John van Aardenne and Valerie Gros
Max Planck Institute for Chemistry, Mainz, Germany.
(contact: aardenne@mpch-mainz.mpg.de)
1. Description of the current GEIA emission databases for NMVOC
The group Non-Methane Volatile Organic Compounds consists of a large variety of different organic compounds which play a significant role in atmospheric chemistry due to their chemical reactivity. Emissions of NMVOC from anthropogenic origin are caused by combustion processes, production of fossil fuels, solvent use and industrial processes [1].
Due to the variety of different compounds it is a rather difficult task to compile a global inventory of NMVOC including all species and the variety of emission source categories. The emission calculation is in general performed in two steps. First, an emission factor approach is used to calculate total NMVOC based on application of (inter)national energy and production statistics and by taking into account abatement technologies in the selection of representative emissions factors. Second, the different species of NMVOC are categorized into NMVOC groups and the emissions per group are derived as fraction of the total NMVOC emissions based on source category specific NMVOC profiles.
The current GEIA set of anthropogenic NMVOC emissions contains data from the EDGAR 2.0 emission inventory [2],[3]. This emissions data is representative for the year 1990. According to the EDGAR 2.0 calculations the overall 1990 emission of NMVOC is 177.5 Tg. Major sources are: (i) biomass burning (~20% of global total), (ii) road transport (~19%), (iii) use of biofuels (~17%), (iv) solvent use (~13%) and (v) oil production (~12%).
In the dataset the following 25 NMVOC aggregation groups have been identified and based on NMVOC profiles emission by group are included in the dataset [3]
|
a01 alkanols
(alcohols) |
a15
xylenes |
|
a02 ethane |
a16
trimethylbenzene |
|
a03 propane |
a17 other
aromatics |
|
a04 butanes |
a18 esters |
|
a05 pentanes |
a19 alkoxy alkanes
(ethers) |
|
a06 hexanes and
higher alkanes |
a20 chlorinated
HCs |
|
a07 ethene |
a21
methanal |
|
a08 propene |
a22 other
alkanals |
|
a09 ethyne |
a23 alkanones
(ketones) |
|
a12 other
alk(adi)enes and alkynes |
a24 (alkanoic)
acids |
|
a13 benzene |
a25 other NMVOC
(HCFCs, nitriles, etc.*) |
|
a14 toluene |
|
*
Excluding isoprenes and monoterpenes (used for natural sources
only).
2. Shortcomings and uncertainties
Due to the reactive role of NMVOC in global atmospheric chemistry, models dealing with NMVOC chemistry will require a fine and explicit speciation of NMVOC emissions. A major shortcoming in any global set of NMVOC emissions is the fact that this detail is not available; only emissions by aggregated classes are provided. Although the spatial resolution of the current dataset (1°x1°) should suffice for most atmospheric models, the lack of temporal resolution in the dataset is another important shortcoming. The current GEIA dataset provides emission estimates representative for the year 1990. Too allow for modeling of present-day atmospheric conditions these emission estimates should be updated for more recent year emissions.
As is the case with any emission inventory based on emission factor calculations, uncertainties of the emission figures are related to:
- quality of (inter)national energy and production statistics
- information on biomass burning activity
- country specific emission factors
Another important uncertainty is the application of source category specific NMVOC emission profiles. In general NMVOC profiles are available for Northern America and European situation [4], [5] while for other regions no information is available. It is obvious that the emission estimates for other regions could be rather inaccurate.
An indication of uncertainty in activity levels, emission factors and the EDGAR v2.0 global emission inventory is given in [3]. It goes beyond a short review to go into much detail for individual NMVOC species but in the literature several studies discuss possible inaccuracies in the current GEIA inventory (see [6] for an example).
3. Alternative emission distributions
There is an update version of EDGAR available which is not yet included in the GEIA data. EDGAR 3.2 presents updated emissions for the year 1990 and new estimates for the year 1995 [7]. This version of EDGAR presents at this moment only total NMVOC emissions.
Several regional emission inventories are available. For example;
- United States of America: National Emission Inventory (NEI; [8])
- Europe: CORINAIR 1994 inventory [9]
- Asia: TRACE-P emission inventory [10]
Although they provide information on a regional scale and sometimes do not include emissions by grid, it could provide an alternative source of information.
Another source of information on NMVOC emissions are the National Communications under the UNFCCC framework [11].
4. Emission trends
Historical time series of global and gridded (1°x1°) total NMVOC have been calculated for the period 1890-1990. A description of the set and the actual data can be found at [12]
Time series of possible future trends in NMVOC emissions have been calculated by the IPCC in [13].
5. The seasonality of the emissions
Seasonality of emissions is not included in the dataset.
6. Natural emissions
See NMVOC (N) part.
7. New GEIA projects
As far as future emission of NMVOC are concerned, it is expected that anthropogenic NMVOC emissions could drastically increase in the next years/decades due to emissions in developing countries in which an increased economic activity is accompanied with lower emission control as in developed countries. The emissions in these regions are not well-known and calculations are often based on non-representative emission factors for developed regions. In a new GEIA inventory activity these developments should explicitly be addressed.
Given the uncertainties accompanying the GEIA inventory it is important that information is presented on uncertainties in the dataset and that suggestions for improvement of the dataset are provided. This is not only a task of the inventory compiling community. Information from measurement campaigns or results from forward or inverse air quality modeling will provide insight in shortcomings of the GEIA data sets. It is therefore important that these insight are communicated to GEIA.
References:
[1] Friedrich, R and A. Obermeier (1999). Anthropogenic emissions of Volatile Organic Compounds. In: Hewitt., N. (Ed.) Reactive Hydrocarbons in the Atmosphere.
[2]
http://arch.rivm.nl/env/int/coredata/geia/index.html/
[3] Olivier, J., A. F. Bouwman, C. W. M. Van der Maas, J. J. M. Berdowski, C. Veldt, J. P. J. Bloos, A. J. H. Visschedijk, P. Y. J. Zandveld, and J. L. Haverlag, 1996. Description of EDGAR Version 2.0: A set of emission inventories of greenhouse gases and ozone depleting substances for all anthropogenic and most natural sources on a per country basis and on 1° x 1° grid. RIVM Rep. 771060002, Bilthoven: National Institute of Public Health and the Environment.
[4]
http://www.epa.gov/ttn/chief/software/speciate/index.html
[5]
http://reports.eea.eu.int/technical_report_2001_3/en/
[6] Jacob, D.J., B.D. Field, E. Jin, I. Bey, Q. Li, J.A. Logan, and R.M. Yantosca. Atmospheric budget of acetone, J. Geophys. Res., 10.1029/2001JD000694, 2002.
[7]
http://arch.rivm.nl/env/int/coredata/edgar/index.html
[8]
http://www.epa.gov/ttn/chief/trends/
(last modified 02/28/05)