The approach should be to prioritise data requirements and to meet different needs on different time-scales. The following sections show the areas in which prioritisation can help meet the targets. These priorities can be described as follows:

  1. Essential. This is the most important information that must be included in any inventory produced by the EEA.
  2. Desirable. These data items should be included but their priority is lower. These are items that will become available on a longer time frame. For example, the full documentation of an inventory is required but can be published after the results become available.
  3. Useful. Some items of data would be useful to have but are not necessary to the overall project. For example, all the details of a power plants boilers are not needed at the European level and so this data would be useful but not essential or desirable.

6.1 Inventory Aims

The aim for Inventories in the past has been to be complete, consistent and transparent. These are defined as:-

  • "Complete." This means that the inventory should include all sources of each pollutant. In an absolute sense we can never be certain that all sources are included, but we must aim to be as complete as possible. Any source we can find must be included. However it is clearly most important to expend most effort on significant sources and not to spend a lot of time on sources that will not affect the result significantly.
  • "Consistent." Each contribution to the final inventory should be comparable between countries. It is important that differences between countries reflect reality. Thus differences between countries in emission factors or methodologies should reflect actual differences, for example in the emission rates, technology, control or operation. This does not mean that identical methodologies need to be used by all countries.
  • "Transparent." In order to demonstrate that a inventory is complete and comparable it must be transparent. The IPCC states that "enough data should be provided to allow a third party to reconstruct the inventory from national activity data and assumptions" (this is their working definition of transparency). They go on to state that, to limit the volume of data, any documentation should focus on describing fully any differences on method and assumptions from the default method. We propose to follow this. The default method is laid down in the EMEP/CORINAIR Guidebook on Emission Inventories whose first edition will be published in June. This guidebook has already been compiled along SNAP codes. This does not imply that the emission factors should be identical in each country, nor that the same methodologies should be used. Transparency can exist even if each country used completely different database systems.

It is clear that there is a fourth requirement, they must be timely. This is perhaps the most important requirement of all, as inventory data must be produced in time to be useful to policy makers. This requirement may conflict with some of the above. Air Emissions '94 should have as its essential requirements:-

  • Timeliness
  • Consistency
  • Completeness

A desirable requirement is to be transparent.

Completeness and consistency are needed in any inventory for it to be of use. It needs to arrive on time. The transparency can come later. For example, transparency may be desirable in the first year and essential after two years. The involvement of the topic centre in the inventory gathering process should assure users of the validity of the data.

6.2 Emissions Data

Many of the needs are for national total data not spatially disaggregated information. Data should be prioritised to ensure that the most urgent demands are met first.

The types of information can be classed, in order of importance, as follows:-

  • National totals. These are the most important pieces of information for policy purposes. They are essential. Preliminary provisional estimates should be provided within six months, at least for CO2.
  • Sectoral breakdown. This is needed to understand the sources of emissions. Some of this will need to be worked out to produce the national estimates and time series data. This is essential at the 11 sector "EMEP/CORINAIR" level. It is essential to have a more detailed sectoral split for other purposes. However this should not be an unlimited request for data.
  • Time series. These are needed to indicate the ways in which emissions are changing. They indicate how a country is moving to meet its emission and environmental targets. These are essential. They will be built up year by year by the topic centre.
  • Spatial distribution. Here the spatial location of the emissions is determined. This is required to provide input into modelling exercises so that the transport and deposition of pollution can be studied and understood. Major source areas will be identified and the potential of targeted control measures can be assessed. Judged against the other needs from an inventory this requirement can be regarded as desirable, i.e. it can be met on a longer time-scale.

To meet user requirements, the data to be collected should be grouped as follows. (Over time the classification may change as user needs alter.)

  • Group 1. Provisional data for main pollutants within six months. This should include CO2. Although only CO2 is required by the EU greenhouse gas reporting requirements within seven months, the energy data that has to be collected to do this will enable SO2 and NOx to be done at the same time with very little effort. The data required is given by the IPCC minimum tables. Although countries outside the EU are not obliged to meet this deadline it should be possible for them to provide data on this time-scale. Both Bulgaria and Slovenia were able to meet this deadline in 1994. Details on how this may be achieved are given in chapter 7.
  • Group 2. National level emissions by eleven source sectors within 12 months. All the eight pollutants covered by CORINAIR 90 should be included here even if only provisional results can be given for some of them. Table 10 shows an example of this type of data.

Table 10 Czech Republic CORINAIR 90 emissions by 11 source sectors (tonnes).

1 Public Power, cogeneration and district heating 1162943 321247 4327 4326 27639 64963000 8550 0
2 Commercial, institutional and residential 458378 103248 58164 58094 500655 51085000 5562 0
3 Industrial Combustion 173266 144906 3487 829 275856 27127000 1865 0
4 Production Processes 61293 8039 28747 1420 82545 747000 2694 2400
5 Extraction and Distribution of fossil fuels 0 0 5585 844842 0 0 0 0
6 Solvent Use 0 0 93023 0 0 0 0 0
7 Road Transport 0 142675 52588 2605 143081 7667000 784 109
8 Other mobile sources and machinery 5862 51383 7247 341 9067 9121000 90 3
9 Waste treatment and disposal 888 1056 324 34258 5124 757000 101 0
10 Agriculture 0 1 6 507520 58 0 25898 88298
11 Nature 0 50 40180 94663 1 2060000 16350 0
Total 1862630 772605 293678 1548898 1044026 157527000 61894 90810


  • Group 3. Updated data for all pollutants within 18-24 months at the latest. This should include all 8 pollutants described above together with any new pollutants e.g. heavy metals and POPs in Air Emissions ‘94. In the future it should be the aim to promote these extra pollutants to group 2.
  • Group 4. Disaggregated emissions at least once every 5 years completed within 24 months. The first three groups provide data at national level only. This part is analogous to CORINAIR 90. The four year frequency comes from the EMEP requirements. The EEA may wish to reduce this interval. An alternative to reducing this interval would be to use interpolation and extrapolation techniques based on national totals to produce estimates for intermediate years. This should be possible as the large point sources have to have their emissions reported each year and the remaining emissions are unlikely to change too rapidly in four years.

Figure 2 shows the stages in the process and where the Emissions Inventory Topic Centre will provide specific input and support.

Figure 2 - Proposed process for Air Emissions 94

figure2.gif (12532 bytes)

6.3 Principles

Inventory compilation should be tailored to meet the user’s needs and to minimise the effort required by individual countries. The main points are outlined below:-

  • Data should be collected in the individual countries by national experts in a similar manner to CORINAIR 90. (The national expert is the person nominated by the National Focal Point to actually create the national inventory. They would have a similar role to the CORINAIR 90 national experts and may be the same people.) This has a number of advantages. Firstly it utilises the existing expertise and knowledge in a country and secondly it can ensure the acceptability of the estimates. National experts should know, or be able to find out, the technological and abatement changes that are occurring in their countries. This will not be apparent from international statistics. It will be important to ensure the necessary level of effort by the individual states. While the Emissions Inventory Topic Centre will give the support it can, the process will also need the commitment and support of the EEA and national governments.
  • The Topic Centre should provide much more individual assistance than was available in CORINAIR 90. This should be in the form of one-to-one meetings or very small groups of countries with similar situations. This help should cover not just the mechanics of using any software but also assist in producing data that are complete, transparent and consistent with the rest of Air Emissions '94. In the early stages assistance will need to focus on assisting in the production of provisional estimates to meet the tight deadlines.
  • Emissions data should be provided in stages to the Emissions Inventory Topic Centre. This would provide timely data to meet the needs outlined in Section 5.3. If the Emissions Inventory Topic Centre is in frequent contact with the individual countries’ experts then there would be more confidence in the data because part of that contact would be devoted to a checking and validation role.
  • The Emissions Inventory Topic Centre, together with each country, should estimate emissions based on international energy statistics. As this will be separate (to some extent) from each countries’ more detailed estimates this will provide a first stage check on the reasonableness of the data. While the two estimates should be consistent the one based on international energy statistics will not be able to capture all the detailed information that is available to the national experts. In the absence of data for a particular country this would also provide a ‘default’ emission estimate. Any emissions software used should be able to total fuels used to enable easy checking against national energy balances. Hence, in some cases it is an integral part of the process and in some it is separate.
  • The CORINAIR 90 software is too complex for many of its users data needs. As the first call is for provisional national level data, much simpler systems can be considered. To collect initial, provisional data, countries’ existing systems should be considered in order to meet the tight deadlines for the provision of data. In addition, this will reduce the additional effort required by each national expert. Countries should be encouraged to make their internal systems compatible with Air Emissions '94. The Emissions Inventory Topic Centre should consider the provision of specific tools to assist this process. They are unlikely to be available immediately and so can be specified and developed in the light of experience and agreed priorities.
  • Use as much of the EMEP/CORINAIR Task Force Guidebook as possible. While time-scales may limit this, it should be possible to use much of the guidebook for Air Emissions '94. Not only should this provide the best default methodology it will also test the guidebook and any problems should be fed back to the Task Force so that the guidebook can be refined. In the longer term the experience of Air Emissions '94 and other work of the Emissions Inventory Topic Centre will be used in improving the guidebook’s methodology and emission factors. All the outputs should be in UNECE reporting formats.
  • Make biomass and greenhouse gas sink definitions and methodologies compatible with the IPCC definitions. As the UNECE Task Force on emission Inventories has agreed to use IPCC definitions and default methods for these sectors Air Emissions '94 should do the same. This is in fact an extension to the previous point. In areas where the IPCC and CORINAIR 90 source definitions conflict then extra data will be needed to enable both needs to be met.
  • Complete energy use data should be collected. This would enable the emissions database top be checked against energy balances. This would provide each user an initial check on the completeness of energy related emissions.
  • It will be necessary to prioritise changes to the software so that the most important changes are made quickly because Air Emissions '94 needs to start collecting data as soon as possible for completion by the end of December 1996. The revised software needs to be developed, checked and distributed by the end of December 1995. The system will be made compatible with as much of the EMEP/CORINAIR Task Force guidebook as possible in the time-scale (the guidebook is based on SNAP90). Draft chapters are available so a start could be made now. (The first published version will be available before the end of the year.) Therefore it is proposed to base Air Emissions '94 on the existing software, modified where necessary. Annex D gives details of the proposed changes. The Emissions Inventory Topic Centre will not distribute empty files but attempt to complete the tables with existing data. For example, emission factors, ‘rubrics’, fuel definitions and surrogate data are unlikely to have altered much since CORINAIR 90. There is also the possibility that default emission factors from the EMEP/CORINAIR Guidebook could be entered into the database. Users would then only need to alter factors that are different in their country. This would leave 6 months (January to June 1996) for the data collection phase with the final data being collected within 2 years after the end of 1994. After June 1995 there will be work in parallel on development and on inventory collection to provide national level data for 1994 and 1995.
  • Software documentation needs to be improved. In particular it needs to be expanded to include a detailed explanation of each algorithm used. This would enable users to correctly assess any advantages to be gained by using the software.
  • National emission estimates and trends are important. These are important information and should be provided on a faster time-scale than the spatial distribution. Ideally this information should be provided as the first step in producing the full spatial map of emissions. Initially it may be necessary to do this as a separate exercise based on national systems and work to a closer integration in the future.
  • It is vital that all parties recognise that these data will all be estimates. They will be the best estimate of emissions that can be made in the time available. There is no point in wonderfully precise data that is ten years old when it is disseminated. At any time in the future the Emissions Inventory Topic Centre will accept revisions to earlier data where these can be justified. The important task is to arrive at consistent time series data. There will need to be a clear procedure to manage updates and revisions. This will be developed by December 1995.
  • Separate tools similar to COPERT should be developed for specific tasks. Such tools could include:-

Power Plant

Paper Pulp



Several of these tools are under development by Expert Panels of the UNECE Task Force on Emission Inventories and they will be available as part of a default methodology but are not intended to replace more sophisticated national approaches.

  • Efforts should be made to develop SNAP90’s compatibility with socio-economic statistics and abatement technologies. The Topic Centre will conduct a review of this problem. It will aim to produce interim proposals by June 1995. This would allow its results to feed into the software development. In the longer term there may need to be more far-reaching changes to the database systems to include this information. Alternative source sector classifications to explicitly describe technologies and controls used in each source sector will be investigated in association with Eurostat. This review will also need to take into account the needs of integrated inventories. It could look at, for example, the ACCOR nomenclature which is being developed with integrated inventories in mind.
  • There should be some form of ‘user forum’ where representatives of the different user groups can discuss the progress of European inventory work and provide input into its development.
  • The central EEA Emissions Inventory Database should not hold confidential data. As this data cannot be distributed to other users, its usefulness is very limited at that level. One possible solution to the problem is that each countries database should hold all the data including confidential data. Countries would then transfer to the central database only the data that is not confidential. This would be done in co-operation with the Air Emissions Topic Centre. This is outlined in Figure 3.

Figure 3 - Distributed databases for confidential data

figure3.gif (3966 bytes)

This is similar to the CORINAIR 90 system except that all the data was transferred to the central database. The country databases could potentially be held in each country and accessed electronically at an appropriate time by the Emissions Inventory Topic Centre for validation and verification. Alternatively they could be held centrally with the need for the EEA to provide extra security to ensure confidentiality.

The Emissions Inventory Topic Centre may not need access to all the data to validate the inventories. Paper records could be used. Data ranges and summaries could also be used for validation if the complete datasets were not open.

The master database could, in principle, be situated anywhere with networked links between the EEA, the Emissions Inventory Topic Centre partners, the NFPs and NRCs and any other users. However it would be appropriate that the database were located at the EEA. This would provide the EEA with the quickest access to the data especially as the form of the connection of the EEA to international networks has not been decided.

Confidentiality is a complex issue, with different requirements in each country . The Topic Centre will look at a range of possible solutions.

lijn.gif (900 bytes)

up.gif (859 bytes)
left.gif (869 bytes) right.gif (869 bytes)


Document Actions