors for WLTM vehicle categories - Saturdays.
Road type Car LGV Taxia HGV
SFAM SFPM SFAM SFPM SFAM SFPM SFAM SFPM
Motorway 4.896 5.561 5.071 5.880 4.896 5.561 5.677 6.152
A_1 4.539 5.323 5.033 5.240 4.539 5.323 6.179 5.769
A_2 4.702 5.257 4.086 4.828 4.702 5.257 4.885 6.113
A_3 5.869 7.227 4.848 7.729 5.869 7.227 5.468 7.019
A_4 4.774 5.944 4.252 5.310 4.774 5.944 4.329 5.859
B_3b 5.296 6.692 4.638 6.142 5.296 6.692 4.662 6.791
B_4 5.296 6.692 4.638 6.142 5.296 6.692 4.662 6.791
Minor road 5.024 6.351 3.918 5.449 5.024 6.351 5.228 6.480
a Assumed to be same as cars.
b B_3 assumed to be same as B_4.
Table C9. Scaling factors for WLTM vehicle categories - Sundays.
Road type Car LGV Taxia HGV
SFAM SFPM SFAM SFPM SFAM SFPM SFAM SFPM
Motorway 5.414 6.156 5.088 6.661 5.414 6.156 5.558 7.239
A_1 5.957 5.658 5.858 5.741 5.957 5.658 6.078 6.727
A_2 6.655 5.479 5.461 005.908 6.655 5.479 5.706 6.686
A_3 9.005 6.922 6.549 7.190 9.005 6.922 6.695 8.366
A_4 7.721 5.850 6.149 7.434 7.721 5.850 5.274 6.225
B_3b 9.308 6.356 8.079 5.297 9.308 6.356 5.585 6.355
B_4 9.308 6.356 8.079 5.297 9.308 6.356 5.585 6.355
Minor road 8.177 6.330 7.136 5.661 8.177 6.330 4.688 7.718
a Assumed to be same as cars.
55 PPR601
b B_3 assumed to be same as B_4.
Step 6: Calculation of HDV split
The HDV component was sub-divided into rigid HGVs, articulated HGVs, buses and
coaches. Splits were based on road type in accordance with Transport Statistics of Great
Britain (Dft, 2009). Road types applied included motorways, „A‟ roads and minor roads.
According to Public Transport Statistics Bulletin (DfT, 2009a) there are around 57,200
buses and 23,200 coaches in Great Britain. Therefore, coaches represent 28.8% of all
buses and coaches. This value was used for A_1 and A_2 road categories. For
motorways, it is assumed that only coaches are present, and for all other roads it is
assumed that only buses are present.
C4.3 Determination of traffic speed
Scaling factors for traffic speed were not determined by Boulter et al. (2008), as speeds
were calculated internally in the TEEM software using the speed-flow relationships in
NAOMI. For this study, scaling factors for traffic speed were not calculated from the TRL
database, as this approach was considered to be unreliable. For example, the road
classification approach used by Boulter et al. (2008) did not take into account the
possibility that different roads within a given category could have a different speed limit,
and therefore any scaling factors may have been influenced by the actual roads included.
Consequently, it was decided that the outputs from WLTM would be used directly.
The speed data in WLTM corresponded to PCUs, and therefore traffic composition was
dealt with implicitly. However, the values were average values for the WLTM periods and
therefore conversion to an annual mean was required.
In order to calculate the annual average traffic speed, the period speeds were weighted
by traffic flow using the equation:
Where:
SAVG is the annual average speed (km/h)
SAM, SIP, SPM and SOP are the speeds in the corresponding periods (km/h)
FAM, FIP, FPM and FOP are the traffic flows in the corresponding periods
FTOTAL is the total daily traffic flow (FAM + FIP + FPM + FOP)
The following assumptions were made concerning traffic speed:
The speed during the period 07:00-10:00 (SAM) was taken to be the same as the
WLTM speed for the period 08:00-09:00.
The speed during the period 10:00-16:00 (SIP) was taken to be the same as the
WLTM speed for this period.
The speed during the period 16:00-19:00 (SPM) was taken to be the same as the
WLTM speed for the period 17:00-18:00.
The speed during the period 19:00-07:00 (SOP) was taken to be the same as the
WLTM free-flow speed.
56 PPR601
As the OP flow was not provided by WLTM, this had to be calculated. The following
equation was used:
Where SFOP is a scaling factor for weekday traffic (see Table C10), again derived
from the TRL database.
Table C10. Scaling factors to derive OP flows (weekdays only).
Road type Scaling factor (SFOP)
Motorwaya 0.336
A_1 0.289
A_2 0.337
A_3 0.428
A_4 0.446
B_3b 0.446
B_4b 0.446
Minor road 0.327
a M4 only
b Small sample, therefore value for A_4 roads used.
Speeds on Saturdays and Sundays were not taken into consideration.
57 PPR601
Appendix D ARTEMIS rail model
The ARTEMIS rail model calculates energy consumption and emissions for any train and
any type of operation, based upon a train operation matrix entered by the user. The
principles of the model are illustrated in Figure B1.
Figure D1: The basic principles of the ARTEMIS rail emission and
energy consumption model (Lindgreen and Sorenson, 2005).
Train operation between any two locations consists of a number of periods of
acceleration, operation at a roughly constant speed, and braking. In general, for a given
type of train and operation the distribution of these operational modes will be similar
between every stop. The ARTEMIS model uses a matrix of speed and acceleration
intervals, and each operational mode of a driving pattern can be described in terms of
these intervals.
Energy consumption is calculated using driving resistances. These include:
Rolling resistance – this is mainly due to deformation processes that occur at
the contact point between the wheel/tyre and the surface. For wheels on rail,
this is relatively low (in comparison to pneumatic tyres on tarmac) and is
simple a product of the mass force (weight times gravity) and the rolling
resistance coefficient (taken to be 0.002 for wheel on rail).
Aerodynamic drag (wind resistance) – this is proportional to the square of the
speed of the train, and is also dependant on the frontal area of the train and
its drag resistance. It is also affected by the air density, which may vary with
altitude (it will also vary slightly with day-to-day variations in meteorological
conditions), though for the current work, the air density has been kept
constant in the equations.
Climbing resistance – climbing resistance (when going uphill) and downhill
force (when going downhill) are dependent on the mass force of the train
(mass times gravity) and sine of the gradient angle. For the small angles
58 PPR601
likely to be encountered, the gradient (rise/fall [m] over the length of the line
[m]) can be used.
Acceleration – the forces due to acceleration (or deceleration) is dependent on
the acceleration rate and the mass of the train.
Thus, by specifying the speed and acceleration of the train, all the forces involved in
train movement can be obtained for each matrix element
Based upon the technical characteristics of the train, a load factor for the train type, a
speed-acceleration matrix, and the characteristics of the railway line, the forces acting
on the train are calculated for each cell of the speed-acceleration matrix. These are
weighted according to the distribution. The results per cell are summed and multiplied by
the link length to obtain the total energy consumption (at the wheels) over the line. The
power required is obtained by dividing this number by the efficiency of the drive system.
Emissions are calculated by multiplying the energy consumption in GJ (gigajoule37) with
emission factors (in g/GJ) for diesel engines for or electricity generation in the case of
electric trains. The same model can be used for individual trains on short routes, train
fleets on long routes, for single runs, or annual averages. To improve accuracy, gradient
resistance should also be included.
Passenger trains have other sources of energy consumption than propulsion. These
include the heating and cooling loads, and electrical power to operate lights, instruments
and other electrical accessories. The heating power varies significantly throughout the
year, and can be used to provide an order of magnitude estimate of auxiliary energy
consumption. The results indicate the relative consumption for auxiliary power is fairly
low, and is comparable to the uncertainties in the basic energy consumption and
emissions of passenger trains. Therefore, it was not incorporated into the ARTEMIS
model specifically, although information is provided to allow its importance to be
estimated.
Five types of train are currently included in the model. The model contains default
average speed-acceleration matrices for these train types, but matrices can also be
specified by the user. It was intended that a wider variety of trains would be included,
but the relevant information could not be obtained. The train types included are felt to
be representative of a wide spectrum of European trains (Lindgreen and Sorenson,
2005). In principle, the model allows various levels of regional differentiation. Calculating
regional emissions from rail transport requires the availability and input into the model
of region-specific data.
37 Joule - derived unit of energy or work in the International System of Units. It is equal to the energy
expended (or work done) in applying a force of one newton through a distance of one metre (1 newton metre
or N·m
59 PPR601
Appendix E Scenarios considered for emission assessment
Scenario 1: Change in traffic flows from 2010 to 2013 without any intervention
For this scenario, TEEM would be run with scaled traffic activity data in 2010 and 2013.
Traffic growth is assumed to be in line with national trip end forecasts provided by DfT‟s
transport model TEMPRO programme, v6.2.
Scenario 2: Meeting LIP targets for modal shift from 2010 to 2013/4 compared
to a baseline.
The Council will work with TfL and interested parties to enhance the public transport
system to encourage modal change towards sustainable travel. Travel by car is currently
nearly 73 percent with only 14 percent of the population using public transport to work38.
The Council aims to improve the bus service provision generally and on north south
routes in particular to increase 8.8% of work trips by bus to 10% in 2014. The Council
hosted two special public transport events in July and September 2010 offering local
representatives, stakeholders and transport providers an opportunity to assess the
current situation and identify topics for future liaison and development. The top three
issues that were identified are to improve transport links to Uxbridge, provide North-
South Sustainable Transport and to integrate public transport with neighbouring local
authorities. These issues will be taken forward by all the relevant partners including the
Public Transport Liaison Group.
Scenario 3: Reduction of cars on the Uxbridge Road.
This scenario affects only road transport and would look at the effect of a 1% reduction
in cars travelling on the Uxbridge Road.
Scenario 4: Dedicated bus service on the Uxbridge to Heathrow airport route.
This would involve transfer of 1% of car passengers to a dedicated bus service, with an
associated increase in bus occupancy.
Scenario 5: Improving transport interchanges to increase public transport use
One of the main theme‟s in LBH‟s LIP is to encourage modal shift as a result of
improving local transport interchanges. There are many interchanges in the borough
including Uxbridge, Hayes, Eastcote, South Ruislip, Ruislip Gardens and links along the
Grand Union Canal
38 http://www.hillingdon.gov.uk/media/pdf/k/a/EB52-_LBH_Local_Implementation_Plan__April_2011.PDF
60 PPR601
Uxbridge station is considered to be the largest and most significant interchange
between the Underground and bus services with 140 bus and 12 train services per peak
hour (over 5.4 million trips per year39).
This scenario would look at the impact of increasing passengers using existing services
by 1% on buses and trains and as a result cause a reduction in car passengers.
Scenario 6: Reduction in road traffic due to introduction of Crossrail.
Little evidence has been provided examining the likely effects of modal switch through
the introduction of Crossrail. However, a 2010 Summary Update Report on the benefits
of Cross Rail suggests a traffic reduction of 2% in London in general with higher benefits
accruing on parallel roads and in the Heathrow area in 2017. This scenario would look at
the emission impacts of Crossrail in Hillingdon by assuming a 2% reduction in traffic flow
on all modelled roads, with a further 2% reduction on the M4 and A4 corridors.
Scenario 7: Implementation of business travel plans that lead to modal shift.
It is difficult to establish what aspects of travel plans are quantifiable in terms of
emissions savings. The I-Trace database40 provides statistics which indicate year on year
trends which could be quantified and businesses are obligated to updating these plans
annually as part of their planning application. Travel plans generally relate to personal
employee travel.
Table D1 shows targets extracted from the proposed Slough International Freight
Exchange (SIFE) Framework Travel Plan (Spencer, 2010) which can act as a guide for
this scenario. For this development, the applicant agreed to pay financial penalties if
these targets were not met. The Table provides 2001 Census proportions shown for
comparison. Staff surveys would need to be conducted to set suitable working targets in
terms of car sharing rather than simply car passengers. These targets could then be
used to assess emissions changes on a person vehicle kilometre basis. This would
involve deriving the total number of employees and making assumptions concerning the
distance travelled by mode.
Table D1. Indicative targets for modal shift achieved by business travel plans.
Year Walk Bicycle Bus Rail Motorcycle Car
driver
Car
passenger
Census
(2001)
2.2% 1.7% 4.5% 1.4% 1.5% 81.5% 6.5%
1 3.0% 2.5% 6.0% 2.0% 2.0% 67.0% 17.0%
3 4.0% 5.0% 9.0% 3.0% 3.0% 60.0% 15.0%
39 Pre-Submission Core Strategy Consultation: Transport http://hillingdon-
consult.limehouse.co.uk/portal/planning/preub_core_strategy/consultation_feb_2011?pointId=1289317489473
40 Westtrans Travel Plan Monitoring (iTRACE and
TRAVEL)http://www.westtrans.org/wla/westtrans.nsf/pages/wt-17. Note that TfL has recently stopped its
funding of the i-Trace database.
61 PPR601
5 4.5% 6.0% 11.5% 3.5% 4.0% 55.0% 14.0
Scenario 8: Re-investigating the impact of major planning developments.
This scenario would re-assess the air quality assessments conducted for large planning
applications that were subject to an Environmental Statement. To do this, the changes in
emissions estimated in the original assessment could be compared with changes using
the EDB (assuming that all the necessary pre and post traffic data are provided). This
scenario could involve examining major schemes proposed in Yiewsley, West Drayton,
Hates or Ruislip Manor.
Scenario 9: Introduction of a north-south bus service and corresponding
reduction in car travel.
The Draft LIP reports that north to south road and public transport accessibility within
the borough is severely constrained. It states that;
„The current public transport provision in Hillingdon has an east-west orientation whilst
north-south journeys are usually slow and often indirect. Road traffic pressures across
the A40 create a major barrier severing the north from the south. Better north-south
links will help connect local people with jobs in the borough, for example connecting
Stockley Park to Hayes and West Drayton will make it a more accessible and sustainable
office location. It is vital that LIP2 addresses the challenge of poor north-south transport
links in Hillingdon.‟
Improving transport links between the north and south of the borough depends on
whether TfL will improve traffic management on the TLRN. However, for this scenario it
could be assumed that a continuous bus service is introduced on a selected route
between two urban centres (one in the north and one in the south). Four single decked
buses will continuously operate the route allowing for a maximum wait time of 20
minutes. The buses will be assumed to carry an average 20% of bus capacity. This will
equate to removing a certain amount of passenger cars from the route. The removal of
passenger cars allows journey speeds to be increased slightly. The test will compare this
with a baseline for a suitable year.
Scenario 10: Relieving traffic congestion from Heathrow villages
To assess this scenario, the first step would be to consider what traffic data are available
for any of the villages. As the WLTM output is peak hour, this could be used as an
indication of traffic flows at busy periods. Therefore, as a start, the data available could
be used to assess the baseline emission levels at selected village locations and compare
these at different locations. This could help develop and prioritise more targeted
measures to relieve traffic congestion.
Scenario 11: Overall impacts of generalised policies from the baseline.
To look at generalised policies, the baseline emissions could be manipulated to test
changes in the base year from a range of LIP policies, such as:
Reducing all traffic by 1%
62 PPR601
Reducing cars by 5%
Increasing buses by 5%
Reducing/increasing all peak vehicle speeds by 10%
Reducing all HGVs by 5%
Increasing LGVs by 2%
The scenario could include providing aggregated emissions to assist policy makers.
Scenario 12: Urban freight consolidation centre for Heathrow to reduce HGV
travel.
There are very few studies that have attempted to evaluate the emission impacts of
Urban Consolidation Centre. In Heathrow, the UCC received 20,000 vehicle deliveries
which resulted in 45,000 store deliveries being made from the centre on 5,000 vehicle
trips in 2004. 190 out of 240 of the retail outlets are using the centre. Vehicle trip
reduction of approximately 70% is being achieved for those goods that flow through the
centre. This was estimated to result in 87,000 vehicle kilometres saved in 2003, and
144,000 vehicle kilometres saved in 2004. Vehicle emissions reductions have also
increased as goods throughput has grown, with CO2 savings of 1,200 kg per week in
2003 and 3,100 kg per week in 2004.41
This scenario would re-examine the types of vehicles using the UCC and collect data
such as average trip distance and load factors (if possible) to assess the emission
benefits. The impact of policies to use cleaner vehicles could also be tested with the EDB.
41 BESTUFS Good Practice Guide on Urban Freight Transport
(http://www.bestufs.net/download/BESTUFS_II/good_practice/English_BESTUFS_Guide.pdf)
TRL Crowthorne House, Nine Mile Ride Wokingham, Berkshire RG40 3GAUnited Kingdom
T: +44 (0) 1344 773131 F: +44 (0) 1344 770356E: [email protected] W: www.trl.co.uk
ISSN 0968-4093
Price code: 3X
Published by IHS Willoughby Road, Bracknell Berkshire RG12 8FB United Kingdom
T: +44 (0) 1344 328038 F: +44 (0) 1344 328005E: [email protected] W: http://emeastore.ihs.com PP
R6
01
A transport emissions database for Hillingdon
The aim of this project was to produce an emissions database (EDB) for the London Borough of Hillingdon to assess changes in emissions and fuel consumption of measures in their Transport Local Implementation Plan (LIP) and Air Quality Action Plan. he EDB includes link-based road transport emissions and a rail module that can be used to determine the contribution of emission sources in the borough and to assess the emissions impact of selected measures from the LIP.
It was found that of the road and rail sources within the EDB that are within the borough, road traffic contributes to over 77 percent NO
x emissions and 69 percent of PM emissions. Of these
road traffic sources, vehicles travelling on the motorways contributed to over 60 percent of each pollutant. For CO
2, road transport contributed to 95 percent of the total emissions. Emissions from
the diesel railway network contributed 23 percent of overall NOx emissions, 31 percent to PM emissions and 5 percent of CO
2 emissions in the borough.
The EDB was also used to assess the impact of four measures from the LIP on emissions over a given area. These measures assessed were modal shift from cars to trains and buses, introducing a dedicated Uxbridge to Heathrow bus service, improvements to a bus and train transport interchange at Uxbridge and the introduction of Crossrail in 2017.
Other titles from this subject area
PPR490 The acoustic durability of timber noise barriers on England’s strategic road network. P A Morgan. 2010
PPR490 Technical Annex to PPR490 – The acoustic durability of timber noise barriers on England’s strategic road network. P A Morgan. 2010
PPR485 The performance of quieter surfaces over time. M Muirhead, L Morris and R E Stait. 2010
PPR432 A future ‘quiet HGV’ permissive certification scheme – phase 1 report. P A Morgan, M Muirhead, M J Ainge and P G Abbott. 2010
PPR394 An examination of the monetised benefit of proposed changes to type approved noise limits for tyres. M Muirhead, P G Abbott and M Burdett. 2009
