"Generation and Evaluation of Emission Data"


  approved by SSC on February, 20th,1997

approved by IEC on June, 24th, 1997



1 List of Principal Investigators

2 Aims of GENEMIS

3 Executive Summary

4 Project Description

5 Organisation of the subproject

6 Proposals for contributions to GENEMIS


1 List of Principal Investigators

Task 1: Measurement of emission data

Dieter Hassel: Analysis, development and evaluation of emission models and data sets for road traffic in Germany and Europe

Werner Engewald: Determination of emission factors of volatile organic compounds from the residential burning of brown-coal briquets (lignite)

Klaus Schäfer : Determination of area source emissions from agriculture and of precursors of photo-oxidants using remote sensing techniques and inverse dispersion modelling

Åke Sjödin: Influence of Maintenance and Driving Behaviour on Ozone Formation Potential of Real-World Road Vehicle Emissions

Gabriele Bruckner : Emission rates of NH3, HNO3 , N2O and particles from agriculture, industry and traffic


Task 2: Evaluation of emission data

Guenter Baumbach: Evaluation of calculated emission data by air pollutant flux measurements around agglomerations of emission sources

Dieter Klemp: Examination of anthropogenic emission inventories by ground-based measurements of relevant ozone precursors

Michael Moellmann-Coers: Determination of man made emissions of trace gases using the ‘source tracer ratio method’

Michiel Roemer: Estimates of VOC emissions in the Rijnmond area based on ambient air measurements

Johannes Staehelin: Evaluation of an emission inventory of a city by field measurements

Franz Slemr: Validation of emission inventories of a city area by comparison with measurements

Szabina Török: Experimental validation of emission data

Task 3: Update of emission data

Kristin Rypdal: Validation of changes in emission inventories considering statistical input and indicators

Wilfried Winiwarter: Application of economic indicators on emission inventories in order to obtain emission updates

Task 4: Emission modeling/European and regional scale

Adolf Ebel: Application of the EURAD model for GENEMIS

Rainer Friedrich: Development of methods and models for the generation of European emission data for episodes

Aysen Müezzinoglu: Preparation of emission inventory of Turkey

Gerhard Smiatek: Mapping Land Use for Modelling Biogenic and Anthropogenic Emissions

Rainer Steinbrecher Biogenic emissions

Roberto San Jose: Automatic Satellite land use classification for biogenic emissions in mesoscale air quality models (BIOEMI)


Task 5: urban emission inventories

Rémy Bouscaren: MEHARI - Modelisation des Emissions de Polluants dans l'Atmosphère avec une Haute Résolution d'Inventaire

Jose M. Baldasano: Development of an upgrade urban source emission model for atmospheric pollutants

Rainer Friedrich Evaluation of urban emission inventories

Peter Sturm: Improved emission inventories for the purpose of modelling air pollution

Wilfried Winiwarter: Development of a high performance emission inventory for assessing urban plumes; further contributors to this task (SATURN PI’s): Carlos Borrego, Giavanna Finzi, Eugene Genikhovic, Steinar Larssen, Nicolas Moussiopoulos, Jan van Rensbergen

Task 6: NH3 emissions

Willem A.H. Asman: Modelling diurnal and seasonal variations of the ammonia emission

Pierre Cellier: Estimating ammonia emissions and deposition at field and regional scale

Nick Hutchings: Modelling the emission of ammonia from agriculture

Mark A. Sutton: Mapping of atmospheric ammonia emissions on regional and local scales

2 Aims of GENEMIS

General aim of GENEMIS is to support the generation of emission data, that can be used for the development of air pollution abatement strategies in Europe. This includes

  • the improvement of methods, models and emission factors for the generation of emission data,
  • the assessment of the accuracy of emission data and
  • the development and improvement of tools, that generate emission data for atmospheric models.


3 Executive Summary

To support the aims of EUROTRAC-2 accurate emission data are needed. These emission data should have special features, especially a high temporal and spatial resolution. Emission data for current and future years should be provided and the accuracy of the data should be known. In addition to emissions of SO2, NOx and CO reliable information on VOC components, NH3 and particulates - classified according to size and chemical composition - should be available. The accuracy of the emission data should be assessed.

During EUROTRAC-I, due to work within its subproject GENEMIS the understanding of the temporal and spatial resolution of emission has been improved considerably. However, a number of other important issues related to the generation of emission data is still unsolved. These include especially the following topics, which will be treated in the new GENEMIS project:

A number of emission factors for important emission sources are still very uncertain. So, for these sources, measurements for deriving representative emission factors will be carried out.

It is still not state of the art to quantify the accuracy of emission data. So, important tasks are

- the development and application of methods to assess the variation/range of the calculated emission data;

- a comprehensive validation of emission data by comparing calculated and measured data; this includes as well the use of data from already existing monitoring stations as special measurement campaigns in the vicinity of agglomerations of emission sources. Such campaigns are organised as subtasks of this subproject with limited duration.

For short term predictions of ozone episodes actual emission data are needed. Therefore methods to forecast (preliminary) emission data for current episodes (emission updates) from existing inventories for past years will be developed resp. improved.

Changes and improvements of the methods and models to calculate emissions will be developed and tested. This includes the improvement of calculations of biogenic emissions. With the improved methods, emission data as needed for the other EUROTRAC II projects will be generated and supplied. In addition, the improvement of the data transfer and the development of emission modules, that can be coupled directly to the atmospheric models and that can take over part of the emission calculations, will help to improve the interface between emission calculation and atmospheric modeling.

Methods for the generation of emission data for particulate matter (including information on the chemical composition on the distribution of the particle sizes and for NH3 will be de-veloped.

To be able to explain the development of measured ambient concentrations of air pollutants in the past, the development of emissions in Europe in past decades has to be investigated.

Once atmospheric models are developed and validated, the next step is to apply them to explore future trends in air pollution and to identify efficient air pollution control strategies. As input data for these analyses scenarios of future emissions and emission reduction scenarios have to be developed and provided.


All these tasks will be carried out based on existing inventoties (e. g. CORINAIR) and in close cooperation and coordination with other activities in the field of emission inventorying (e. g. IGAC, EEA/CORINAIR, EMEP, LOTOS).


4 Project Description

4.1 Introduction: the Need for Reliable Emission Data

Although progress in the implementation of emission reduction measures has been achieved in Europe in the last decade, air pollution is still a major problem in Europe. All over Europe, the formation of photochemical oxidants is causing damage to human health and significant reductions in crop yield, the deposition of acidic substances and nutrients is too high for many sensitive ecosystems and there are indications, that the damage to human health caused by aerosols is currently underestimated. In parts of Central and Eastern Europe, the extensive use of coal and lignite in power plants and heating systems causes severe air pollution problems.

Estimated damage due to air pollution in Europe - if expressed in monetary terms - adds up to hundreds of billions of ECU. This underlines the necessity for further reductions of emissions of air pollution. However, increasing emission reduction requires more and more expensive measures. Thus, air pollution control gets more and more in conflict with other aims of society. With increasing control costs it will get more difficult to achieve acceptance for the necessary measures; so, it will be increasingly important to establish proof,

  • that the fullfilment of environmental goals is assured and
  • that the emission reduction is reached efficiently, i.e. with the least possible costs.

One prerequisite to show this is the improvement of scientific knowledge about source-receptor relationships for photooxidants, acidifying substances and aerosols. For that, obviously accurate information on the sources, i.e. on the emissions of the precursors, is needed. Furthermore, the only way to show in advance, that emission reduction measures are efficient and successful, is to apply carefully validated atmospheric models, that are able to calculate the expected changes in ambient air concentrations and deposition of pollutants due to changes in emissions. For validating and applying these models, accurate emission data in high temporal and spatial resolution are needed; this includes current and future emissions and information on the accuracy of the data.

It is the aim of GENEMIS to improve the scientific basis for generating emission data, to assess the uncertainty of emission data and to provide emission data with features according to the requirements of the users of emission data in other subprojects of EUROTRAC-2.

4.2 State of the Art

During the last years, several European emission inventories have been set up. These include ECE-EMEP, TNO-LOTOS and EC-CORINAIR. The LOTOS inventory is not institutional, it was designed to support the Dutch LOTOS project. The EMEP and CORINAIR inventory are currently harmonized, so that the future source for European annual emission data will be CORINAIR/EMEP. It is planned, that data for the year 1994 will be available in 1997.

However, the currently available inventories are not able to provide emission data with the features as needed for the validation and application of the EUROTRAC atmospheric models. So, within EUROTRAC I, the subproject GENEMIS(1) was set up to generate emission data with a higher temporal and spatial resolution than usually available. The application of the developed methods showed, that strong variations of emissions with time occur and that these variations can be estimated by using available socioeconomic, technical and meteorological data. The understanding of the temporal and spatial resolution of the emissions of NOx, VOC and SO2 was improved considerably.

The data base of the emissions in Central and Eastern European countries was improved with the help of a network of scientists in these countries.

In addition, reliable VOC-split factors were collected and put at the disposal of the EUROTRAC community. In a large tunnel experiment in the Gubrist tunnel emission factors of NOx, CO and VOC-species were measured and compared with the GENEMIS emission factors. A good agreement - within the uncertainty bounds of the emission model calculations and of the measurements - between measured and calculated emissions could be proven for gasoline powered cars.

Other achievements include the improvement of methods for the update of emissions and the test and improvement of international data bases, esp. CORINAIR and LOTOS.

4.3 Work Programme

4.3.1 General Remarks

Whilst in the former GENEMIS(1) project the main emphasis was on the temporal resolution of emissions, main topics now are the assessment of the accuracy of emission data on the basis of measurements, the investigation of further species (NH3 and particulates) , the development of harmonized methods to generate urban emission inventories and the measurement of still uncertain emission factors. So, although the acronym of the former and the new GENEMIS project remains the same, the content - and also the full title- are quite different.

The studied compounds are derived from the scientific tasks of EUROTRAC-2. As photooxidants, acidifying substances and aerosols shall be investigated, emissions of SO2, NOx, CO, NH3, VOC (split into components as needed by chemical models) and particulates (including information on size distribution and composition) are analysed. The investigation focuses more on NH3, VOC-components and particulates, as the uncertainty of emission data for these species is in general larger than for the other species.

Of course, all emission sources of the compounds mentioned above have to be regarded. However, for biogenic emissions only model developments and calculations, but no measurements are planned, as this topic is covered by the BIATEX subproject.

Within GENEMIS, it is planned to use as much as possible existing information on emissions, e.g. the CORINAIR inventory, the ongoing work of the UN/ECE Task Forces and various results from research of the European Commission. Instead of creating own data bases it is planned to generate models and data, that build on the existing data bases and disaggregate, update and improve them. A close cooperation with the EEA and the UN/ECE is strived for. One of the advantages of this procedure is, that - after modules for e.g. a temporal resolution of emission data in CORINAIR structure have been developed within GENEMIS - they can be applied to future updates of the EC and ECE inventories. So, these benefits of the GENEMIS work can be preserved to times after EUROTRAC-2 has ended.

In the following, the working tasks are described in detail.

4.3.2 Measurements of Emission Factors

For some of those emission factors, that are considered quite uncertain and that are relevant (i.e. belonging to a source category which causes a considerable share of total emissions), measurements are made to improve the accuracy of the factors. According to the larger uncertainty of emission estimates of VOC components, NH3 and particulates, measurements focus on these compounds. In the beginning, measurements of VOC components will be made. Source categories, of which emission factors will be measured in the first phase of the projct (until middle of 1999), include:

  • VOC-split of car and truck emissions:

Sampling measurements and a tunnel experiment for determining VOC components from cars and trucks are performed to enhance the accuracy of available emission factor data bases.

  • *VOC-emissions and split for lignite firings:

Combining the information obtained from Gas Chromatography - Mass Spectroscopy coupling, selective detection, and chromatographic data as independent identification methods the exhaust gases of lignite stoves are examined.


  • VOC-emissions from gasoline stations:

By means of active Fourier Transform Infrared Spectroscopy (FTIR) and Differential Optical Absoption Spectroscopy (DOAS) the concentrations of VOC in the exhaust gas plumes of gasoline stations are measured.


In further phases, particulate emissions from transport (e. g. exhaust gases from diesel engines, tyres, road dust) and NH3 emissions after manure spraying will be analysed.


4.3.3 Evaluation of Emission Inventories with Field Measurements


The accuracies of emission inventories are generally difficult to assess because they are based on many informations of different origin with variable and sometimes unknown uncertainties. Estimates using the classical propagation of errors are difficult to apply because uncertainties of several important variables are not well known. The gained insight in uncertainty ranges might be incomplete, as systemic errors might occur (e. g. unknown sources, errors in base data, missing information) and as often the same methods, emission factors a.s.o. are used for the estimation of emissions in different inventories. So, the comparison of estimated emission data with measured data is indispensable.

Ambient air measurements of CO/NOx and NMHC/NOx concentration ratios in several American cities showed differences up to a factors of 2 to 7, respectively, to ratios derived from emission inventories (National Research Council: "Rethinking the Ozone Problem in Urban and Regional Air Pollution", National Academy Press, Washington D.C., 1991). Similar and more detailed studies for European cities are rare. In almost all studies with information on VOC/NOx only NMHC were reported and the obtained VOC/NOx ratios thus neglect the emissions of oxygenated compounds such as aldehydes and ketones, which are produced by inefficient fossil fuel burning, and alcohols and esters, which are frequently used as solvents. In addition to road traffic emission, many other individual emission sources are relevant in a city area such as evaporation of gasoline, evaporation of solvents, emissions from residential heating, and industrial emissions. Some of them are less well characterized both in the fingerprint of the released compounds and in the total emitted amount. The complex emission structure of a city area poses several questions with respect to the evaluation of emission inventories: a) how complete are the sources considered in complex emission models, b) if complete, do the models reproduce correctly the emission fingerprints and strengths, and c) do the emission models reproduce correctly the temporal variations of both emission fingerprints and strengths.

Indications of deficiencies in the current emission models such as mentioned above demonstrate the need of experimental validation of the emission inventories. Experimental validation of emission inventories from a city area is thus one of the major objectives of GENEMIS. The accuracy of emission inventories of a city can be directly evaluated by comparison of estimated emissions with emissions determined by field measurements. For experimental determination of emissions from a city, however, suitable experimental strategies have to be developed at first and, if possible, intercompared to assess the uncertainty of the experiment. Three experimental strategies for determination of emissions from a city are planned to be developed and used within GENEMIS.

1. Determination of emission fluxes from a city from the divergence of downwind and upwind fluxes. The first study of this type is planned for Augsburg (Germany). Using aircraft, tethered balloons, air ships, radio sondes, ground stations, and remote sensing techniques, distribution of chemical species and meteorological parameters upwind and downwind of a medium sized city will be determined. The emission fluxes will then be calculated from the divergence of the inflow and outflow fluxes.

2. Determination of emission fluxes using known trace emissions.

A first study of this type is also planned for Augsburg. A tracer (SF6) will be released with a known release rate at several sites distributed over the city area. Using aircraft and ground based stations, concentration ratios of individual substances to the tracer will be measured downwind of the city. The emission fluxes of the individual substances will then be calculated from the observed concentration ratios and the tracer release rate. This experimental strategy has been used with success for small towns with about 50.000 inhabitants and seems to be suitable also for determination of the diurnal emission variations. One objective within GENEMIS 2 is to extend this experimental strategy to larger cities. The other objective is to define the range of meteorological conditions amenable to the use of the tracer technique.

3. Determination of relative emission fluxes from concentration ratios of individual substances. Concentration ratios of individual substances downwind of the city reflect the ratios of substance emissions within the city and have to be consistent with the estimated emission inventories. This experimental strategy can be used for larger cities where it is difficult to apply the tracer and the divergence techniques because of the not well defined boundaries of the city area and for logistical reasons. Studies of this type are planned for the cities of Augsburg and Zürich (Switzerland). A city plume study is also planned in Hungary.

The experiments will be accompanied by development of analytical techniques for determination of oxygenated and halogenated substances, in addition to the currently measured NMHC compounds. The extension of the spectrum of measured VOC compounds is important for application of Chemical Mass Balance models (CMB) and other techniques for apportionment of observed emissions to the individual sources. Alcohols and esters may be particularly helpful for determination of the solvent contribution to the VOC emissions. Charaterization of additional source fingerprints (relative pattern of VOC composition) is another task of the experimental work within GENEMIS 2.

The combination of all three of the above mentioned methods within the planned ‘Augsburg’-experiment gives the opportunity to compare and assess the different methods.

4. Receptor modeling based on data from existing monitoring stations

Receptor models based on principal component anlysis will be applied to use measured ambient concentrations from monitoring stations to determine the dominating emission sources and the related source profiles. The results will be compared with emission model results.


4.3.4 Emission Update for Obtaining a Contemporized Emission Data Base (Real time emission inventories)

Creating an emission inventory is usually a very time consuming process.. In addition, updates are only performed every few years. For the CORINAIR inventory, the results from base year 1985 were finally published in '95. At about the same time already the '90 data became available, which is still the most recent spatially disaggregated data base available for Europe. For CORINAIR '94 it looks as if the time lag may be reduced to three years. Only starting from this information, the emissions for specific periods can be retrieved. This may be good enough for a general trend assessment, or for model evaluation using available experimental data as from routine monitoring. For a number of applications, however, what is needed is emission data for the actual year. Such applications include models accompanying field measurements, but also photochemical models used for ozone forecasts. In contrast to emission projections (scenarios), this activity does not deal with future emission data. Emissions are reported at the same time they occur, thus they are termed real-time or contemporized emission inventories.

As in the creation of the emission inventory as such, emission changes are constituated by changes in two parameters. Firstly, There are the technological changes, which are reflected in a changing emission factor of an activity. Such a change normally occurs gradually, with the turnover of old equipment being replaced by new one.

The second important influence comes from changes in the activities, that cause the emissions.

In order to transform the source activities in a way corresponding to classical emission updates, firm statistical parameters have to be selected which make up for the lacking detailed data on the European scale, EUROSTAT data will be used. Actual activity data may be gained by trend analysis.


The following working steps will be carried out:

  • Creation and selection of applicable statistical indicators
  • Application of contemporized inventory to a selected test area (local/urban scale)
  • Application of contemporized inventory to a large test area (e.g. European Union, CORINAIR inventory)
  • Validation of statistical indicator's relevance to emerging economic statistics
  • Validation of emission data taking advantage of future emerging emission inventories (CORINAIR '94, CORINAIR '98)


4.3.5 Development, Improvement and Application of Methods and Models for Calculating Regional Emission Data

A) Development and Improvement of Methods

The standards and requirements, that emission data have to meet with respect to accuracy, spatial and temporal resolution, sectoral disaggregation, species a.s.o. are constantly growing. Furthermore, new knowledge about emission processes and factors and new activity data are created as well within GENEMIS as within other projects is steadily arising. So, the currently used methods and models for generating emission data with high temporal and spatial resolution should be revised, improved and completed. Methods will be adapted to improved activity data and new emission factors.

For improving the calculation of biogenic emissions, BVOC emission algorithms, that are based on meteorological and biophysical parameters combined with land use data (ecosystem and plant specific) and probably correction terms for the turbulent transport and air chemistry will be developed. The resulting emission data will reflect the specific European conditions.

Land use data are important input parameters for the estimation of anthropogenic and biogenic emissions. To meet the requirements of the emission models, the thematical and geometrical resolution of land use data in existing data bases will be improved.

B) Improvement of the Interface between Emission Models and Atmospheric Models

The amount of data that has to be transfered from the emission model to the atmospheric models is tremendous. To change this, the generation of emission data might be divided into two steps:

In a first step a set of base data is generated. For the preparation of this set the knowledge of emission experts, time consuming data collections and data analyses are needed.

This data set is then delivered to the users of the atmospheric models together with a software package. This enables atmospheric modelers to calculate emissions parallel to their model for the currently active time step using calculated ambient air temperatures of their models to generate emission data for temperature dependent processes.

Of course, the software for generating emission data with high temporal resolution could also be installed at authorities like e.g. the EEA.

C) Generation of European Emission Data

Starting from existing data (e.g. CORINAIR) the developed models are applied to generate the emission data, that are needed for the EUROTRAC-2 project, esp. for the models for transport and chemical transformation of pollutants.

Firstly, data from the existing emission data bases have to be obtained and transformed into the spatial units needed. Next, the data are updated and transformed into time-dependent data by means of the models developed. VOC components are allocated to classes. The results are emission data for episodes in 1994 and 1998.

D) Assessment of Accuracy and Variability of Emission Data

For the use of emission data in transport and transformation models, in addition to the emission data information about the accuracy of the data is needed. This information would help to clarify, whether differences between measurements and calculated model results stem from inaccuracies in the model equations or whether they might also be caused by inaccurate emission data (of course the interpretation of such differences is difficult, as measurements usually are made on one location, whereas model results are average data for a cuboid).

However, at present standard deviations of emissions are not given for any emission data base, mainly because information about deviations of activities or emission factors are not known. Besides, emission factors may not always be normally distributed, as sometimes a small part within a collective of emission sources may be sources with rather high emission factors.

To arrive at first estimates of the accuracy of emission data, the following procedures are possible and should be followed simultanously:

a) The ‘classical’ approach is, to try to determine a probability distribution for the parameters (activities and emission factors) used to calculate the emissions. This has to be done by analysing as many measured data as possible. Of course, it is not possible to treat all parameters, as for some of them available measurements are too few for a statistical analysis. The analysis should focus on those parameters, that are used for the estimation of the emissions of the most important source categories. Furthermore, for some parameters it might be necessary to use expert judgements of the possible band width of parameter values.

b) Another method, that gives some insight in possible uncertainties of emission data is the comparison of emission data, activity data or emission factors, that are calculated with different methods or for different plants or areas. If e. g. for a smaller area a very detailed emission inventory exists, the comparison of the results with those of a coarser inventory will give some hints on uncertainties. The comparison of the emission factors for different sources of the same source category might explain the variability of the factor.


4.3.6 Urban Emission Inventories

In several EUROTRAC subprojects, including SATURN, LOOP and also GENEMIS (see chapter 4.3.3), air pollution in urban areas is investigated. For these activities, urban emission data with high spatial and temporal resolution is needed. However, existing urban emission inventories are often highly uncertain, furthermore their quality varies for different urban areas. Whereas for the generation of regional and European inventories standardized procedure description exist (e. g. the Atmospheric Emission Inventory Guidebook of the EEA), no such standardized guidelines exist on the international level for urban areas. Therefore, to provide default emission factors and to develop tools and procedures for generating urban emission inventories and to ensure the quality of the data , the following working steps are carried through within this task:


A) Development of requirements and guidelines for urban emission models

As the projects in question are intended to be applicable at European level, characteristics of commonality and compatibility have to be taken into consideration when selecting models for computation within emission inventories. It will hardly be possible to capture the emission behaviour of all sources in the planned project areas across Europe with a single computational model since the variations in existing methodologies, available activity, and emission data as well as emission standards are simply too large. However, to ensure compatibility within the inventories used within EUROTRAC-2, minimum requirements have to be defined and applied. The requirements for the harmonised method include area under consideration, grid scale of the inventory, time scale of the inventory, pollutants treated within the inventory, definition of the input data for stationary sources, mobile sources, biogenic emissions, VOC-profiles for the sources, specification of the necessary software tools for emission modelling, specification of the necessary software tools for displaying the results, specification of the format for the output data. This assures compatibility within the project and offers the possibility to compare the emission inventories.


B) Comparison of urban emission inventories

Emission inventories which will be made according to the requirements will be analysed in order to point out regional differences and similarities. Within SATURN, LOOP and GENEMIS it can be expected that emission inventories from cities in Germany (Augsburg, Berlin), Spain (Barcelona), Greece (Athens), Austria(Graz), Italy(Milano), Norway (Oslo), Russia (St. Petersburg), Portugal (Lisboa) and other European cities will be established or used. The following tasks will be performed:

  • Analysis and assessment in order to distinguish particularities and define similarities.
  • Specification of the necessary software tools that are needed for such tasks (e.g. data bases and GIS).
  • Analysis and comparison of specific patterns of activity data collected at different urban areas (e.g. streets, roads, highway specific traffic loads, industry operation patterns, etc.).
  • Specific chapter for the road traffic emissions, which usually constitute a major part of emissions at urban scale.
  • Consideration of additional pollutants which are important at urban scale and not usually considered, such as particulate matter, heavy metals (lead, etc.) and reactive organic pollutants.
  • Estimation of the influence of different methodological approaches and input data used for the preparation of emission inventories.


C) Basic investigations concerning different methodological approaches and quality of emission data.

Although minimum requirements for emissions investigations will be defined as a first step different methodological approaches can be used in order to achieve the results required for air quality modelling. E.g. it is possible to use so called "top-down" approaches to calculate emissions from road traffic in a mesoscale application, it is also possible to use "bottom-up" models for the same case. Influences of such methodological aspects will be analysed.


4.3.7 Ammonia emissions

Ammonia (NH3) and its reaction product ammonium (NH4+) contribute both to eutrofication and to acidification. Moreover, NH3 is the most important basic component in the atmosphere, which has a large influence on precipitation scavenging and the formation of aerosols. The NH3 emissions in Europe are of the same order of magnitude as the NOx-emissions. The main sources of NH3 on a European scale are manure and fertilizers. In most cases in the past annual NH3 emissions were calculated by using emission factors (kg NH3 animal-1 year-1 or kg NH3 tonne fertilizer-1) which were more or less the same for whole Europe. NH3 emission are highly variable in space and time. Recently, new research on NH3 emission processes has been performed.. This makes it possible to develop NH3 emission models that are based on a description of physical, meteorological and chemical processes.

Temporal (diurnal , seasonal, interannual) and spatial variations in the NH3 emissions are not very well known. It is important to know these variations for the correct modelling of atmospheric processes. NH3 emissions from the application of manure and fertilizers depend to a large part on meteorological parameters like temperature, friction velocity and Monin-Obukhov length that play an important role in reactions and dry deposition in atmospheric transport models. Therefore, models will be developed describing the emission of this part including their spatial and temporal variability, that can directly be coupled to atmospheric transport models and use the values of the meteorological parameters provided in these models.

It is the aim of this task to improve existing methods and to develop new the methods for estimating the NH3 emission and its temporal and spatial variations and possible emission reductions.

This will be done by:

  1. developing emission models that are based on a description of physical, meteorological and chemical processes. This will make it possible to study the temporal variation of the emissions (diurnal, seasonal, interannual variations). These models need as input meteorological measurements (temperature, turbulence), information on agricultural practise (amount of manure/fertilizer applied, time of application) or on the compensation point (for NH3 emission from plants and (sea)water). These models will be based on data published in the literature and additional field experiments.
  2. developing procedures to obtain the best possible estimate of emissions on a small scale (5 km or less). Atmospheric transport models often use grid elements of 50x50 km2 or larger. A typical situation is that a minor part of a grid element consists of natural areas , where usually no NH3 emissions occur (forest, moorland etc.), whereas the main part consists of agricultural areas with a high NH3 emission density. Model calculations using NH3 emissions on a coarse grid may lead to large errors in the modelled dry deposition of NH3 to the natural areas with a wrong estimate of the exceedence of the critical load as a result.. Spatially detailed emission inventories will give a possibility to study the errors made with the atmospheric transport models that use a coarse grid. Moreover, these detailed emission inventories will make it possible to redistribute the dry deposition of NH3 within a coarse grid element, so that more realistic dry depositions to nature areas are obtained.
  3. evaluating the technical possibilities of emission reduction technologies and their associated costs.
  4. combining a. and c. to study the possibilities for emission reductions.


4.3.8 Emissions of Particulate Matter


The environmental issues that will be addressed in EUROTRAC 2 include the analysis of aerosols and acidifying substances. The chemistry of these substances involves particulate matter stemming from anthropogenic and biogenic sources. However, data on emissions of particulates are rare and usually incomplete (i.e. not all emission sources are regarded). Furthermore, to be useful for the analysis of transport and chemical transformation of pollutants, emissions of particulate matter have to be known subdivided into classes of species and of particle sizes.

Major aims of this task are to provide

  • ! a systematic collection of available knowledge on the sources of particulates and the emissions released from these sources, including the identification of priorities of further research work required for an appropriate improvement of present knowledge.
  • ! improved and completed emission factors for particles, including a detailed characterization of the emitted particles (chemical composition, particle size distribution a.s.o.). To achieve this aim additional measurements of emissions are planned.
  • ! first estimates of anthropogenic and - as far as possible - biogenic emissions of particles for European regions, including a breakdown of the temporal and spatial variability of the emissions as well as a characterization of the chemical composition, the particle size distribution and other relevant parameters needed to describe the emitted particles. However, as knowledge regarding emissions of particulates contains still many gaps, the resulting inventory will be incomplete.
  • ! detailed emission inventories for 'secondary' sources of aerosols in Europe and European regions. emissions of NH3 (see chapter 4.3.7) and HCl have to be regarded in this context additionally to ongoing activities concerning inventories of SO2, NOx and VOC.
  • ! a comparison of ambient measurements of particle and calculated emission data for deriving conclusions with regard to the completeness and realistic description of the source collectives that are considered in the emission estimates.


4.3.9 Emission Scenarios


A) Trends in past decades

In order to explain the development of measurements of ambient concentrations of pollutants in the past, the development of emissions in the past decades in Europe should be known. Although some emission estimates for different years in the past exist, these are not comparable as they are usually generated with different methods and parameters. So, using the currently available advanced methodology and available statistical data, emission inventories for past years are generated.


B) Future trends

Once models for transport and chemical transformation have been verified and used for past and present episodes, it is obvious that the next step will be to look for future trends in air pollution. In Western Europe, significant reductions of emissions will take place due to new laws and regulations concerning air pollution control. In Eastern Europe, increases or decreases of emissions are possible due to changes in the economy, traffic growth and due to a growing concern about environmental pollution.

The calculation of future air pollution will help to identify areas, where air pollution will still be or become a problem in the future. This will help to identify future problems in an early stage and will indicate, where further efforts are necessary for protecting the environment.

The preparation of scenarios of future emissions includes the estimation of the future activities. So, models have to be developed, that calculate the course of the activity values by using basic variables such as the number of inhabitants, the number of cars per 1000 inhabitants, economic growth of the different industries and the tertiary sector, growth of dwelling space per inhabitant, changes in the specific energy demand, substitution of energy carriers, changes in industrial processes etc. It may be sensible to couple economic models to the models for the emission calculation.

Future emission factors are influenced by technical innovations and by the different emission reduction measures, that are realized due to the new laws and regulations in air pollution control.

Of course, existing models and scenarios will be used. This includes the use of the FORMOVE model for transport emissions, the CASPER model and of the economy and energy scenarios of the Commission of the European Union.

It is intended to prepare a reference scenario for 2005. This reference scenario estimates the emissions that occur, if current trends in technologies, behaviour, economic and population growth and environmental policies continue until 2005. Special emission reduction measures, that exceed the measures now planned or in operation are not considered. The scenario will be prepared by a working group with members from the different participating countries.


C) Emission reduction scenarios

The determination of the future concentration of trace constituents in the troposphere will lead to an identification of areas and species, where a further reduction of pollution is desirable. As economic resources are limited, it is one of the aims of a rational environmental policy, to achieve the desired air pollution control with a minimum of expenditures. This aim can be supported by identifying strategies for efficient air pollution control.

These strategies are determined by examining the possible emission reduction measures for each source or source category. Costs and effect (emission reduction) of the measures are identified. Then the measures for the different source categories are combined to different emission reduction strategies. Optimal strategies combine all measures with marginal abatement costs per kg of pollutant below a limit. Different settings of the limit lead to different emission reduction.

The different emission reduction strategies are then used as an input for the atmospheric models to evaluate the impacts of the strategies on the concentration and deposition of air pollutants in Europe. From the results, recommendations for rational and efficient environmental policies can be deduced.


4.3.10 Cooperation

A) Cooperation with other institutions

Cooperation with other institutions that are involved in the generation of emission data like e.g. IGAC, EEA/CORINAIR and ECE/EMEP is planned. Especially a close cooperation with the European Environmental Agency (EEA) and EEA’s Topic Center for Emissions is essential. This might include the exchange of data, methods and computer models. Furthermore results from other EC activities (COST319, MEET, AUTO/OIL and others) are used.


B) Collaboration with other subprojects

European emission data with high temporal and spatial resolution are provided to REMAPE and to TROMEDA. With SATURN and LOOP urban emission inventories are developed. BIATEX results will be used to improve the calculation of biogenic emissions. The collaboration with further subprojects is planned and will be concretized during the planning phase of additional subprojects.


5 Organisation of the subproject

Coordinator of the subproject is Dr.-Ing. habil. Rainer Friedrich. Dr. Friedrich, who studied physics, is deputy director of the Institute for Energy Economics and the Rational Use of Energy (IER) of the University of Stuttgart and heads the department of Technology Assessment and Environment of IER. His main working areas are the investigation of impacts of human activities on the environment, the generation of emission inventories , the identification of efficient emission reduction strategies, the assessment of technologies with regard to environmental impact and risks to human health, the identification of sustainability indicators and the estimation of external costs of energy and transport systems. As a university lecturer he teaches in the fields of energy and environment, technology assessment and environmental economics. He has a large experience in coordinating international research projects. He was coordinator of the subproject GENEMIS of EUROTRAC 1 and is currently coordinator in two EC-DGXII research projects , member of the Scientific Steering Committee of EUROTRAC-2 and one of the subproject coordinators of the German Tropospheric Research Programme.

To facilitate the coordination of the subproject, it is divided into tasks. These tasks correspond to the tasks described in chapters 4.3.2 to 4.3.9 of this proposal. Each task is coordinated by a task leader, who is also member of the steering committee. The following task leaders are responsible for the different tasks:

  1. 1 Measurements of emission factors (4.3.2): Franz Slemr
  2. 2 Evaluation of emission data (4.3.4): Johannes Staehelin, Franz Slemr
  3. 3 Emission Update (4.3.4): Winfried Winiwarter
  4. 4 Methods and models /European scale (4.3.5): Rainer Friedrich
  5. 5 Urban emission inventories (4.3.6): Peter Sturm
  6. 6 NH3 - emissions (4.3.7): Willem Asman
  7. 7 Emissions of particulate matter (4.3.8): Jan Berdowski
  8. 8 Future emissions (4.3.9): Tinus Pulles.


From the list of principal investigators in chapter 1, the allocation of PI’s to the tasks can be seen. However, some PI’s contribute to several tasks, in that case only the most important allocation is given. Activities in tasks 7 and 8 will start later; task 8 has to build on inventories for current episodes developed in task 4. So, further PI contributions for these tasks, but also for the other tasks, are planned.

Within task 2: ‘evaluation of emission data’, two subtasks are treated separately, as they describe larger self contained activities. These are the ‘Augsburg experiment’, coordinated by Franz Slemr and the ‘Zürich experiment’coordinated by Johannes Staehelin. Subtasks usually have a shorter duration than the task they belong to.

Within task 5:‘urban emission inventories’, a close cooperation with the subprojects SATURN and LOOP is necessary. This is ensured e.g. by allowing double allocations of individual contributions to two subprojects. In addition, the task leader is member of as well the SATURN as the GENEMIS steering committee.

Information exchange within the subproject is achieved by regular meetings and workshops. A GENEMIS workshop will be held once a year. Task and subtask contributors will meet more often, meetings will be planned and coordinated by the task leaders.


Time schedule:

The GENEMIS tasks will be actively processed throughout the whole duration of EUROTRAC_ 2, i.e. until 2002. However, the subtasks and the projects, that are the basis for the work on the tasks, usually have a shorter duration of 3 to 4 years.

Task 1: Results of the measurements described in task 1 will be available until middle of July 1999.

Task 2: The first ‘Augsburg’ experiment will be carried out in February/March 1998, the ‘Zürich’ experiment in spring 1998. Further experiments are planned for 2000. In Hungary, experiments will be performed 1998, 1999 and 2000.

Task 3: A methodological framework will be available until 2000. First updates (European scale) for 1998 will be ready during 1998.

Task 4: First results, i.e. emission data for episodes are generated and supplied in agreement with the users of the data in the other subprojects. Preliminary European data for 1994 will be available before the end of 1997.

Task 5 : The first deliverable will be the guidelines for the set up of urban emission inventories. These guidelines will be available by end of 1997. Collection and comparison of different urban emission inventories will be done in the years 1998 to 2000. The basic investigations concerning different methodological approaches will be ongoing until 2002.

Task 7 will start in 1998. Task 8 will begin in a later phase of the project (ca.1998/1999).

For carrying out the tasks described above, the following personnel is required:

personnel - man-years
yearly costs/kECU



6 Proposals for contributions to GENEMIS

In the following the proposals for contributions to GENEMIS are listed in the same order as in the list of principal investigators (chapter 1).