www.defra.gov.

uk

Local Authority CO2 emissions estimates 2006

Statistical Summary

18 September 2008
Published by:
Department for Environment, Food and Rural Affairs
Nobel House
17 Smith Square
London
SW1P 3JR Telephone
+44 (0) 20 7238 6609
Internet service http://www.defra.gov.uk/

© Crown copyright 2008

Copyright in the typographical arrangement and design rests with the Crown.
This publication (excluding the logo) may be reproduced free of charge in any format or
medium provided that it is reproduced accurately and not used in a misleading context.
The material must be acknowledged as Crown copyright with the title and source of the
publication specified.

Extracts of this publication may be made for non-commercial in-house use, subject to the
source being acknowledged. For a guide about Crown copyright please contact Defra
Communications Directorate and also: http://www.defra.gov.uk/copyright.htm

Further information and Applications for reproduction should be made in writing to:

Environment Statistics Service

Department for Environment, Food and Rural Affairs
5F Ergon House
17 Smith Square
London
SW1P 3JR

Email enquiries about this publication: enviro.statistics@defra.gsi.gov.uk

Published by the Department for Environment, Food and Rural Affairs.

Produced in the UK, September 2008 in electronic format.

This is a National Statistics publication.

National Statistics are produced to high professional standards set out in the National
Statistics Code of Practice. They undergo regularly quality assurance reviews to ensure
that they meet customer needs. They are produced free from any political interference.

2
 2006 National Statistics on Carbon Dioxide emissions at
Local Authority and Regional Level

Defra Statistical Summary

National Statistics of carbon dioxide emissions for local authority areas for 2006, and
revised figures for 2005, have been produced on behalf of Defra by AEA. This paper
explains the background to the estimates, summarises some of the results and
conclusions, and discusses some of the issues which need to be considered when
using the data.

Full details of the results and methodology are available in the AEA research report
and accompanying spreadsheet (see
http://www.defra.gov.uk/environment/statistics/globatmos/galocalghg.htm).

Contents
1. Introduction

2. Improvements since last year

3. Use of the estimates

4. Results

5. Reconciliation with the UK inventory

6. The future

1. Introduction

The UK compiles an annual inventory of its greenhouse gas emissions in order to monitor
progress against domestic and international targets such as the Kyoto Protocol.
Disaggregated versions of the UK inventory are also produced for England, Scotland, Wales
and Northern Ireland, and maps estimating the geographical distribution of the sources of
emissions are also produced.

Carbon dioxide (CO2) is the main greenhouse gas, accounting for about 85 per cent of the
UK total, and the vast majority of CO2 emissions come from the burning of fossil fuels. In
recent years, increasing emphasis has been placed on the role of regional and local
government in contributing to energy efficiency improvements, and hence reductions in CO2
emissions. The level of interest in local estimates has therefore increased accordingly.

This project combines data from the National Atmospheric Emissions Inventory (NAEI) with
local energy consumption data from the Department for Business, Enterprise and Regulatory
Reform (BERR) to produce a nationally consistent set of CO2 emission estimates down to
local authority level.

The statistics show emissions allocated on an “end-user” basis - the general principle here is
that emissions are distributed according to the point of energy consumption (or point of

3
emission if not energy related). Except for the energy industry, emissions from the production
of goods are assigned to where the production takes place – thus as with the national
inventories, emissions from the production of goods which are exported will be included, and
emissions from the production of goods which are imported are excluded.

The local authority statistics now include consistent figures for more than one year – 2005
and 2006. This, along with some improvements to the quality of the input data, mean that
they have now been classified as full National Statistics under the terms of the National
Statistics Code of Practice. The statistics published in 2005, 2006 and 2007 which estimated
the emissions in 2003, 2004 and 2005 respectively, were all labelled as “Experimental
National Statistics”, indicating that they were still under development. The revised estimates
for 2005 being published this year supersede those published in 2007.

2. Improvements since last year

Improvements in the 2006 and revised 2005 estimates compared with those previously
published for 2005 include:

Accuracy

This is the main area where there have been improvements since the 2005 estimates were
released in 2007. Accuracy considerations fall in two main areas:

1. Input datasets and sources that have been improved since the 2005 estimates were
produced:

• BERR have established their Local Authority (LA) gas and electricity consumption
data as National Statistics. These are a major input into the LA emissions estimates
(accounting for approximately half of total emissions), and have been upgraded from
experimental status. This has resulted from improved geo-coding of postcode and
address data associated with meter locations and reductions in the number of meters
where an “estimated annual consumption” is needed due to lack of readings.
Aggregated 2006 meter data corresponded well with UK based data in the Digest of
UK Energy Statistics (DUKES).

• The road transport emissions component of the model is estimated by combining fuel
consumption and emissions factors by vehicle type, with traffic flow data from DfT.
The estimates have been updated with some methodological improvements:

ƒ Improved fuel consumption factors, particularly for HGVs by using more

comprehensive haulier survey data from DfT. These improved factors have
resulted in the total aggregated emissions using this bottom up modelling are
within 3 per cent of the total indicated by national fuel sales figure in DUKES.
This is compared to as much as a 17 per cent difference (for diesel) last year.

ƒ Significant time has also been spent validating data on allocation of traffic counts
to the road network

It is acknowledged that there are still some limitations with this input – i.e. the same
fleet composition (proportions of different sizes and ages of vehicles) is assumed to
be consistent across the whole country, and it is still not known how much error
inadequate capture of fuel tourism (hauliers buying fuel abroad) may be introducing to
the model.

4
 • Accurate point level data on emissions from large installations such as power stations
or where industrial processes are regulated are available. For smaller industrial and
commercial activity, these emissions must be modelled and mapped using IDBR
employment data and BERR sectoral emissions which can be matched using SIC
codes. The maps have now been updated for the first time since 2005 along with
some methodological improvements:

ƒ Use of gas availability / network data to validate the emissions associated with
gas and aid the modelling of emissions from other fuels (i.e. use of these fuels is
more likely where gas is not available)

ƒ Smoke Control Area maps have also been produced to similarly validate maps of
emissions from coal consumption.

ƒ More accurate data on which industrial sectors use which kind of fuel have
improved

2. AEA have introduced new processes of database management, and quality assurance of
data and modelling processes that aim to remove error over and above that found in the
data inputs, and meet National Statistics quality protocols:

• Promoting accessibility and transparency through detailed metadata on spreadsheets

and comprehensive manuals

• Quality control through sense checking within and between tables by more than one
member of staff, and by comparing with other comparable data for consistency

• Uncertainty analysis to assess accuracy of individual components of the model and

peer review to ensure the most appropriate and up to date methods are used

The uncertainty analysis in the 2006 estimates report found that for the majority of LAs, the
percentage error was less than 2.5%. Some input datasets do have a high degree of error
but the largest sources of emissions tend to be those with the most accurate data (and those
to which improvements have been made since the publication of the original 2005 figures in
2007).

Comparability

The other key improvement to the estimates other than accuracy, is the development of a
consistent time series, not available in previous releases.

As such, the new methodology being employed to model the 2006 emissions will also be
used to revise the 2005 estimates. This will provide LAs with a new baseline for 2005 and
allow comparisons of the 2006 estimates with those for 2005. Future updates will also
incorporate any necessary revisions back to 2005, if and when further methodological
improvements are introduced.

3. Use of the estimates

The purpose of these estimates is to assist those using local emissions accounting as a tool
in developing emissions reduction strategies and for raising awareness of greenhouse gas
emissions as an issue. The estimates and methodological report should be useful to those
who are already working on local inventories, and encourage others to do more by providing
a useful starting point for further work.

5
Despite the important improvements made since last year, these estimates are not perfect.
They stretch the information available to the limit in order to provide estimates for each
authority. Some of the limitations of the 2006 estimates include:

• confidentiality constraints on data for some large electricity and gas customers which
either prevents their allocation to local authority boundaries, or introduces some
additional uncertainty into the allocation, which may have a significant impact on
results for a few authorities

• road transport emission estimates rely on national traffic statistics, and distribution of
traffic on minor roads has had to be imputed at local level from regional level data

• the local distribution of emissions from sources other than gas, electricity generation
or road transport (these residual sources are about 20 per cent of total emissions)
largely has to be estimated from proxy information such as population or employment
data

• some of the key sources used for mapping emissions do not cover the whole of the
UK, and therefore alternative methods have had to be used for authorities in Northern
Ireland, and in the case of domestic solid fuel use, Scotland as well.

Further detail on data quality and the methods used is available in the main report.

It is also very important to bear in mind that circumstances vary enormously between
authorities, and also that local authorities have relatively little influence over some types of
emissions. In some cases more accurate data may be obtainable from locally available
information, on which Defra would welcome feedback.

For all these reasons, these statistics should be interpreted with caution. However, used
with care they can provide help in setting priorities. In particular, the dataset is sufficiently
robust to set a baseline against which to monitor action on climate change at a local level.

It should be noted that the results at Government Office Region level are much more robust.
Most of the difficulties in allocating data to local authorities have little impact at regional level.
Problems of interpretation, such as economic activity or road transport taking place across
boundaries, still exist but are less acute at the regional level than at the local level.

4. Results

Estimates have been produced for each local authority in the UK of CO2 emissions from the
following broad source categories:

• industry, commercial & public sector (including electricity-related emissions)

• domestic (including electricity-related emissions)

• road transport

• land use, land use change and forestry

The level of detail is constrained by the detail available in BERR statistics on local electricity
and gas use. To estimate a further breakdown would have involved further general
assumptions about energy use for different sectors, since local data is not available.

6
However, further detail, mostly in terms of fuel types, is shown in the main report in order to
provide additional insight into how the estimates are constructed.

Because land use, land use change and forestry can act as a sink, removing carbon from the
atmosphere, as well as a source of carbon emissions, net emissions for this sector may
sometimes be negative. Because this sector is rather different in nature from others (for
which emissions are largely attributable to fuel consumption), it may be convenient to use
figures which exclude the land use, land use change and forestry sector.

Emissions in 2006

Table 1 shows a summary of the end user results for Government Office Regions. Estimates
are also shown per resident, based on the Office of National Statistics (ONS) mid-year
population estimates for 2006, in order to make some allowance for the different size of
regions. However, it should be noted that while emissions per resident may be a useful
measure for domestic emissions, emissions from industry and road transport are driven by
many factors other than resident population. Therefore industrial and commercial and road
transport emissions per resident should be interpreted with caution.

Results for individual local authorities can be found in the spreadsheet supplied with the main
report.

There is a great deal of variation between local authorities. In particular a significant amount
of industrial emissions are concentrated in a few areas, so for most local authorities the
contribution of industrial and commercial emissions in 2006 was lower than the overall
averages in Table 1 suggest.

Table 1: End user carbon dioxide emissions 2006: Government Office Region summary

Total emissions Per capita emissions

(million tonnes carbon dioxide) (tonnes carbon dioxide per resident)
Industrial, Industrial,
Government commercial Road LULUCF commercial Road LULUCF
Office Region & public Domestic transport Total & public Domestic transport Total
North East 21 6 5 -0 33 8.4 2.5 1.9 0.0 12.7
North West 27 17 15 1 59 3.9 2.5 2.2 0.1 8.7
Yorkshire and
the Humber 28 13 12 0 53 5.5 2.5 2.3 0.1 10.4
East Midlands 18 11 11 0 41 4.2 2.5 2.6 0.1 9.3
West Midlands 19 13 13 0 45 3.5 2.4 2.4 0.1 8.4
East of England 17 14 14 1 45 3.0 2.5 2.5 0.1 8.1
Greater London 22 17 11 0 50 2.9 2.3 1.4 0.0 6.6
South East 25 21 20 -0 66 3.0 2.6 2.4 0.0 8.0
South West 17 13 12 1 42 3.3 2.5 2.3 0.2 8.3
England (1) 193 126 113 3 434 3.8 2.5 2.2 0.1 8.6
Wales (1) 19 8 6 -0 33 6.4 2.6 2.2 -0.1 11.1
Scotland (1) 22 14 11 -4 43 4.3 2.8 2.2 -0.8 8.5
N. Ireland (1) 6 6 5 -0 16 3.2 3.5 2.7 -0.1 9.4
UK (1) 245 154 135 -2 532 4.0 2.5 2.2 0.0 8.8
% of UK total 46.1% 28.9% 25.4% -0.4% 100%

(1) Sum of local authority emission estimates differ from the official inventory for the UK (see Reconciliation section below)
(2) These data are not comparable with the inventories for the Devolved Administrations, which are compiled on an ‘at
source’ basis rather than ‘by end user’

7
Table 2 shows the number of LAs with different proportions of emissions coming from the
different sectors. This shows more clearly the wider distribution in the proportion of
emissions caused by industrial and commercial activity – i.e. it accounted for over half of
emissions in 74 LAs whereas domestic emissions only accounted for over half in 8 and road
transport for over half also in 8. In the majority of cases, industry and commerce made up
30-50% of emissions, while domestic and road transport were usually the source of 20-40%
each.

Table 2: End user carbon dioxide emissions 2006: Proportion of emissions by sector
frequency table

Proportion of emissions Sector (number of LAs where sector accounts for corresponding proportion of emissions)
Industrial and commercial Domestic Road transport
Less than 20% 5 42 101
20-30% 88 127 150
30-40% 169 183 121
40-50% 98 74 54
50-60% 41 8 8
More than 60% 33 0 0
Total 434 434 434

Emissions per resident vary least between areas for the domestic sector, and are dominated
by gas and electricity consumption, for which real local data are available. Therefore these
are perhaps the most interesting results, and although there are still minor problems with
data quality, local fuel and energy use statistics produced by BERR continue to improve in
accuracy and are now classed as National Statistics. Domestic emissions here represent
emissions from energy consumption in and around the home, including emissions
attributable to the use of electricity, but not activities by private individuals elsewhere, such
as personal travel.

Domestic emissions are important. In 38 per cent of local authorities, domestic end user
emissions were greater than industrial and commercial end user emissions, and in a further
27 per cent of authorities the domestic end user emissions were at least three quarters of
industrial and commercial end user emissions. This is little changed compared with 2005.

In 2006, domestic end-user emissions were less than 2 tonnes per person in 1 per cent of
LAs, between 2 and 2.5 tonnes per person in 37 per cent, between 2.5 and 3 tonnes per
person in 52 per cent and above 3 tonnes per person in 9 per cent. This can be influenced
by the fuel types used, the type and condition of the housing (including its insulation), the
average temperature (urban areas can be much warmer and therefore easier to heat than
rural areas), average household size, type of household and the income and preferences of
the occupiers.

In 2006, about 47 per cent of domestic end user emissions arose from gas use, 43 per cent
from electricity, and 10 per cent from consumption of other fuels. This changed little since
2005, and the pattern varies locally with the availability of different fuel types.

Road transport emissions include both freight and passenger transport, both private and for
business purposes. The estimates are made on the basis of the distribution of traffic,
therefore some of the emissions within an authority represent through traffic, or part of trips
into or out of the area whether by residents or non-residents. In some authorities this can be
particularly significant, and the issue has to be borne in mind when looking at either totals or
per capita estimates. The main report shows how the estimates break down between major

8
and minor roads, to help with consideration of this point. On the end user basis, road
transport emissions include a share of emissions from oil refineries.

Changes between 2005 and 2006

When the LA emissions are aggregated, estimated total emissions remained about the same
– falling slightly from 532.4 million tonnes to 531.7 million tonnes between 2005 and 2006.
Table 3 compares 2005 with 2006 at the regional level. Most regions have seen little change
although there is some variation – the largest decrease in per capita terms of 0.7 tonnes per
person was seen in the North East and the largest increase, of 0.3 tonnes per person, was
seen in Northern Ireland.

Table 3: End user carbon dioxide emissions 2005 and 2006: Government Office
Region comparison

Region / country 2005 2006

Total emissions Per capita Total emissions Per capita
(millions of tonnes) (millions of tonnes)
North East 34 13.4 33 12.7
North West 59 8.8 59 8.7
Yorkshire and the Humber 53 10.5 53 10.4
East Midlands 40 9.4 41 9.3
West Midlands 44 8.2 45 8.4
East of England 45 8.1 45 8.1
Greater London 48 6.4 50 6.6
South East 66 8.1 66 8.0
South West 41 8.2 42 8.3
England (1) 430 8.5 434 8.6
Wales (1) 35 11.7 33 11.1
Scotland (1) 43 8.5 43 8.5
N. Ireland (1) 16 9.1 16 9.4
UK (1) 532 8.8 532 8.8

(1) Sum of local authority emission estimates differ from the official inventory for the UK (see Reconciliation section below)
(2) These data are not comparable with the inventories for the Devolved Administrations, which are compiled on an ‘at
source’ basis rather than ‘by end user’

At the Local Authority level, there was a larger variation by LA with some areas estimated
emissions increasing considerably, and some decreasing considerably:

• 42 LAs experienced an increase of more than 5% - maximum increase of 17% was

seen in Tower Hamlets.
• 35 LAs experienced a decrease of more than 5% - maximum decrease of 26% was
seen in Ribble Valley
• In absolute terms 21 LAs saw an increase of more than 100 thousand tonnes, and a
further 21 saw a decrease of more than 100 thousand tonnes.

9
Changes by sector overall

In absolute terms, the 3 largest increases were in the following sectors:

• Industry and commercial electricity (5.8 million tonnes)

• Domestic electricity (2.7 million tonnes)
• Industrial and commercial non-fuel (1.4 million tonnes)

During 2006, the fuel mix at the national level used to generate electricity resulted in greater
emissions per unit of electricity than in 2005. Therefore the factor applied to all electricity
use to convert that use into emissions was larger for 2006. This is the main driver behind the
increased emissions attributable to industry and commercial, and domestic electricity use.

The 3 largest decreases were in the following sectors:

• Industry and commercial gas (-3.5 million tonnes)

• Industry and commercial large gas users (-2.2 million tonnes)
• Industry and commercial solid fuel (-1.6 million tonnes)

Changes by sector at the Local Authority level

As discussed above, several LAs have seen large changes between 2005 and 2006. Taking
the ten LAs that have seen the biggest changes in each direction, an assessment was made
of which specific sub-sectors were driving the changes so they could be individually verified.

Table 4: LAs that saw large increases between 2005 and 2006

Local Authority Percentage Sector(s) responsible (proportion of all increases in that area)
increase
Tower Hamlets 17 Industrial and commercial electricity (>90%)
South Gloucestershire 14 Industrial and commercial large gas users (82%)
Gravesham 13 Industry and commercial non-fuel (47%) and electricity (34%)
Scottish Borders 13 Land use change (65%)
Newport 13 Industry and commercial process gases (29%), solid fuel (27%) and non-
fuel (22%)
Middlesbrough 13 Industrial and commercial large gas users (84%)
Tynedale 13 Land use change (56%)
St. Edmundsbury 12 Industrial and commercial non-fuel (53%) and solid fuel (27)
Moray 12 Industry and commercial gas (46%) and land use change (22%)
Telford and Wrekin 11 Industry and commercial non-fuel (39%) and electricity (23%)

For other LAs not mentioned in Table 4 that have seen a large increase, the major
contributors are similarly industrial and commercial emissions from electricity, solid fuel, oil
and other non-fuel related activity. In a few cases, land use change is also a major
contributor to the increase.

10
Table 5: LAs that saw large decreases between 2005 and 2006

Local Authority Percentage Sector(s) responsible (proportion of all decreases in that area)
decrease
Ribble Valley -26 Industry and commercial non-fuel (45%) and solid fuel (43%)
Copeland -24 Industry and commercial gas (>90%)
Neath Port Talbot -21 Industry and commercial process gases (49%) and solid fuel (39%)
Exeter -21 Industry and commercial gas (>90%)
Hartlepool -15 Industry and commercial large gas users (51%) and gas (45%)
Stockton-on-Tees -15 Industry and commercial large gas users (72%)
Argyll and Bute -15 Industry and commercial oil (69%)
Redcar and Cleveland -14 Industry and commercial process gases (72%)
Clackmannanshire -14 Industry and commercial oil (>90%)
Vale of Glamorgan, The -14 Industry and commercial large gas users (75%)

For other LAs not mentioned in Table 5 that have seen a large decrease, the major
contributors are similarly industrial and commercial emissions from gas (large users and
more generally) and solid fuel. In a few cases, land use change is also a major contributor to
the decrease. Domestic gas, which has seen a relatively large decrease overall, is more
evenly spread with most areas seeing a small decrease (0-5%) and hence tends to make up
a smaller proportion of the total decrease in areas where there has been a large fall in
emissions.

5. Reconciliation with the UK inventory

These local estimates are designed to be as consistent as possible with the national
inventory for the UK. However, some differences are unavoidable.

A number of emission sources included in the UK inventory are not included in the local
estimates, as there is no obvious basis for doing so. Excluded sources are principally linked
to aviation and shipping.

A small proportion of the gas and electricity consumption allocated to the domestic sector in
these estimates would be attributed to business in the UK inventory. This is because it is
impossible to distinguish between domestic customers and smaller businesses in the meter
point consumption data used in these local estimates.

Table 6 shows a summary of the reconciliation between the UK inventory and the local
inventory. The different elements of this reconciliation should be interpreted as follows:
“Unallocated” emissions are those which would have been allocated to local areas if relevant
and suitable geographical information had been available for this purpose; “Excluded”
emissions are those which have been deliberately excluded from the local level allocation, as
it would not have been appropriate to include them; and “Statistical differences” are
differences which have become apparent due to the different methodological approaches
used in deriving the UK inventory and the local level estimates. More detail can be found in
the main report.

11
Table 6: Reconciliation of 2006 local emission estimates with UK
inventory

(million tonnes CO2)

Details Totals

End user emissions allocated to local areas 527.1

Unallocated
unallocated consumption (largely industrial
1.0
and commercial electricity)
large electricity users with unknown location 3.6
Total unallocated 4.6

Total UK end user emissions (local method) 531.7

Excluded from local allocation:

domestic shipping 6.1
domestic aviation 2.6
military transport 3.1
fuel exports 10.1
international aviation and shipping 5.1
Total excluded 27.1

Methodological differences:
higher domestic in local method -4.6
lower industry & commercial in local method +0.4
Total methodological differences -4.2

UK Greenhouse Gas Inventory total 554.6

6. The future

It is intended that further sets of these National Statistics will be produced annually. There is
also a commitment to maintain a comparable time series starting in 2005 into the future, so
the series for each year back to 2005 will be reviewed each year to ensure consistency.

12
Comments

We would welcome comments on these statistics. These should sent to:

ESS4
Environment Statistics Service
Defra
Area 5F
Ergon House
17 Smith Square
London
SW1P 3JR
e-mail: enviro.statistics@defra.gsi.gov.uk

Useful links

Details of BERR’s programme of work on estimates of local and regional energy statistics are
available here:

http://www.berr.gov.uk/energy/statistics/regional/index.html

The home pages of the UK National Atmospheric Emissions Inventory are here:

http://www.naei.org.uk

13