Category Title

General guidance Spatial mapping of emissions

Version Guidebook 2016

Lead authors
Nele Veldeman, Wim van der Maas

Contributing authors (including to earlier versions of this chapter)

John Van Aardenne, Justin Goodwin, Katarina Mareckova, Martin Adams, Paul Ruyssenaars, Robert
Wankmüller, Stephen Pye

EMEP/EEA air pollutant emission inventory guidebook 2016 1

 7. Spatial mapping of emissions

Contents
1 Introduction ...................................................................................................... 3
2 Terminology ...................................................................................................... 4
2.1 General terms .......................................................................................................................... 4
2.2 Geographic features ................................................................................................................ 5

3 Methods for compiling a spatial inventory ..................................................... 7

3.1 Compiling point source data .................................................................................................... 9
3.2 Computing diffuse emissions ................................................................................................. 11
3.3 Distributing diffuse emissions ................................................................................................ 11
3.4 Combining different spatial features ...................................................................................... 32

4 Finding key spatial data sources................................................................... 33

4.1 General .................................................................................................................................. 33
4.2 National datasets ................................................................................................................... 34
4.3 International datasets ............................................................................................................ 35

5 Overview of available spatial emissions data (updated 2013) ..................... 39

6 References ...................................................................................................... 40
7 Point of enquiry .............................................................................................. 41

EMEP/EEA air pollutant emission inventory guidebook 2016 2

 7. Spatial mapping of emissions

1 Introduction
The aim of this ‘Spatial emission mapping’ chapter is to further elaborate on the gridding of emissions as
to:
 Support the link between emission data and air quality models that need emissions information
at a proper spatial, temporal and sectoral resolution;
 Facilitate countries in (improving) the gridding of their emission inventories of air pollutants under
the United Nations Economic Commission for Europe (UNECE) Convention on Long-range
Transboundary Air Pollution (CLRTAP).

Sub-national spatial emissions are increasingly important because:

 reported spatial emissions data are an input for models used to assess atmospheric concentrations
and deposition, as the spatial location of emissions determines to a great extent their atmospheric
dispersion patterns and impact area. The results of model assessments inform national and
international policies used to improve the environment and human health;

 regular reporting of spatial emissions is required under the Emission Reporting Guidelines for Parties
to the LRTAP Convention.

This chapter provides guidance on compiling spatial emissions datasets. It focuses on methods suitable
for generating and reporting spatial data required under the LRTAP Convention that are consistent with
nationally-reported inventories under CLRTAP.

This chapter starts with the definition of terms used when dealing with spatial datasets (Section 2). In
Section 3 a generic set of methodologies for deriving spatial datasets from national emissions inventories
is established. A sector-specific tiered approach for estimating spatial emissions is discussed, moreover,
some sector-specific issues are dealt with. Furthermore, approaches to combining spatial datasets are
presented to enable the inventory compiler to derive an aggregated spatial dataset combining sectoral
emissions into a unified gridded dataset like that needed for European Monitoring and Evaluation
Programme (EMEP) reporting. All methods rely on the identification and use of important spatial datasets.
Therefore, generic data sources for this type of data are outlined in Section 4. In Section 5 an overview of
available spatially disaggregated emission inventories, which could be used as an example, is given.

When preparing spatial data for reporting under EMEP, this chapter should be used in conjunction with
the EMEP Reporting Guidelines (1). These guidelines define the reporting requirements for spatially
resolved data.

The Reporting Guidelines give requirements concerning:

 the EMEP grid (0.1° x 0.1° longitude/latitude);

 the sectoral definitions for gridded and large point sources;
 additional large point source information requirements, e.g. height class;
 the required pollutants (main pollutants, PM, Pb, Cd, Hg, PAHs, HCB, dioxins/furans and PCBs);
 years for which reporting of gridded data is required.

(1) The Reporting Guidelines are available on the CEIP website (http://www.ceip.at)

EMEP/EEA air pollutant emission inventory guidebook 2016 3

 7. Spatial mapping of emissions

2 Terminology
2.1 General terms

CEIP - EMEP Centre on Emission Inventories and Projection (http://www.ceip.at).

CLRTAP - Convention on Long-range Transboundary Air Pollution.

Diffuse sources - Diffuse sources of a sector are defined as the national total of a sector minus the
reported point sources. This definition is in agreement with the definition used in E-PRTR (see below) and
implies that diffuse sources may contain (non-reported) point sources, line sources, and area sources.

EMEP - the Cooperative Program for Monitoring and Evaluation of the Long Range Transmission of Air
Pollutants in Europe.

EMEP grid - the EMEP grid is the geographical extent covering the EMEP area at a resolution of 0.1° ×
0.1° longitude-latitude in the WGS84 geographic coordinate system. The domain covers the geographic
area between 30°N-82°N latitude and 30°W-90°E longitude.

E-PRTR - E-PRTR is the European Pollutant Release and Transfer Register, established under EU
Regulation (EC) No 166/2006 of the European Parliament and of the Council of 18 January 2006, and is
intended to fully implement the obligations of the UNECE PRTR Protocol.

GIS - Geographical Information Systems.

HDV - Heavy Duty Vehicles, vehicles with a gross vehicle weight of > 3 500 kg.

IPPC - Integrated Pollution Prevention and Control. This EU Directive (the ‘IPPC Directive’) imposes a
requirement for industrial and agricultural activities with a high pollution potential to have a permit which
can only be issued if certain environmental conditions are met, so that the companies themselves bear
responsibility for preventing and reducing any pollution they may cause. More recently, the Directive on
Industrial Emissions 2010/75/EU (IED) was adopted by the European Union. The IED replaces the IPPC
Directive and a number of other sectoral directives as of 7 January 2014, with the exception of the LCP
Directive, which will be repealed with effect from 1 January 2016.

LCPD - Large Combustion Plant Directive: Directive 2001/80/EC of the European Parliament and of the
Council of 23 October 2001 on the limitation of emissions of certain pollutants into the air from large
combustion plants.

LDV - Light Duty Vehicles, vehicles with a gross vehicle weight of ≤ 3 500 kg.

NACE - Classification of Economic Activities in the European Union.

NFR - Nomenclature For Reporting.

NUTS - Nomenclature of Units for Territorial Statistics, which is a hierarchical classification of

administrative boundaries developed by Eurostat. The idea behind NUTS is to provide a common
designation for different levels of administrative geographic boundaries across the EU regardless of local
language and naming conventions.

SNAP - Selected Nomenclature for sources of Air Pollution — developed as part of the CORINAIR project
for distinguishing emission source sectors, sub-sectors and activities.

Surrogate spatial dataset - a geographically resolved dataset of statistics by grid, link, point or boundary
such as land use coverage percentage by grid, vehicle flow by road link, employers number by industrial
point, population by administrative boundary. Applied as alternative data source to spatially allocate
emissions, when direct spatial information on the emission source is not available.

EMEP/EEA air pollutant emission inventory guidebook 2016 4

 7. Spatial mapping of emissions

2.2 Geographic features

Geographical features will be used to represent emission sources. These features will define the
geographical structure of the spatial dataset.

Point sources: A point source is an emission source at a known location, represented by x and y
coordinates that indicate the main point of emission. Examples of point sources are industrial plants or
power stations.

Emissions from point sources represent sectors of a national inventory either fully (e.g. often for power
stations where the sector is made up of only large sites for which emissions reporting is mandatory) or in
part (e.g. such as combustion in industry, for which only the large sites within the sector are typically
required to report emissions). In the latter case, the remainder of the emissions for the sector are mapped
as an area source.

Large point sources (LPS): LPS are defined in the UNECE reporting guidelines (1) as facilities
whose combined emissions, within the limited identifiable area of the site premises, exceed
certain pollutant emission thresholds. Note: although stack height is an important parameter for
modelling emissions, it is not a criterion used for selecting LPS.

Area sources: An area source is an emission source that exhibits diffuse characteristics. For example,
sources that are too numerous or small to be individually identified as point sources or from which
emissions arise over a large area. This could include forests, residential areas and
administrative/commercial activities within urban areas.

Area sources as polygons: area polygons are often used to represent data attributed to
administrative or other types of boundaries (data collection boundaries, site boundaries and other
non-linear or regular geographical features).

Residential fuel combustion is an example of a sector that can be represented in this way, using
population census data mapped using the polygons defining the data collection boundaries.

Polygons are vector- (line-) based features and are characterised by multiple x, y coordinates for
each line defining an area. Examples of areas defined by polygons are the regions as defined by
the NUTS classification (Nomenclature of Units for Territorial Statistics). According to this
classification, the economic territory of the EU is divided in several zones, presented by polygons:

 NUTS 1: major socio-economic regions

 NUTS 2: basic regions for the application of regional policies
 NUTS 3: small regions for specific diagnoses

EMEP/EEA air pollutant emission inventory guidebook 2016 5

 7. Spatial mapping of emissions

Area sources as grids: area sources can be represented in a regular grid of identically-sized
cells (either as polygons or in a raster dataset). The spatial aspects of grids are usually
characterised by geographical coordinates for the centre or corner of the grid and a definition of
the size of each cell.

Agricultural and natural emissions sectors can be represented using land use data derived from
satellite images in raster format.

Grids are often used to harmonise datasets: point, line and polygon features can be converted
to grids and then several different layers of information (emission sources) can easily be
aggregated together (see section 3.4).

Line sources: A line source is an emission source that exhibits a line type of geography, e.g. a road,
railway, pipeline or shipping lane. Line sources are represented by vectors with a starting node and an
end node specifying an x, y location for each. Line source features can also contain vertices that define
curves between the start and end reference points.

EMEP/EEA air pollutant emission inventory guidebook 2016 6

 7. Spatial mapping of emissions

3 Methods for compiling a spatial

inventory
This section provides general guidance on the approaches to deriving spatially resolved (e.g.
sub-national) datasets. First, an introduction on general good practice is given. Consequently, a concrete
approach is outlined. Hereto, a schematic overview of consecutive steps is provided (see Figure 3-1). The
different steps as outlined in the scheme are then discussed in more detail in separate paragraphs.

It is good practice to consider the elements below when defining an efficient spatial distribution project.

1. Use key category analysis (see Chapter 2, Key category analysis and methodological choice) to
identify the most important sources and give the most time to these.
2. Make use of GIS tools and skills to improve the usefulness of available data. This will mean
understanding the general types of spatial features and possibly bringing in skills from outside the
existing inventory team for the production/manipulation of spatial datasets.
3. Make use of existing spatial datasets and carefully consider the merits versus costs of extensive new
surveying or data processing to derive new spatial datasets. It is often more important to generate a
timely dataset based on less accurate data than a perfect dataset that means reporting deadlines are
missed or all resources are consumed.
4. Select the surrogate data that is judged to most closely represent the spatial emissions patterns and
intensity, e.g. for combustion sources, surrogate spatial datasets that most closely match the spatial
patterns of fuel consumed by type should be chosen.
5. Surrogate spatial datasets that are complete (cover the whole national area) should be preferred.
6. Use, when possible and when no other more accurate data is available, the spatial surrogate that
was used for spatial mapping in previous years. This is to guarantee consistency.
7. Issues relating to non-disclosure may be encountered (at a sectoral or spatial level) that may impose
barriers to acquiring data (e.g. population, agriculture, employment data). As only highly aggregated
output data is needed for reporting, signing of non-disclosure or confidentiality agreements or asking
the data supplier to derive aggregated datasets may improve the accessibility of this data. It is
important that issues relating to this are identified and dealt with in consultation with the national
statistical authority.
8. It is advisable to consider the resolution (spatial detail) required in order to meet any wider national
or international uses. Aggregation to the present EMEP 0.1 x 0.1 degree longitude/latitude grid could
be done, for example, from more detailed spatial resolutions that might be more useful in a national
context. Most nationally reported emissions datasets are based on national statistics and are not
resolved spatially in a manner that could be readily disaggregated to the required 0.1 x 0.1 degree
EMEP grid. Possible exceptions in some countries are detailed road transport networks and reported
point source emissions data..
9. When updating a spatial inventory it is often not possible to update all the spatial datasets every year
(for economic reasons). A yearly data acquisition plan (DAP) can describe which surrogate data is
updated with which frequency, depending on its importance, costs and variation in time.
10. When the budget is very limited the available datasets in section 5 can act as a starting point when
they are used as a surrogate data for the spatial allocation of the national total for some sectors. The
limited resources can then be used for the most relevant sectors.
Different methods for compiling spatially resolved estimates can be used depending on the availability of
data. However, the general approach always contains the same basic steps. Therefore, a general scheme
can be followed. The schematic overview is presented in Figure 3-1.

EMEP/EEA air pollutant emission inventory guidebook 2016 7

 7. Spatial mapping of emissions

Figure 3-1 General approach for compiling a spatial emission inventory

National reported point source emissions

(E-PRTR Annex I activities)

1. Compile emissions for point sources National reported emissions dataset

See § 3.1 (NFR 09 categories)

2. Compute diffuse emissions

by subtracking the emissions from the
facilities from the national totals
See § 3.2

All emissions can be Consistency check:

allocated to point No diffuse emissions > 0? Yes
locations

3. Spatially disaggregate the diffuse emissions

See § 3.3

Spatially resolved emission map Spatially resolved emission Spatially resolved emission
of point sources map of line sources map of area sources

4. Combine different spatial features

See § 3.4

Spatially disaggregated
emissions map

First, an emissions inventory for point sources should be compiled. Hereto, a number of different data
sources are available. In general, for many European countries, the best starting point is the European
Pollutant Release and Transfer Register (E‐PRTR) database, which has been established by Regulation
166/266/EC from 18 January 2006. It should be noted that E-PRTR is based on data officially submitted
by national competent authorities, and so internal national contact points will be able to provide point
source data. In other countries, many industrial point sources and their emission reports are similarly
available through the relevant national or regional competent authorities, especially for those countries
that are Parties to the UNECE Aarhus Convention PRTR protocol. In order to obtain a comprehensive
point source emissions inventory, the E-PRTR or other point source emissions should however be
combined with emissions stemming from point sources that are regulated but for which there are no annual
emissions reporting requirements and from point sources for sites or pollutants not reported or regulated.
In section 3.1 a specific method for compiling point source data is described.

In a second step, the shares of releases from diffuse sources should be determined. The point source
emissions compiled in the first step should not exceed national total emissions reported under LRTAP
Convention, which include all anthropogenic emissions occurring in the geographical area of the country
(large point sources, linear and area sources). However, due to inconsistent reporting under different
reporting obligations and due to different sector classification systems, point source emissions might
exceed the national totals. Therefore, linking national point source data and national total emissions turns
out to be a challenging task: computation of diffuse emissions is not as straightforward as a simple
subtraction per sector. In section 3.2 some guidance to develop a subtraction methodology for diffuse
emissions to air is outlined.

Once the national diffuse emission totals are determined per source category, a gridding methodology for
each category should be developed. National emission estimates will need to be distributed across the
national spatial area using a surrogate spatial dataset, according to a common basic principle which can

EMEP/EEA air pollutant emission inventory guidebook 2016 8

 7. Spatial mapping of emissions

be presented in a straightforward formula that is referring to the specific surrogate spatial dataset. The
methods used can range in quality from Tier 3 to Tier 1 depending on the appropriateness of the spatial
activity data being used. An extensive description of the basic principle and the different methodologies
is provided in section 3.3. Furthermore sectoral guidance is given and some examples are listed.

Finally, different spatial features need to be combined, in order to obtain a spatially disaggregated
emissions map. Information on how this could be done is provided in section 3.4.

3.1 Compiling point source data

Emissions for point sources can be compiled using a number of different data sources and techniques.
For convenience, the point source data can be divided into three groups.

1. Regulated point sources such as those regulated under the Integrated Pollution Prevention and
Control (IPPC) Directive regulatory regime and/or where there is a requirement for centralised
annual emissions reporting (e.g. for E-PRTR/the Large Combustion Plant (LCP) Directive);
2. Point sources that are regulated but for which there are no annual emissions reporting
requirements (e.g. E-PRTR does not cover all point source emissions as it uses emission
thresholds, emissions below the specified threshold are not included);
3. Point sources for sites or pollutants not reported or regulated.

To obtain a detailed point source emission data set, point source emissions of all three groups should be
combined.

First the regulated point sources with requirement for reporting should be considered. As outlined in
the introduction, the best starting point is the E‐PRTR or equivalent national database. Such data can be
used directly: emission data are known at exact locations, represented by x and y coordinates indicating
the main point of emission on the site. As such there is no need to further spatially disaggregate the data
to obtain spatially resolved emission maps of point sources per specific sector.

The E-PRTR or equivalent national data represent the total annual emission releases during normal
operations and accidents. For E-PRTR, releases and transfers must however be reported only if the
emissions of a facility are above the activity and pollutant thresholds set out in the E-PRTR Regulation.
Therefore, sources may not need to report emissions if these are below a specified reporting threshold or
reporting is not required for the specific activity undertaken at a facility. Consequently point source
emissions from smaller plants or from specific activities might not be included in the E-PRTR database.

Emissions from regulated point sources without annual emissions reporting requirements can
however often be estimated based on centralised data on process type and/or registered capacities and
initialisation reports associated with the original application for emission permits. Estimating point source
emissions for non E-PRTR sources and representing them with x and y coordinates according to their
exact location, results in spatially resolved maps of the smaller point sources.

In some cases, data sets are not complete. Furthermore, some point sources are not regulated. Emissions
from not reported or non-regulated sources can be modelled by distributing national emission
estimates over the known sources on the basis of capacity, pollutant correlations with reported data (e.g.
particulate to PM10/PM2.5) or some other 'surrogate' statistic, such as employment. The following box
(Example 1) provides some examples of approaches used to derive emissions for point sources in the
absence of reported data.

EMEP/EEA air pollutant emission inventory guidebook 2016 9

 7. Spatial mapping of emissions

EXAMPLE 1: ESTIMATING POINT SOURCE EMISSIONS FOR SOURCES/POLLUTANTS THAT ARE NOT REPORTED
In some cases, datasets are not complete. Furthermore, some point sources are not regulated. In these
cases, point source data is generated using national emission factors and some ‘surrogate’ activity
statistic. Examples of approaches used are given below.
 Estimates of plant capacity can be used to allocate the national emission estimate. This approach
can be used, for example, for bread bakeries where estimates of the capacity of large mechanised
bakeries can be made or gathered from national statistics or trade associations.
 Emission estimates for one (reported) pollutant can be used to provide a weighted estimate of the
national emission estimate of another pollutant. For example, emissions of PM10 from certain coating
processes can be estimated by allocating the national total to sites based on their share of the
national VOC emission.
 Deriving point source estimates based on pollutant ratios can be used to fill gaps in reported
emissions data. In some cases known PM10/PM2.5 ratios can be established to estimate emissions
for PM10 and PM2.5 for similar processes. Where no other data is available, other pollutants, such as
NOx and SO2, can be used to distribute other pollutant emissions.
 Assuming that all plants in a given sector have equal emissions: in a few cases where there are
relatively few plants in a sector but no activity data can be derived, emissions can be assumed to be
equal at all of the sites.
With the possible exception of using plant capacity, many of the approaches listed above will yield
emission estimates that are subject to significant uncertainty. However, most of the emission estimates
generated using these methods are, individually, relatively small and the generation of point source data
by these means is judged better than mapping the emissions as area sources.

Finally, the obtained point source inventories and maps of the three different groups should be combined.
It is therefore recommended to compile the different data sets based on the same sector classification. In
principle any classification can be chosen initially, however, it is advisable to consider the categories
required under different reporting obligations when compiling the point source emission data. The derived
point source datasets should be structured such that it is possible to differentiate point source emissions
into the relevant reporting sectors.

According to E-PRTR regulations, point source emissions have to be reported in categories covering 65
economic activities within 9 different industrial sectors. In order to allow computation of diffuse emissions,
the E-PRTR or equivalent national LPS emission data will however need to be reconciled with the national
totals and sectoral definitions in the inventory as reported under CLRTAP. Hereto, the data will need to
be classified into process or Nomenclature For Reporting (NFR) categories (see section 3.2). Therefore
it might be advisable to model and/or estimate non reported emissions (group 2 and 3) using the E-PRTR
and/or NFR categories as well. Also, it might be worth to consider any other national or international uses,
before deciding on the sector classification. The following link maps different classifications from the
various reporting obligations:
http://www.ceip.at/fileadmin/inhalte/emep/xls/Spreadsheet_for_reporting_formats.xls.

As outlined in the introduction, the purpose of spatial emission mapping is twofold. On one hand, countries
are obliged to report gridded emissions under the LRTAP Convention. On the other hand, spatial emission
maps are crucial input to air quality models. With respect to the latter, an important note should be made.
For modelling purposes, source characteristics such as stack height, source diameter and source heat
capacity are important parameters. Nevertheless these characteristics are not required under most
reporting obligations, it is highly recommended to include them when compiling the national point source
emission inventory for modelling purposes.

EMEP/EEA air pollutant emission inventory guidebook 2016 10

 7. Spatial mapping of emissions

3.2 Computing diffuse emissions

Point source emissions, as compiled in the first step, represent sectors of a national inventory either fully
or partly. In the latter case the remaining emissions are to be considered as diffuse emissions. Therefore
a methodology to distinguish the point-source releases and diffuse releases needs to be developed.

Review of different datasets (e.g. E-PRTR inventory versus CLRTAP inventory) can reveal that the total
or sector specific E‐PRTR emissions of some countries exceed the emissions officially reported by the
same countries to CLRTAP (e.g. CEIP, 2010). If this occurs, a straightforward subtraction could not be
used. Instead of developing potentially complicated procedures to overcome this problem, it is strongly
advised to first solve the issue before moving further in the process of spatial disaggregation of emissions.
As the problem might occur due to different reasons and as it is impossible to provide a check list covering
all causes, only some general guidance can be given in this section.

In most cases, exceedances of national totals by point emissions are caused by:

 inconsistent reporting under different reporting obligations;

 different sector classification systems;
 missing data (e.g. not all point sources included in national total);
 inconsistent data updates.

It is therefore suggested to first compare the different reporting obligations, making sure the same
activities are taken into account. Furthermore, it should be checked whether the sector classifications are
applied in a consistent manner (e.g. quite often total E-PRTR emissions do not exceed the national total
whereas sector specific totals do, and this can be a consequence of inconsistent sector conversions).
Moreover, it is advised to check whether the point source contribution in the national total equals the totals
that are effectively reported to E-PRTR. It might be that the point source data were revised after reporting
to E-PRTR, and that the update is only taken into account in the national totals.

Different methodologies to identify the diffuse shares of the national emissions (CLRTAP) which are not
covered by the reporting to E-PRTR have been developed in the past. In all approaches, firstly the different
categorizations (NFR for CLRTAP and Annex I E-PRTR) used for reporting requirements, have been
analysed. Based on this correlation of activities (link NFR – E-PRTR) several subtraction approaches
have been applied. An overview of different procedures can be found in Theloke et al., 2009.

Some European countries (e.g. BE, NL or UK) already apply methods to identify sector specific shares of
diffuse emissions. Additional information can be obtained via contacting emission experts in these
countries.

3.3 Distributing diffuse emissions

There will be many cases where emissions cannot be calculated at a suitably small spatial scale or
estimates are inconsistent with national estimates and statistics. Hence, national emission estimates will
need to be distributed across the national spatial area using a surrogate spatial dataset. The methods
used can range in quality from Tier 3 to Tier 1 depending on the appropriateness of the spatial activity
data being used, the basic principle behind the different methods is however the same for all cases. In
this section, both, the basic principles as some derived methodologies are outlined.

3.3.1 Basic principles

The basic principle of distributing emissions is presented in the formula below using a surrogate
spatial dataset x:

EMEP/EEA air pollutant emission inventory guidebook 2016 11

 7. Spatial mapping of emissions

valueix
emissionix  emissiont  ix

 value
jx

Where:

i : is a specific geographic feature;

emissionix : is the emissions attributed to a specific geographical feature (e.g. a grid,

line, point or administrative boundary) within the spatial surrogate
dataset x;

emissiont : is the total national emission for a sector to be distributed across the
national area using the (x) surrogate spatial dataset;

valueix – jx : are the surrogate data values of each of the specific geographical
features within the spatial surrogate dataset x.

The following steps should be followed:

1. determine the emission total to be distributed (emission t) (either national total for sector or where
a sector is represented by some large point sources: national total — sum of point sources, as
outlined in sections 3.1 and 3.2);

2. distribute that emission using the basic principles above using a suitable surrogate statistic
(according to the detailed guidance by sector given below).

3. Keep the surrogate spatial data in its original shape as long as possible in the calculation. This
makes it easy to correct for mistakes or add new information without a big effort later on.

This approach effectively shares out the national emissions according to the intensity of a chosen or
derived spatially resolved statistic.

3.3.2 Decision tree

Different tiered methods for compiling spatially resolved estimates can be used depending on the
data available. A general decision tree for prioritising approaches for each sector is presented below
in Figure 3-2.

EMEP/EEA air pollutant emission inventory guidebook 2016 12

 7. Spatial mapping of emissions

Figure 3-2 General decision tree for emissions mapping

Start

Do yo u have repo rted o r

measured spatial emissio ns
o r clo sely related activity
data (e.g. traffic flo w by
Yes Apply Tier 3:
vehicle type, fuel
Use this data for
co nsumptio n by type) data the Mapping
at the appro priate spatial
detail?

Do you have spatial

data that relates to the
national emissions Yes Apply Tier 2
using related
source? E.g.
employment by statistical data
relevant sector

No

Key Apply Tier 1

Yes source?
No Using basic surrogate
Try to get National
data e.g. Satellite Land
Statistics
Cover, Population or
General Employment

Tier 3 methods will include estimates that are based on closely related spatial activity statistics, e.g.
road traffic flows by vehicle type, spatial fuel consumption data by sector (e.g. boiler use data).

Tier 2 methods will be based on the use of surrogate statistics. However, for Tier 2, these statistics
need to relate to the sector and could include detailed sector specific employment, population or
household size and number (for domestic emissions).

Tier 1 methods will include the use of loosely related surrogate statistics such as urban rural land
cover data, population (for non domestic sources).

These principals apply to the general methods for estimating spatial emissions. Detailed methods are
provided for each sector in section 3.3.3. The following box (Example 2) provides some general
examples of distributing national emissions.

EMEP/EEA air pollutant emission inventory guidebook 2016 13

 7. Spatial mapping of emissions

EXAMPLE 2: DISTRIBUTION OF NATIONAL EMISSIONS

SOx emissions from residential combustion may be allocated based on a gridded or administrative
boundary (e.g. NUTS) spatial dataset of population density. However, emissions of SO 2 may not
correlate very well to population density in countries where a variety of different fuels are burned (e.g.
city centres may burn predominantly gas and therefore produce very low SO x emissions per head of
population). Additional survey information and or energy supply data (e.g. metered gas supply) could
be used to enhance the population-based surrogate for residential emissions and achieve a better
spatial correlation to ‘real’ emissions.
National transport emissions may be allocated to road links (road line maps) based on measured or
modelled traffic flow, road type or road width information for each of the road links. Again, the closer
the distribution attributes for each road link correlate to the actual emissions, the better. For example,
road width and road type only loosely relate to traffic emissions and provide a poor distribution
method. Being able to distinguish between the numbers of heavy goods vehicles and cars using
different road links each year, and the average speeds of traffic on these links, will improve the
compiler’s ability to accurately allocate emissions.

In many cases the combination of more than one spatial dataset will provide the best results for
distributing emissions. For example, where traffic count/density information is not available, basic
road link information can be combined with population data to derive appropriate emission distribution
datasets to provide a Tier 1 methodology.

3.3.3 Sectoral guidance

Table 1 gives general guidance for the tiered mapping of emissions for different sectors. The first two
columns list and describe the NFR sector codes. The third column reveals the corresponding GNFR
sector (2). The fourth column gives an indication on whether the sector is fully, partly, or not covered
by E-PRTR point sources, by assigning a category (A-D) to each NFR sector. All NFR categories are
represented with the following corresponding categories:

 A – E-PRTR point source relevant: covered by E-PRTR sources

 B – E-PRTR point source relevant: covered by E-PRTR sources but also including diffuse
sources (below E-PRTR reporting threshold)
 C – E-PRTR relevant like industrial facilities (B), but only agricultural installations
 D – Diffuse sources, which are not covered by E-PRTR

This gives a clear indication on whether diffuse emissions that need spatial disaggregation are to be
expected.
Moreover, Table 1 contains a sector specific overview of different spatial mapping approaches
ranging from Tier 3 to Tier 1.

(2) http://www.ceip.at/reporting-instructions/ - table for mapping between GNFR and NFR

EMEP/EEA air pollutant emission inventory guidebook 2016 14

 7. Spatial mapping of emissions

Table 1: General tiered guidance for the spatial distribution of emissions by sector

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
1.A.1.a Employment data
A combination of tiered
Public Electricity and Heat A_PublicPower A approaches might be needed
Production e.g. for 1.A.1.c:
depending on the availability
number of employees
Reported point source of a complete dataset of point
1.A.1 by economic activities
1.A.1.b data or national totals sources. Where only partial
B_Industry A (EUROSTAT Land cover
Petroleum Refining distributed using plant- datasets are available for
Employment statistics
specific capacity or point sources use Example 1
Energy industries - Manufacture of coke
other activity statistics to make best use of point
1.A.1.c oven products)
source data whilst maintaining
Manufacture of Solid Fuels and B_Industry B consistency with national
Other Energy Industries See also section 0 for
emissions estimates.
an example
1.A.2.a
Stationary Combustion in
B_Industry B
Manufacturing Industries and
Construction: Iron and Steel
1.A.2.b
Stationary Combustion in Employment data A combination of tiered
1.A.2 B_Industry B
Manufacturing Industries and approaches might be needed
Construction: Non-ferrous Metals e.g. for 1.A.2.a: depending on the availability
1.A.2.c Reported point source number of employees of a complete dataset of point
Manufacturing
Stationary Combustion in data or national totals by economic activities Population sources. Where only partial
industries and B_Industry B distributed using plant- (EUROSTAT or Land cover datasets are available for
Manufacturing Industries and
construction Construction: Chemicals specific capacity or Employment statistics point sources use Example 1
other activity statistics - Manufacture of basic to make best use of point
1.A.2.d iron and steel and of source data whilst maintaining
Stationary Combustion in ferroalloys) consistency with national
B_Industry B
Manufacturing Industries and emissions estimates.
Construction: Pulp, Paper and Print
1.A.2.e
Stationary Combustion in
Manufacturing Industry and B_Industry B
Construction: Food Processing,
Beverages and Tobacco

EMEP/EEA air pollutant emission inventory guidebook 2013 15

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes

e.g. for 1.A.2.f:

number of employees
by economic activities A combination of tiered
1.A.2 1.A.2.f.i (EUROSTAT approaches might be needed
Stationary Combustion in Employment statistics depending on the availability
B_Industry B
Manufacturing Manufacturing Industries and Reported point source - Manufacture of other of a complete dataset of point
Construction: Other data or national totals non-metallic mineral Population sources. Where only partial
industries and distributed using plant- products; of glass & or Land cover datasets are available for
construction specific capacity or glass products; of point sources use Example 1
1.A.2.f.ii other activity statistics other porcelain & to make best use of point
I_OffRoad D
Mobile Combustion in Manufacturing ceramic products; of source data whilst maintaining
Industries and Construction cement, lime & consistency with national
plaster; of central emissions estimates.
heating radiators and
boilers; of industrial
ashes)
Emissions from domestic
cruise and from international
aircraft flights should be
1.A.3.a.i (i) H_Aviation D
Allocate national excluded from the mapping as
International Aviation (LTO) Estimated detailed
emissions estimates Land cover these are estimated centrally
emissions by airport
to each airport based by EMEP. Where possible,
for Landing and Take-
on Take-Off and additional weightings to take
1.A.3.a.ii (i) H_Aviation D Off by type of aircraft
1.A.3 Landing statistics account of the average size of
Civil Aviation (Domestic, LTO)
aircraft at each airport should
Transport be used (e.g. emission factor
to aircraft size ratios)
Different tiered approaches
1.A.3.b.i F_RoadTransport D Using road network will usually be needed for
Road Transport: Passenger cars information and different road types. Major
Traffic flows and types population based Population roads will often have traffic
1.A.3.b.ii of vehicles (3) traffic intensity and Land cover counts or modelled flows,
Road Transport: Light duty vehicles F_RoadTransport D while minor roads will not.
Countries that have traffic
count/flow information will

(3) Spatial traffic flow statistics that have a basic vehicle type split should be used in order to enable appropriate emissions allocation for the key NFR reporting categories. Where traffic composition
data (flows by vehicle type) are unavailable, averaged composition % can be applied to total flow by road type based on national survey data or on data available on the COPERT4/FLEETS
webpages Gkatzoflias et al (2007), or from the European MEET (Methodologies to Estimate Emissions from Transport), a European Commission (DG VII) sponsored project in the framework of the
4th Framework Programme in the area of Transport project.

EMEP/EEA air pollutant emission inventory guidebook 2013 16

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
1.A.3.b.iii usually need to apply a Tier 2
Road Transport: Heavy duty F_RoadTransport D method for minor roads
vehicles

1.A.3.b.iv
Road Transport: Mopeds & F_RoadTransport D
Motorcycles Different tiered approaches
will usually be needed for
1.A.3.b.v Using road network different road types. Major
Road Transport: Gasoline F_RoadTransport D Traffic flows and types information and Population roads will often have traffic
evaporation of vehicles population based and Land cover counts or modelled flows,
traffic intensity while minor roads will not.
1.A.3.b.vi Countries that have traffic
Road Transport: Automobile tyre F_RoadTransport D count/flow information will
and brake wear usually need to apply a Tier 2
method for minor roads
1.A.3.b.vii
1.A.3 Road Transport: Automobile road F_RoadTransport D
abrasion l
Transport Diesel rail traffic on Rail networks that have been
Population-
the rail network electrified should be excluded
Rail network and weighted
1.A.3.c I_Offroad reconciled with from the distributions where
D population-based distributions of
Railways national mobile possible. This may only be
traffic weightings land cover class
locomotive important if large areas are all
for rail
consumption data electrified (e.g. cities)
Tier 1 & 2 methods will need
to make assumptions about
the weighting of in-port vs. In-
transit emissions. Tier 3
1.A.3.d.i (i) D methods will need to use ship
Route-specific ship Assign national
International maritime Navigation z_Memo movement data from
movement data and Port arrival & emissions to Land
centralised databases and
details of fuel quality destination statistics cover classes for
1.A.3.d.i (ii) account for in port emissions
by region, used to weight port ports and coastal
International inland waterways G_Shipping D for loading and unloading of
consumption and and coastal shipping shipping areas
ships. Harbourmaster or coast
emission factors by areas
1.A.3.d.ii guard data can sometimes
type of ship & fuel
National Navigation (Shipping) G_Shipping D provide details of ship time in
ports and operations.
Emissions will need to be
broken down into national and
international data. Useful

EMEP/EEA air pollutant emission inventory guidebook 2013 17

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
information for the mapping of
shipping data can be obtained
1.A.3 from Entec UK (2005).

Transport 1.A.3.e Reported/collected

B_Industry D Land cover Land cover
Pipeline compressors data

Large point source data is

1.A.4.a.i C_OtherStationary D likely to be minimal unless
Reported/collected
Commercial / Institutional: Stationary Comb Employment for public there are large district heating
point source data, Land cover
and commercial or commercial/institutional
boiler census or
1.A.4.a.ii services heating plant included in the
surveys
Commercial / Institutional: Mobile I_OffRoad D national inventory under
sectoral 1A4a
Tier 1 & 2 methods assume
that a linear relationship
between emissions and
Detailed fuel deliveries
1.A.4.b.i C_OtherStationary D population density or land
for key fuels (e.g.gas) Population or
Residential: Stationary plants Comb cover exists. This assumption
and modelled household density
will be most realistic if a
estimates for other combined with land Land cover
country has a uniform
fuels using data on cover data if smoke
1.A.4.b.ii distribution of fuel use by
population density control areas exist in
Residential: Household and I_OffRoad D type. Where there is a broad
and/or household cities.
gardening (mobile) variation of fuel type use in
1.A.4 numbers and types.
different areas, the accuracy
Other sectors of the simple method will be
much lower
Where land cover is used,
emissions for agriculture and
1.A.4.c.i
forestry should be split and
Agriculture/Forestry/Fishing: C_OtherStationary D For 1.A.4.c.i and
distributed according to the
Stationary Comb 1.A.4.c.ii: For 1.A.4.c.i and
relevant classes. Where this
Detailed fuel deliveries 1.A.4.c.ii: Land cover
is not possible, emissions
for key fuels (e.g. gas) Employment data for
should be distributed
1.A.4.c.ii and modelled the agricultural and
according to a combined land
Agriculture/Forestry/Fishing: Off- estimates for other forestry sectors
cover class for agriculture and
road Vehicles & Other Machinery fuels using
forests or allocated to the
I_OffRoad D employment data
dominant class e.g. ‘arable
land’ for countries where
emissions from agricultural
combustion dominate.

EMEP/EEA air pollutant emission inventory guidebook 2013 18

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
Where employment data is
used, care should be taken to
ensure that the employment
classes are representative of
the national sector for
Agriculture and Forestry.
Employment statistics will
often include the financial and
administrative head offices
(often located in cities) while
energy statistics based
national emissions from these
For 1.A.4.c.iii: ‘head offices’ may be included
Allocation of For 1.A.4.c.iii: For 1.A.4.c.iii: under 1.A.4.a.i Commercial /
1.A.4.c.iii emissions to ports on Allocation of Assign national Institutional: Stationary. Care
Agriculture/Forestry/Fishing: G_Shipping D fish landings and to emissions to ports on emissions to Land should be taken to ensure that
National Fishing geographic areas of fish landings cover classes for the emissions are located
fishing grounds ports. where they occur.
The use of employment data
will locate emissions at
registered places or regions of
work and may tend to focus
emissions inappropriately to
urbanised areas

Emissions from fishing are

likely to be associated with
fishing grounds rather than
port activities. Tier 3 methods
will need to allocate emissions
to ports and fishing grounds

C_OtherStationary
1.A.5.a
Comb D
Other, Stationary (including Military)
1.A.5 Population Population Population
and Land cover and Land cover and Land cover
Other 1.A.5.b
Other mobile (Incl. military, land I_OffRoad D
based & recreational boats).

EMEP/EEA air pollutant emission inventory guidebook 2013 19

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
Locate point sources
and use employment
1.B.1.a data for specific
Fugitive emission from Solid Fuels: D_Fugitive B sectors to allocate Mines, fuel transformation
Coal Mining and Handling emissions plant, depots and distribution
centres are likely to be
1.B.1
e.g. for 1.B.1.b: Total employment regulated and significant
Reported point source
1.B.1.b number of employees for the mining and national industrial sites. In
data or using plant-
Fugitive Fugitive emission from Solid Fuels: D_Fugitive B by economic activities transformation these cases site-specific
specific capacity or
Solid Fuel Transformation (EUROSTAT industry as a information can be collected
emissions from other activity statistics
Employment statistics whole and used to distribute national
solid fuels - Electric power emission estimates over a
1.B.1.c generation, transmis- number of point sources or
Other fugitive emissions from solid D_Fugitive B sion and distribution; grids
fuels Manufacture of coke
oven products)

Location of off shore

extraction installations
and allocation of
1.B.2.a.i D_Fugitive B Exploration and emissions using
Exploration Production, Transport production distribution proxies such as
centres should be employment or
1.B.2 identified and capacity. Distribution
1.B.2.a.ii D_Fugitive B emissions estimated to be allocated evenly Many of the exploration,
Fugitive Identification of
Refining / Storage allocated to the point over the distribution refining and storage facilities
large point
emissions oil and locations using network will be regulated and data
sources and
production activity available from energy
natural gas distributed evenly
1.B.2.a.iii D_Fugitive D data. e.g. for 1.B.2.a.i: ministries or regulators.
over locations.
Distribution of oil products number of employees Operators or regulators for
Use population or
Distribution emissions by economic activities distribution of fuels will often
1.B.3 employment data
should be mapped (EUROSTAT be able to provide network
for distribution
Other fugitive 1.B.2.b D_Fugitive B using details of Employment statistics maps
Natural gas distribution network - Manufacture of gas;
emissions and leakage rates or distribution of
1.B.2.c losses along the gaseous fuels through
Venting and flaring D_Fugitive B systems mains; Manufacture of
refined petroleum
products)

EMEP/EEA air pollutant emission inventory guidebook 2013 20

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
2.A.1 B_Industry B
Cement production

2.A.2 B_Industry B
Lime production
Employment data Where possible, try to use
point source data as the basis
e.g. for 2.A.1 and for estimating process
2.A.3 B_Industry B 2.A.2: number of emissions. The methodology
Glass production employees by for Tier 2 relies heavily on
economic activities detailed sectoral employment
Integrate reported (EUROSTAT Population data for surrogate spatial
point source data or Employment statistics or Land cover distributions. However, in
2.A.4.a B_Industry B derive emissions - Manufacture of many cases these are not
2.A Quarrying and mining of minerals using plant specific, paper and paper specific to the processes
other than coal activity, throughput, products; producing emissions, as
Mineral Products
production, capacity or Manufacture of emissions are likely to be
other activity statistics cement, lime and highly specific to particular
plaster; Manufacture plants and processes.
2.A.4.b B_Industry D of abrasive products Employment data will also
Construction and demolition and non-metallic distribute emissions to
mineral products; locations that may have
Manu-facture of basic administrative or head office
iron and steel and of activities only where process
2.A.4.c ferro-alloys) emissions do not occur
Storage, handling and transport of B_Industry D
mineral products

2.A.4.d B_Industry B
Other mineral products

EMEP/EEA air pollutant emission inventory guidebook 2013 21

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
2.B.1 B_Industry B
Ammonia production

2.B.2 B_Industry B
Nitric acid production

2.B.3 B_Industry B
Adipic acid production
Employment data
2.B.5 B_Industry B
Carbide production e.g. for 2.B.1, 2.B.2,
2.B 2.B.3 and 2.B.4:
2.B.6 B_Industry B number of employees
Chemical Industry
Titanium dioxide production by economic activities
(EUROSTAT
2.B.7 B_Industry B Employment statistics
Soda ash production - Manufacture of other
inorganic basic
2.B.10.a B_Industry B chemicals)
Other chemical industry
Population
2.B.10.b B_Industry B Integrate reported or Land cover
Storage, handling and transport of point source data or
chemical products derive emissions
2.C.1 B_Industry B using plant specific, e.g. for 2.C.1 and
Iron and steel production activity, throughput, 2.C.2: number of
2.C production, capacity or employees by
2.C.2 B_Industry B other activity statistics economic activities
Metal Production (EUROSTAT
Ferroalloys production
Employment statistics
2.C.3 B_Industry B - Manufacture of basic
Aluminium production iron and steel and of
ferro-alloys)
2.C.4 B_Industry B
Magnesium production

2.C.5 B_Industry B
Lead production

2.C.6 B_Industry B
Zinc production

EMEP/EEA air pollutant emission inventory guidebook 2013 22

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes

2.C.7.a B_Industry B
Copper production
B_Industry B Integrate reported Employment data Population
2.C.7.b point source data or or Land cover
Nickel production derive emissions e.g.: number of
using plant specific, employees by
2.C.7.c B_Industry B activity, throughput, economic activities
Other metal production production, capacity or (EUROSTAT
other activity statistics
2.C.7.d B_Industry B
Storage, handling and transport of
metal products
2D3a E_Solvents D
Decorative coating application Employment or
appropriate
2D3b E_Solvents B population data
Industrial coating application For Tier 2 where processes
Integrate reported e.g. for number of are industrial and there is a
2.D – 2.L 2D3c E_Solvents B point source data or employees by good employment dataset,
Other coating application derive emissions economic activities Land cover use this.
Solvent use and using plant-specific, (Employment for
Other production 2D3d B_Industry D activity, throughput, coating industries For emissions that result from
Asphalt roofing production, capacity or including metal consumption of products in
other activity statistics packaging, vehicle the home, use population
2D3e B_Industry D refinishing, rolling
Road paving with asphalt mills, vehicle repair,
wood coating, ... )

EMEP/EEA air pollutant emission inventory guidebook 2013 23

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
2D3f E_Solvents B
Degreasing

2D3g E_Solvents D
Dry cleaning

2D3h E_Solvents D
Domestic solvent use including
fungicides

2D3i E_Solvents B
Chemical products

2D3j Printing E_Solvents B Employment or

appropriate
2D3k population data For Tier 2 where processes
Other solvent use E_Solvents D Integrate reported are industrial and there is a
point source data or e.g. for number of good employment dataset,
2.D – 2.L 2G Other product use E_Solvents B derive emissions employees by use this.
using plant-specific, economic activities Land cover
Solvent use and
2H1 B_Industry B activity, throughput, (Employment in
Other industrial Pulp and paper industry production, capacity or newspaper and For emissions that result from
other activity statistics magazine industry) consumption of products in
production
2H2 B_Industry B the home, use population
Food and beverages industry e.g. for 3.D.2 and
3.D.3: population
2H3 B_Industry B density
Other industrial processes

2I Wood processing B_Industry B

2J Production of POPs B_Industry B

2K
Consumption of POPs and heavy B_Industry D
metals (e.g. electrical and scientific
equipment)

2L Other production, consumption, B_Industry D

storage, transportation or handling of
bulk products

EMEP/EEA air pollutant emission inventory guidebook 2013 24

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
3.B.1.a
K_AgriLivestock D
Manure management - Dairy cattle

3.B.1.b
K_AgriLivestock D
Manure management - Non-dairy
cattle
Employment statistics When using these statistics
3.B.2 Reported emissions or land cover and care should be taken to
K_AgriLivestock D
Manure management – Sheep from regulated farms agricultural production Land cover for account for possible over
or detailed spatial statistics arable land allocations to head/offices or
3.B.3 farm livestock survey or land cover and market employment in urban
K_AgriLivestock C
Manure management - Swine statistics livestock statistics areas that will distort the
pattern of emissions and
3.B.4.a allocate too many emissions
K_AgriLivestock D
Manure management – Buffalo to urban areas

3.B.4.f
K_AgriLivestock D
Manure management – Goats

3.B.4.g
K_AgriLivestock D
Manure management – Horses
Reported emissions Employment statistics Land cover for When using these statistics
3.B.4.i from regulated farms or land cover and arable land care should be taken to
K_AgriLivestock D
Manure management - Mules and or detailed spatial agricultural production account for possible over
asses farm livestock survey statistics allocations to head/offices or
statistics or land cover and market employment in urban
3.B.4.j livestock statistics areas that will distort the
K_AgriLivestock C
Manure management – Poultry pattern of emissions and
allocate too many emissions
3.B.4.n to urban areas
K_AgriLivestock D
Manure management - Other
animals
3.D.a.1 L_AgriOther D
Inorganic N-fertilizers (includes also
urea application)
3.D – 3.I
3.D.a.2.a L_AgriOther D
Other agriculture Animal manure applied to soils

3.D.a.2.b L_AgriOther D Reported emissions Land cover for For Tier 3 survey statistics for
Employment statistics
Sewage sludge applied to soils from regulated farms arable land crop production can be

EMEP/EEA air pollutant emission inventory guidebook 2013 25

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes
3.D.a.2.c L_AgriOther or detailed spatial or land cover and combined with fertilizer
Other organic fertilisers applied to D farm crop/fertilizer use agricultural production use/stubble burning rates to
soils (including compost) survey statistics statistics estimate weightings by crop
or land cover and type. Farm level data is often
3.D.a.3 L_AgriOther livestock statistics commercially sensitive and
Urine and dung deposited by grazing D may need to be aggregated
animals

3.D.a.4 L_AgriOther
Crop residues applied to soils D

3.D.b L_AgriOther
Indirect emissions from managed D
soils

3.D.c L_AgriOther
Farm-level agricultural operations D
including storage, handling and
transport of agricultural products

3.D.d L_AgriOther D Reported emissions Land cover for For Tier 3 survey statistics for
Off-farm storage, handling and from regulated farms Employment statistics arable land crop production can be
transport of bulk agricultural or detailed spatial or land cover and combined with fertilizer
products farm crop/fertilizer use agricultural production use/stubble burning rates to
survey statistics statistics estimate weightings by crop
3.D.e Cultivated crops L_AgriOther D or land cover and type. Farm level data is often
livestock statistics commercially sensitive and
3.D.f Use of pesticides L_AgriOther D may need to be aggregated

3.F
Field burning of agricultural residues L_AgriOther D

3.I
Agriculture other L_AgriOther D
5.A J_Waste
Waste disposal to land Most countries with regulated
5.A – 5B Biological treatment of waste - Solid Population
statistics and landfill land disposal will have
waste disposal on land D Evenly distributed statistics weighted
Biological disposal records by records of landfill sites in use.
emissions over landfill with discontinuous
treatment of site It may be more difficult to
5.B.1 site locations urban fabric land
identify disused sites or sites
waste Biological treatment of waste – J_Waste D cover
Composting records that are not regulated
Composting

EMEP/EEA air pollutant emission inventory guidebook 2013 26

 7. Spatial mapping of emissions

Best -------------------------------------------------------------- Worst

NFR sector NFR sector name GNFR sector Cat. Tier 3 Tier 2 Tier 1 Notes

5.B.2
Biological treatment of waste - J_Waste D
Anaerobic digestion at biogas
facilities
J_Waste A
5.C.1.a
Municipal waste incineration
J_Waste A
5.C.1.b
Industrial waste incineration Incineration 6Ca–d is
generally regulated or
controlled. Regulators or trade
J_Waste D Employment data
5.C.1.c associations will hold site
5.C Emissions distributed for the specific
Clinical waste incineration location details and often
Regulated process over known sites industry or
Waste records of activity. Small scale
J_Waste D emissions by site based on capacity or population/ farm
5.C.1.d waste burning (6Ce) should
incineration population statistics for small
Sewage sludge incineration be distributed using
scale burning
population or farm statistics
J_Waste D
5.C.1.e Cremation depending on the dominant
small scale burning sector
5.C.1.f
J_Waste D
Other waste incineration

5.C.2 Open burning of waste

J_Waste D
5.D.1
J_Waste Employment statistics
Domestic wastewater handling D Many countries now regulate
for wastewater
5.D Regulated process wastewater treatment plant.
treatment location of
5.D.2 D information and data Population Locations should be well
Waste-water J_Waste plant and capacity or
Industrial wastewater handling on wastewater statistics known and activity
some surrogate of
handling treatment plant data/emissions available by
capacity-based on
5.D.3 D site
J_Waste population density
Other wastewater handling
6.A
Other (included in National Total for R_Other D
6 Other Entire Territory)
Detailed surveys of National emissions
11
land use types and distributed using land
Biogenic Basic land cover
burned area combined cover data and
emissions and data
with (Ing 2007) factors burned area data from
forest fires
for emissions NatAir

EMEP/EEA air pollutant emission inventory guidebook 2013 27

 7. Spatial mapping of emissions

3.3.4 Example

Gridding methodology of the emissions from road transport (NFR sector 1.A.3.b)

Several European countries (e.g. BE, NL or UK) apply country specific methods to grid emissions from
road transport. In this section, the methodology developed by Theloke et al. (2009) is described, as it is
general and widely applicable over the whole of Europe. However, note that some information sources
quoted in this section are only available commercially.

Sector description

The emissions from road transport arise from the combustion of fuels such as gasoline, diesel, liquefied
petroleum gas (LPG), and natural gas in internal combustion engines (EMEP/EEA, 2013). The sector road
transport considers all on road transportations of passengers as well as good from all vehicle classes
driven by fuel combustion. This sector is considered as diffuse (see Table 1, NFR sector 1.A.3.b),
therefore, according to the scheme presented in Figure 3-1, compilation of point source emissions is not
relevant and consequently the national reported emissions can be considered as diffuse emissions.

Exhaust emissions from road transport (non exhaust emissions are not considered here) are reported
according to the following NFR sector codes:

 1.A.3.b.i Road Transport: Passenger cars;

 1.A.3.b.ii Road Transport: Light duty vehicles;
 1.A.3.b.iii Road Transport: Heavy duty vehicles;
 1.A.3.b.iv Road Transport: Mopeds & Motorcycles.

The allocation of emissions distinguishes on-road activities on the following street types:

 highways;
 rural roads;
 urban roads.

Furthermore, the vehicle types are distinguished by fuel type for each relevant NFR code:

 passengers cars (LPG, diesel and gasoline)

 light duty vehicles (diesel and gasoline)
 heavy duty vehicles (diesel)
 motor cycles (two and four-strokes, gasoline)

Emission input data

Emission data used for the gridding procedure are reported sectoral totals from the CLRTAP NFR sectors
1.A.3.b.i, Passenger cars; 1.A.3.b.ii, Light duty vehicles; 1.A.3.b.iii, Heavy duty vehicles; 1.A.3.b.iv,
Mopeds & Motorcycles.

Additional preparation of the emission data sets is necessary to distinguish between different road and
vehicle types. The first intermediate calculation is the allocation of the emissions reported to CLRTAP into
different road classes (highway, rural and urban) based on the TREMOVE model (TREMOVE, 2010). The
results are road and vehicle class specific shares for each pollutant and country.

The next step is to harmonize the road and vehicle classes with the road network from the TRANSTOOLS
model (TRANSTOOLS, 2010). The road classes (highway, rural and urban) are allocated to both roads
which are covered and those that are not covered by TRANSTOOLS. As TRANSTOOLS covers only
highways and major rural roads the emissions from rural roads based on TREMOVE calculation therefore
have to be partly allocated to the roads covered by TRANSTOOLS. The assumption is that 50 % of the

EMEP/EEA air pollutant emission inventory guidebook 2016 28

 7. Spatial mapping of emissions

rural roads are allocated to the TRANSTOOLS roads and the remaining 50 % are categorised as rural
road emissions.

The second intermediate calculation is for the distinction of the rural and urban road traffic into line and
area sources. The assumption for the rural roads not covered by TRANSTOOLS is that 70 % of the
emissions are categorised as line sources and the remaining 30 % of the rural roads not covered by
TRANSTOOLS are categorised as area sources. The urban roads are categorised 50 % as line and 50 %
as area sources. This strategy is clarified in the upper part of the scheme presented in Figure 3-3.

Applied methodology
The spatial distribution of emission data into target grid cells is carried out using the approach given in the
formula:

valueix
emissionix  emissiont  ix

 value
jx

Where:

i : is a specific geographic feature; here: road segments within the grid

layer.

emissionix : is the emissions attributed to a specific geographical feature (e.g. a grid,

line, point or administrative boundary) within the spatial surrogate
dataset x; here: emission of road transport (for each vehicle and road
type)x at the specific road segmenti.

emissiont : is the total national emission for a sector to be distributed across the
national area using the (x) surrogate spatial dataset; here: national total
emission of road transport (distinguished by vehicle and road
category)x.

valueix – jx : are the surrogate data values of each of the specific geographical
features within the spatial surrogate dataset x; here: the fraction of
traffic volume for each individual road segment within each target grid
cell of the sum of the overall traffic volume within the country.

The specific geographic feature is here the line (road segment) and grid level. The use of surrogate data
(proxy variables) allows determining the share of annual emissions which are attributed to each grid cell.
The proxy data are correlated with the emission source sectors and defined by means of:

 geographical resolved statistical information or

 land cover/land use data

Spatial distribution and gridding

The methodology of the spatial distribution of the road transport can be summarised into the following
main steps:

1. Regionalisation of national emission releases: allocation of the emission values based on

traffic volume data for each road segment and also population density related to roads not
covered by TRANSTOOLS. This allocates the share of the national totals of each specific
pollutant to each road segment in the TRANSTOOLS model and to NUTS level 3 regions for the
urban and rural traffic not covered by TRANSTOOLS.

EMEP/EEA air pollutant emission inventory guidebook 2016 29

 7. Spatial mapping of emissions

2. Gridding: spatial distribution of the regionalised emission values on grid cell level is based on
TRANSTOOLS, GISCO (ROAD) (GISCO, 2010) road network and gridded population density
from JRC (Gallego, 2010). The result are gridded emissions for each road segment and regional
unit to each grid cell, using the following underlying parameters:
 Traffic volume and road network from TRANSTOOLS for highways and partly for rural
roads;
 Road network divided by road type from GISCO (ROAD) for the roads not covered in
TRANSTOOLS (secondary and local roads);
 Gridded population density as weighting factor for line sources in relation to rural and
urban roads not covered by TRANSTOOLS. Additionally as distribution parameter for
rural and urban area sources.
 Degree of urbanisation (GISCO, 2010) for the categorization of the roads from GISCO
and the gridded population from JRC (Gallego, 2010) into urban and rural.

The methodology for the spatial distribution of national total emissions from road transport activities to a
grid for each vehicle and road type is shown in Figure 3-3.

Figure 3-3 Overview of the applied methodology for the spatial distribution of the road
transport, Theloke et al. (2009).

Gridding methodology for the diffuse emissions from Stationary Combustion in Manufacturing
Industries and Construction: Iron and Steel (NFR sector 1.A.2.a)

Several European countries (e.g. BE, NL and UK) apply country specific methods to grid diffuse emissions
from industrial releases. In this paragraph, a general methodology based on employment data is
described. NFR sector 1.A.2.a is chosen as an example, but the methodology can be applied for several
sectors (see Table 1) and moreover is widely applicable over the whole of Europe.

Sector description

The NFR sector 1.A.2.a, ‘Stationary Combustion in Manufacturing Industries and Construction: Iron and
Steel’, is dominated by E-PRTR related point sources, but is also containing diffuse emissions (see

EMEP/EEA air pollutant emission inventory guidebook 2016 30

 7. Spatial mapping of emissions

Table 1). This comprises industrial releases from industrial facilities not covered by the E-PRTR
Regulation and its Annexes..

In this section, a methodology to spatially allocate the diffuse emissions from stationary combustion in
iron and steel industry is described.

Emission input data

As input emission data for the spatial distribution of diffuse industrial releases, national data reported to
the LRTAP Convention can be taken. To obtain the diffuse part of the country specific emissions, not
covered by the E-PRTR regulation, a subtraction methodology, as described in 3.2, first needs to be
applied.

Applied methodology

Diffuse industrial releases can be spatially distributed according to different proxy data. The basic principle
of distributing emissions is presented in the formula below using a surrogate spatial dataset:

valueix
emissionix  emissiont  ix

 value
jx

Where

i : is a specific geographic feature; here: a high resolution (e.g. 3 km x 3

km) grid cell.

emissionix : is the emissions attributed to a specific geographical feature (e.g. a grid,

line, point or administrative boundary) within the spatial surrogate
dataset x; here: emission value attributed to each grid cell.

emissiont : is the total national emission for a sector to be distributed across the
national area using the (x) surrogate spatial dataset; here: when starting
from the CLRTAP data, the diffuse industrial emissions within a
member state (subtraction methodology required).

valueix – jx : are the surrogate data values of each of the specific geographical
features within the spatial surrogate dataset x; here: a spatial surrogate
based on
 number of employees according to the activity (e.g. EUROSTAT
Employment statistics - Manufacture of basic iron and steel and
of ferroalloys) and
 "Industrial or Commercial units" from CORINE land cover class
(CLC2000/2006, 2010).

Thus, a combination of land use and employment statistics: the share

of employees in each ‘Industrial or Commercial’ CORINE land cover
unit, related to the national total of the employees.

The use of surrogate data (proxy variables) allows determining the share of annual emissions that have
to be attributed to each cell. The crucial part in the methodology described above, is the construction of
the spatial surrogate. It basically consists of the following steps:

 Select the appropriate land cover types in the CORINE land cover map. Here: "Industrial
or commercial units" (CLC2000/2006, 2010);
 Calculate the share of employees in each polygon of the selected units;

EMEP/EEA air pollutant emission inventory guidebook 2016 31

 7. Spatial mapping of emissions

 Use the employee weighted industrial zones as spatial surrogate in the above formula.

3.4 Combining different spatial features

Following the emission-distribution process, emissions at this stage are likely to be in a number of different
spatial forms including different sizes of grids, polygons, lines and even point sources (where emissions
are derived by allocating national estimates according to published capacity or employment information),
each determined by the spatial characteristics of the source data. There is then a need to combine these
together into a unifying format. This is usually a regular grid at a resolution appropriate to the spatial
accuracy of the data inputs.

Spatial emission data will need to be combined to form emission maps. This is generally done by resolving
the different spatial forms to a common grid so that different sectors/sources can be aggregated. The
common grid can either be the EMEP 0.1 x 0.1 degree grid or another grid based on national coordinates
and or smaller cell sizes. The methodologies for converting the different forms to a common grid are
outlined below. For line and area conversion to grids an intersect operation is needed. This intersects the
boundaries of the polygon or the length of the line with the boundaries of the grid and creates a new set
of features cut to the extent of each grid cell.

3.4.1 Point sources to grids

Point sources can be allocated directly to the grid within which they are contained by converting
(rounding) the x,y values to that of the coordinates used to geo-reference the grid or by intersecting
the point with the grid.

3.4.2 Area source (polygons) to grids

Intersecting the polygon with the grid will produce a dataset of polygons contained within each grid.

The fraction of the area of the new polygons can be used to distribute the emissions/surrogate statistics
from the original polygon to the grid cells. Alternatively, an emission rate/area can be applied to the new
polygon area and that emission/surrogate statistic assigned to the grid cell.

3.4.3 Line sources to grids

Intersecting the line features with the grid will produce a dataset of shorter line lengths contained within
each grid.

EMEP/EEA air pollutant emission inventory guidebook 2016 32

 7. Spatial mapping of emissions

The fraction of the original line length of the new line can be used to distribute the emissions/surrogate
statistics from the original line to the grid cells. Alternatively, an emission rate/unit of length can be applied
to the new line length and that emission/surrogate statistic assigned to the grid cell.

3.4.4 Converting between different spatial projections (e.g. GIS to EMEP)

In a number of cases, an inventory compiler may need to combine data from different spatial datasets and
extents to eventually derive the 0.1 x 0.1 degree EMEP grid.

The Open Geospatial Consortium Inc. provides guidance and standards for coordinate transformation
(see www.opengeospatial.org/standards/ct).

3.4.5 Aggregating to the UNECE GNFR

The aggregated sectors ‘gridded NFR’ (GNFR) for reporting are defined in Annex I to the Guidelines for
reporting emission data under the Convention on Long-range Transboundary Air Pollution (1). These
aggregations can be achieved through the aggregation of the spatially resolved (mapped) detailed NFR
sectors (or sector groups). Aggregation of NFR to GNFR prior to mapping is not recommended as it may
result in reduced accuracy in the placement of emissions.

4 Finding key spatial data sources

There are a number of different sources of spatial data. The first place to look should be national statistical
centres such as demographic, economic, transport, regulatory, energy, regulating bodies and trade
associations, as these data are likely to be most up-to-date.

4.1 General

4.1.1 Administrative boundaries

Statistics may be collected and stored with reference to regional or local government names while the
information that defines the spatial boundaries of these areas (geographies) could be maintained in
separate mapping datasets. Often a national mapping body is responsible for the boundary datasets while
specific statistics attributable to these boundaries can be available from elsewhere. In these cases the
statistics will need to be joined to the boundary dataset using lookups between the statistical dataset and
the spatial structure dataset (e.g. area IDs or names). A number of common national datasets are listed
below and give a starting point for data collection activities.

4.1.2 Geo-referenced data

Some data will be available with grid reference attributes, whereas other data may have postal address
details. A grid reference lookup will be required to place the latter on a map.

Where national data is unavailable or too time consuming to collect, a number of international datasets
can be used (see section 4.3 of the present chapter).

EMEP/EEA air pollutant emission inventory guidebook 2016 33

 7. Spatial mapping of emissions

4.2 National datasets

4.2.1 Population and employment

Most countries will have spatial population and employment datasets based on administrative boundaries
that can be used/combined to derive specific distributions or used as general default distributions where
other methods are not feasible. This is a good basic dataset that can be used in many ways for distributing
emissions from different sources.

4.2.2 Gas distribution networks

Information on gas supply by region or on a GIS basis is often available from energy departments, from
gas suppliers or from national statistical centres. Even national information on the number of households
with/without gas supply can be useful when combined with population to estimate a distribution network.

4.2.3 Agricultural data

Most countries have agricultural census or survey data collected (e.g. livestock numbers, crop production,
fertilizer use) at a detailed spatial scale at administrative boundary level.

4.2.4 Road network information

It is likely that many countries will hold national or commercial road network datasets that includes the
road geography. These can be used to help distribute road traffic emissions in combination with traffic
intensity statistics for administrative boundaries or specific count points.

4.2.5 Rail

Rail networks can be relatively easy to identify and datasets of the network are usually held by national
mapping departments or organisations. Rail activity data is more difficult to obtain but can be part of
national statistics or generated from detailed timetable information.

4.2.6 Airport activity data

Many countries have detailed aircraft movement datasets as part of their national statistics. These can be
used to distribute emissions from Landing and Take-Off (LTO) and from airside support vehicles to airport
areas.

4.2.7 Aviation

National aviation authorities will have detailed databases of aircraft movements, the aircraft type, origin
and destination by airport. These data can be used to distribute the nationally estimated emissions for
Landing and Take-Off for domestic and international aviation and allocate these emissions to the
appropriate grid square.

4.2.8 National shipping

National shipping data is usually in the form of port arrival and departure statistics and available from the
national statistical authority. These will need to separate national and international shipping but can be
used to provide an indication of port activity for attributing national emissions.

EMEP/EEA air pollutant emission inventory guidebook 2016 34

 7. Spatial mapping of emissions

4.2.9 Point source information

Regulation of large point sources is common in most countries and public reporting of regulated emissions
occurs under the requirements of the Aarhus Convention PRTR Protocol, LCPD and E-PRTR, all of which
have established a requirement for regular point source emissions reporting. Regulators within countries
that are parties to these protocols and directives will have publicly available records of reported emissions
data. Alternative large point source information sources can include:

 trade associations
 operators
 statistical energy and productivity publications (capacity).

4.2.10 Local inventory data

In some cases local inventory data can be used to improve the spatial distribution of emissions for
transport and stationary sources by providing smaller process emissions by point source and traffic
information. However, integration of this data with the nationally reported data and resolving emissions
allocated to other areas can be time consuming and difficult to document.

4.3 International datasets

There are a number of different international datasets that can be used to help derive the spatial emissions
of a country. In this section both publically and commercially available spatial data sets are listed. It is
mentioned if a data access fee is required.

Note that existing spatial emission inventories at international or European level can be used as proxy
data (or as such) to derive spatial emissions of a country. Existing spatial inventories are however not
listed here, an overview of spatial inventories is provided in section 5.

4.3.1 INSPIRE

In the future the EU intends to provide access to spatial datasets through the INSPIRE programme
http://inspire.jrc.ec.europa.eu/. A number of different geographical datasets will be available under
this European initiative.

4.3.2 CORINE land cover

If relevant national spatial statistics are not available then a simpler and less accurate method using
land cover data can be used to derive emissions.

The CORINE dataset provides processed satellite images showing different land cover classes. This
data can be accessed from: http://www.eea.europa.eu/data-and-maps/data/corine-land-cover-2006-
raster-2. Among the 43 different land covers provided, CORINE provides the following datasets of
relevance to emissions mapping:

 continuous urban fabric

 discontinuous urban fabric
 industrial or commercial units
 road and rail networks and associated land
 port areas
 airports
 mineral extraction sites
 dump sites
 construction sites

EMEP/EEA air pollutant emission inventory guidebook 2016 35

 7. Spatial mapping of emissions

 green urban areas

 arable land
 sport and leisure facilities.

A number of these CORINE datasets can be used individually or in combination to generate spatial
distributions for sectoral emissions.

The IMPRESAREO project (Dore et al., 2001, 2004) derived weightings to apply to elements of the
CORINE land cover map to represent NFR sectors. These weightings are listed in the table below.

NFR Sector Land cover class Weighting

1.A.1, 1.A.2, 2 and 3 Industry 76 %
Continuous urban fabric 15 %
Discontinuous urban fabric 1%
Ports 4%
Airports 4%
1.A.3.d.ii Ports 100 %
1.A.3.a.ii.(i) Airports 100 %
1.A.4.a Continuous urban fabric 50 %
Discontinuous urban fabric 50 %
4 Arable land 100 %
6.A Discontinuous urban fabric 100 %

Where CORINE or similar data is not available, satellite-based land cover data can often be derived from
raw images using the CORINE methodology.

4.3.3 ESA GlobCover

GlobCover is an ESA initiative which began in 2005 in partnership with JRC, EEA, FAO, UNEP,
GOFC-GOLD and IGBP. The aim of the project was to develop a service capable of delivering global
composites and land cover maps using as input observations from the 300m MERIS sensor on board the
ENVISAT satellite mission.

ESA makes available the land cover maps, which cover 2 periods: December 2004 - June 2006 and
January - December 2009.

See: http://due.esrin.esa.int/globcover/

4.3.4 CORINE population

If relevant national spatial statistics are not available then for some sectors (see Table 1) a simpler and
less accurate method using population density data can be used to derive emissions.

The CORINE population density dataset provides population density derived from the CORINE 2000 land
cover data set.

This data can be accessed from: http://www.eea.europa.eu/data-and-maps/data/population-density-

disaggregated-with-corine-land-cover-2000-2

4.3.5 SEDAC population

Where CORINE or similar data is not available, global population density data can be used.

The Socioeconomic Data and Applications Center (SEDAC) of NASA provides Gridded Population of the
World (GPW). GPWv3 depicts the distribution of human population across the globe. GPWv3 provides

EMEP/EEA air pollutant emission inventory guidebook 2016 36

 7. Spatial mapping of emissions

globally consistent and spatially explicit human population information and data for use in research, policy
making, and communications.

Population data estimates are provided for 1990, 1995, and 2000, and projected (in 2004, when GPWv3
was released) to 2005, 2010, and 2015. The projected grids were produced in collaboration with the United
Nations Food and Agriculture Programme (FAO) as Population Count and Density Grid Future Estimates.
There is also an extensive map collection that includes population density and sub-national administrative
boundary maps (depicting the input units) at country, continental, and global levels.

This data can be accessed from: http://sedac.ciesin.columbia.edu/data/collection/gpw-v3

4.3.6 Eurostat
Eurostat is a Directorate-General of the European Commission located in Luxembourg. Its main
responsibilities are to provide statistical information to the institutions of the European Union (EU) and to
promote the harmonisation of statistical methods across its member states and candidates for accession
as well as EFTA countries.

Employment data is available from Eurostat at NUTS 3 level, split into three categories: services, industry
and agriculture. Employment statistics are reported using the NACE classification system.

Eurostat also has several useful datasets on ‘Industry, trade and services’, ‘Agriculture and Fisheries’,
and ‘Transport’.

This data can be accessed from: http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home

4.3.7 ESRI data

ESRI provides a full spectrum of ready-to-use, high-quality geospatial data for your GIS visualization and
analysis projects. Some data sets are publically available and can be used for spatial distribution of
emissions. Interesting examples are locations of airports at global scale (World Airports), World Roads,
World Railroads, World Urban Areas, Europe Population Density, ...

Data can be accessed from:

http://www.arcgis.com/home/group.html?owner=esri&title=ESRI%20Data%20%26%20Maps&content=al
l

Some data sets are only commercially available, e.g. through ArcGIS licence. An example: European rail
network (included in ArcGIS version 9 installation CD-rom, (Railroads, ESRI Data and Maps, ArcGIS 9).

4.3.8 Open Street Maps

OpenStreetMap (OSM) is a collaborative project to create a free editable map of the world. Rather than
the map itself, the data generated by the OpenStreetMap project is considered its primary output.
Examples of data made available through OSM are locations of airports, road network, railroad network,
…

Geofabrik is a consulting and software development firm based in Karlsruhe, Germany specialising in
OSM services. The OSM data can easily be accessed through the Geofabrik website:
http://download.geofabrik.de/osm/europe/.

4.3.9 MapCruzin

Examples of data made available through MapCruzin are locations of airports, road network, railroad
network, waterways, …

EMEP/EEA air pollutant emission inventory guidebook 2016 37

 7. Spatial mapping of emissions

This data can be accessed from: http://www.mapcruzin.com/free-europe-arcgis-maps-shapefiles.htm

4.3.10 ICAO and EUROCONTROL

Airport statistics for major airports can be obtained from the International Civil Aviation Organisation
(ICAO) website if country-specific data are not available. See www.icaodata.com/Trial/WhatIsICAO.aspx.
Another interesting point of contact with respect to airport statistics is EUROCONTROL
(http://www.eurocontrol.int/).

Airport statistics can be used to distribute nationally calculated emissions to different airports to estimate
LTO emissions for each.

4.3.11 Shipping: Lloyds Register

The Lloyds Register contains detailed ship movements data that can be used to distribute emissions from
shipping. A fee is required to access the data.

See: http://www.lr.org/

4.3.12 COPERNICUS

COPERNICUS, the European Earth Observation programme, (previously called GMES) will provide,
accurate, timely and easily accessible information to improve the management of the environment,
understand and mitigate the effects of climate change and ensure civil security.

See http://copernicus.eu/

EMEP/EEA air pollutant emission inventory guidebook 2016 38

 7. Spatial mapping of emissions

5 Overview of available spatial emissions data (updated 2013)

Table 2: Overview of available spatial emissions data. For each emission inventory the following information is provided: (i) Last and foreseen update, (ii) Sectoral,
spatial, and temporal resolution, (iii) Spatial and temporal coverage, (iv) Underlying emission and proxy data, (v) Developer. The gridding methodology
is discussed in a separate paragraph per inventory

Last Foreseen Sectoral Spatial Temporal Spatial Temporal Underlying emission

Inventory Reference Developer Data access
update update resolution resolution resolution coverage coverage data

Publicly
calculation of emissions
- 0.1 x 0.1 ° available
EDGAR 4.2 2011 Every 5 yrs IPCC sectors Annual Global 1970-2008 using a technology based http://edgar.jrc.ec.europa. JRC
- per country Link to data
emission factor approach eu/index.php
download

Available on
Emissions reported by
2003-2007 request
0.125° Europe European Member States
TNO-MACC 2007 2012 SNAP01 sectors Annual (update to incl. Kuenen et al. (2011) TNO –
x 0.0625° (EU+ non-EU) + gapfilling using IIASA
2008-2009) fee may be
and TNO data
required

Publicly

5 x 5 km² Europe available

E-PRTR diffuse NFR_09 and CRF Emissions reported by University of
2011 Not known yet 1 x 1 km² on Annual (EU27 and EFTA 2008 Theloke et al. (2009) Link to data
emissions sector European Member States Stuttgart
request (EEA) countries) download

1 x 1 km²
Both spatial Available on
Belgium; 5 x 5
distribution and Emissions reported by request –
E-MAP 2012 SNAP sectors km² Central Annual Europe 1990-2009 Maes et al. (2004)
national totals: European Member States fee may be
Europe; 60 x 60
yearly required
km² Europe

EMEP/EEA air pollutant emission inventory guidebook 2016 39

6 References
CEIP, 2010 – Technical Report CEIP 1/2010, Inventory Review 2010, Review of emission data reported
under the LRTAP Convention and NEC Directive, Stage 1 and 2 review - Status of gridded data and LPS
data.

CLC2000, 2010 – EEA: CORINE Land Cover 2000 100 m (CLC2000)

CLC2006, 2010 – EEA: CORINE Land Cover 2006 100 m (CLC2006)

Dore et al., 2001 – IMPRESAREO - Improving the Spatial Resolution of Air Emission Inventories Using
Earth Observation Data, WP 5000: Development of the Method for More General Application to Urban Air
Quality Issues; WP6000: Evaluation, Validation and Refinement of Spatially Resolved Inventories for a
Range of Urban Pollutants. AEAT/ENV/R/0462, Culham.

Dore et al., 2004 – APMOSHERE, Air Pollution Modelling for Support to Policy on Health and
Environmental Risk in Europe (EVK2-2002-00577)). Emission Mapping Methodology Progress Report
(January 2004). AEA Technology, Netcen, UK

EMEP/EEA 2013, ‘1.A.3.b Exhaust emissions from road transport’ in ‘EMEP/EEA Air Pollutant Emission
Inventory Guidebook, 2013’, European Environment Agency, (www.eea.europa.eu/emep-eea-guidebook)

Entec UK, 2005 – European Commission Directorate General Environment Service Contract on Ship
Emissions: Assignment, Abatement and Market-based Instruments. Contract No:
070501/2004/383959/MAR/C1, Task 1 - Preliminary Assignment of Ship Emissions to European
Countries. Final Report August 2005. Entec UK Limited. p.19.

Gallego et al., 2010 – A population density grid of the European Union, Population and Environment
(2010), 31: 460-473.

GISCO, 2010 – Geographic Information System of the European Commission (EUROSTAT),

http://epp.eurostat.ec.europa.eu/portal/page/portal/gisco_Geographical_information_maps/introduction,
2010.

Gkatzoflias et al., 2007 – COPERT 4 model (http://www.eea.europa.eu/publications/copert-4-2014-

estimating-emissions), European Topic Centre on Air and Climate Change (ETC-ACC).

Kuenen et al., 2011 – TNO report, TNO-060-UT-2011-00588, MACC European Emission Inventory for the
years 2003-2007.

Maes et al., 2009 – Atmospheric Environment 43 (2009) 1246-1254, Spatial surrogates for the
disaggregation of CORINAIR emission inventories.

Theloke et al., 2009 – Final Report Diffuse Air Emissions in E-PRTR, Methodology development for the
spatial distribution of the diffuse emissions in Europe.

TRANSTOOLS, 2010 – URL: http://energy.jrc.ec.europa.eu/transtools/index.html, 02.04.2013.

TREMOVE, 2010 – URL: http://www.tremove.org/documentation/index.htm, 02.04.2013.

EMEP/EEA air pollutant emi

7 Point of enquiry
Enquiries concerning this chapter should be directed to the co-chairs of the Task Force on Emission
Inventories and Projections (TFEIP). Please refer to the TFEIP website (www.tfeip-secretariat.org/) for the
contact details of the current co-chairs.

EMEP/EEA air pollutant emi