Emission of trace gases and aerosols from biomass burning

 M. O. Andreae
Biogeochemistry Department Max Planck Institute for Chemistry,
P.O. Box 3060, D-55020 Mainz, Germany

            A large body of information on emissions from the various types of biomass burning has been accumulated over the past decade, to a large extent as a result of IGBP/IGAC research activities. Yet, this information has not been readily accessible to the atmospheric chemistry community because it was scattered over a large number of publications, and reported in numerous different units and reference systems. A discussion of various methodological problems in determining emission factors from field measurements will be presented. Presently available data have been critically evaluated and integrated into a consistent format. On the basis of this analysis, we present a set of emission factors for a large variety of species emitted from biomass fires. We have derived global estimates of pyrogenic emissions for important species emitted by the various types of biomass burning and will compare these estimates with results from inverse modeling studies.

Research on emissions of species into the atmosphere under the Environment Programme of the EC

 G. Angeletti
European Commission, Research Directorate General
Rue de la Loi 200, 1049 Brussels, Belgium

The European Commission (EC) supports atmospheric research within the key Action on Global Change, Climate and Biodiversity of the 5th Framework Programme (FP), 1998-2002.  The content of the work programme in based on a strategic document which was prepared through the EC Science Panels on Atmospheric Research and Stratospheric Ozone and was published in November 1997 (AIRES, doc. 17645).

While atmospheric issues having a regional, global dimension, are integrated with other issues of the key action on global change as climate, ecosystems, carbon cycle, global observing systems, local, urban scale problems related to air quality are dealt within the key action ´City of tomorrow and cultural heritageª.

Two calls for research proposals have been launched in relation to the atmospheric composition issue: the first in Summer 1999 and the most recent with deadline 15 February 2001.

Here, a description is given on the projects supported under the 1st call and a brief indication of the projects likely to be supported under the 2nd one.  Concerning the 1st phase, a clustering has been done, bringing to the formation of 4 groups of projects dealing with aerosols, ozone and its precursors, sources of pollutants and oxidation processes.

A better, more comprehensive clustering can be done once the projects of the 2nd phase can start, integrating so the previous grouping.

Currently the EC is developing a strategy for a frontier-free research policy in Europe, the so-called European Research Area (ERA).  Improved cooperation among researchers in the Members States, access to large research infrastructures, the stimulation of women and young peopleís career in science, as well the networking of centres of excellence are expected to be the main principles for the future of European research.

One of the specific programmes related to the next Framework Programme, FP6 within ERA, as proposed by the EC, deals with Global Changeª; it is hoped that atmospheric research will find there its good place, in particular within the actions envisaged on global climate change.

In the meantime, a new strategic document, follow-up of AIRES, has been prepared recently, defining new research priorities as a possible contribution to ERA and FP6.

Chemical Assessment of Oceanic and Terrestrial Sulfur in the Marine Boundary Layer
over the Subarctic North Pacific during Summer

 K. Aranami, S. Watanabe, S. Tsunogai, A. Ohki, K. Miura and H. Kojima
Laboratory of Marine and Atmospheric Geochemistry, Graduate School of Environmental
 Earth Science Hokkaido University, Sapporo 060-0810, Japan

As integrative tools for the determination of oceanic and terrestrial sulfur emissions, we used dimethylsulfide (DMS) in surface seawater and the air, methanesulfonic acid (MSA) and non-sea-salt sulfate (nss-SO₄^2-) in aerosol, and radon-222 (Rn-222).  We considered DMS and MSA as oceanic sulfur, nss-SO₄^2- as total (oceanic + terrestrial) sulfur, radon-222 (Rn-222) as a tracer of terrestrial air masses.  In the marine boundary layer (MBL) over the subarctic North Pacific during summer (Jul. - Sep. 1997), the mean concentrations of DMS and MSA were higher in the eastern region than the western region and Bering Sea, although these regions did not significantly differ in the mean DMS concentration in surface seawater.  The DMS and MSA difference between the eastern and western region would be produced by sea-to-air DMS flux owing to wind speed, while the difference between the eastern region and Bering Sea could be caused by other factors such as height of MBL, DMS oxidation and vertical convection rate.  The concentrations of nss-SO₄^2- and Rn-222 were higher near the continent, which suggests that nss-SO₄^2- concentrations were regionally influenced by anthropogenic sulfur input.

Physical and Chemical Properties of Natural Biogenic Aerosols in Amazonia

 P. Artaxo, J.V. Martins, A. S. Procupio, T. M. Pauliquevis, M. O. Andreae, P. Guyon
Instituto de Fisica, Universidade de S„o Paulo, Rua do Mat„o, Travessa R, 187
Sao Paulo, S. P., Brazil
Max Planck Institute for Chemistry, Biogeochemistry Department, P. O. Box 3060
 D-55020 Mainz, Germany.

Tropical rain forests are rapidly reducing their area, but still cover a large fraction of the tropical continental region, and influence the global atmosphere in several ways.  In particular, the natural vegetation is a major aerosol source that is poorly characterized from the chemical and physical point of view. The natural Biogenic aerosol is characterized by large particles, with predominance of carbonaceous material with K, S, Ca, P, Zn and other trace elements. It has optical absorption that is difficult to explain in terms of the know composition and size distribution. The mass concentration for PM₁₀ is about 15mg/m, with most (75%) of the mass in the coarse fraction. These natural biogenic aerosol particles carry significant amounts of phosphorus, and it is important in terms of the biogeochemistry of this key tropical forest nutrient. Measurements at a forest tower show large emissions of biogenic aerosols during nighttime at ground level of the tower, and a strong gradient along the forest canopy. Aerosol sun photometry shows that during the wet season the aerosol optical depth averages 0.10 at 500 nm, a very low value, indicative of clean continental conditions.

GHG Emissions From Biomass Burning And Land Use Change
And Forestry Sector In India ñ Estimate From Satellite
Data And Ground Based Measurements

 K.V.S.Badarinath and V.Krishna Prasad, Prabhat K. Gupta*, B.C. Arya, C.Sharma*, S.L.Jain* and A.P.Mitra*
National Remote Sensing Agency, (Dept. of Space-Govt. of India), Hyderabadñ500037, India
*National Physical Laboratory, Dr.K.S. Krishnan Road, New Delhi - 110 012, India.

Biomass burning in tropics is widely recognized as an important source of trace gas emissions. In India, though emissions of trace gases from urban pollution sources have been studied, data on biomass burning has not been reported. In the present study, we report the emission ratios and emission factors obtained for quantification of trace gases from tropical dry deciduous and mixed deciduous forests obtained from ground based measurements along with aerosol and UV-B measurements. Biomass quantities at different sites before and after burning have been quantified in the forest patches cleared for shifting cultivation purposes. Measurements were made for the gases of CO, CO₂, NO_x,CH₄, N₂O, etc. Satellite data pertaining to study area at different resolutions has been studied using IRS-1C LISS-III, IRS-P3 WiFS, IRS-P4 OCM data and NOAA-AVHRR data, for identifying location, intensity and regional extent of fires. Combustion completeness suggested a range of 16.1%-30.04%, indicating low amount of biomass burned during shifting cultivation purposes. The dCO/dCO₂ emission ratios are observed to be relatively high when compared to the other ecosystems such as Boreal, Chapparal and Savanna fires. In the study, comparison of emission ratios and combustion factors for different types of forests along with the implications of biomass burning from land use change and forestry sector in India are discussed. 

Estimate of Global NO_x-Emissions from GOME-NO₂ Observations

 S. Beirle, M. Wenig, T. Wagner, J. Hollwedel, W. Wilms-Grabe, S. Kraus,
B. Johne, U. Platt
Institute for Environmental Physics, University of Heidelberg, INF 229,
D-69120, Germany

Nitric oxides play a very important role among the anthropogenic trace gases. They affect human health and have an impact on ozone chemistry and climatic change. Here we describe a new method for the quantification of the global NO_x budget from image sequences of the Global Ozone Monitoring Experiment (GOME) spectrometer on board the ERS 2 satellite. In contrast to measurements using ground-based or balloon- or aircraft-borne sensors, this instrument provides, for the first time, the possibility of observing global maps of NO₂ column densities.

Independent of traditional methods, a global source strength of 43 TgN/yr is estimated. The accuracy of this method is comparable to that of established statistical approaches.

Satellite-based Global Distribution of Sea Surface
Dimethylsulfide (DMS) concentrations

S. Belviso and C. Moulin
Laboratoire des Sciences du Climat et de l'Environnement, UMR CEA-CNRS, CEN/Saclay,
Orme des Merisiers, B‚t 709, 91191 Gif-sur-Yvette, France

Dimethylsulfide (DMS) is a sulfur-compound produced naturally in the sea. Once in the atmosphere, it is photooxidized to form non-sea-salt sulfate aerosols which affect the radiative budget indirectly as cloud condensation nuclei (CCN). It has been postulated that the planet's climate may be modulated by variations in DMS production resulting from changes of the sea surface temperature (SST). It is shown that surface seawater DMS concentrations can be diagnosed at a near global scale using (1) SEAWIFS observations of Chl a, (2) a SST dependent index of the marine community structure and (3), observational relationships based on cruise data (DMS being a non linear function of Chl a and the index of marine community structure). This procedure is used to characterize the spatio-temporal variations in marine DMS concentrations at the global scale for the period 1998-2000.  The results are compared to the Kettle et al. global database of about 16,000 point measurements of DMS in surface waters. Our satellite-based global distribution of DMS concentrations represents many features of the observed concentrations of DMS. This is a contribution to the SOLAS project.

Compilation of Inventories of Industrial Emissions

 Carmen M. Benkovitz
Atmospheric Sciences Division, Environmental Sciences Department
Brookhaven National Laboratory, Upton, NY 11973, USA

The mathematical modeling of the transport and transformation of trace species in the atmosphere is one of the scientific tools currently used to assess atmospheric chemistry, air quality, and climatic conditions.  From the scientific but also from the management perspectives accurate inventories of emissions of the trace species at the appropriate spatial, temporal, and species resolution are required.  There are two general methodologies used to estimate emissions: bottom-up and top-down (also known as inverse modeling).  Bottom-up methodologies used to estimate industrial emissions are based on activity data, emission factors (amount of emissions per unit activity), and for some inventories additional parameters (such as sulfur content of fuels).  To compile regional and global inventories researchers can either bring together estimates made at the national or sub-national level by national experts or directly estimate emissions based on activity rates from reports compiled by multi-national organizations such as the United Nations and the International Energy agency and on emission factors and other information available in the literature.  In all cases the data used must be checked for completeness, transparency, consistency, comparability, and accuracy.  These emissions estimates must be given finer spatial (usually gridded), temporal, and for some inventories species resolution. The locations of major stationary sources (power plants, industrial complexes) are usually known, so the emissions can be directly assigned to the appropriate grid cell.   For emissions from other activities, such as transportation, spatial resolution is obtained via the use of surrogate information, such as population, land use, traffic counts, etc. which already exists in or can be converted directly to gridded form.  To obtain finer temporal resolution (seasonal, daily, weekday/weekend, etc.) auxiliary information such as plant schedules, traffic counts, etc. is used.  Speciation factors have been and are being developed to speciate inventories of NO_x (NO, NO₂), particulate matter (PM2.5, PM10; by species), and hydrocarbons (individual species or groups of species).  Top-down (inverse modeling) methodologies directly invert air quality measurements in terms of poorly known but critical parameters to constrain the emissions needed to explain these measurements; values of these parameters are usually computed using atmospheric transport models.  Several statistical tools are being used to address the inversion problem, including empirical Bayes framework, synthesis inversion techniques based on Greens function, single value decomposition, and tangent linear and adjoint models.  Currently there are several strong limitations on the use of inverse modeling.  It is harder to address emissions by sector using this methodology; for the long-lived trace gases the signature in the measurements induced by various sources is small, so high precision measurements are needed; emissions are highly variable so their representation needs many degrees of freedom making the inverse problem highly underdetermined; additional a priori information based on model results, secondary observations or simplifications of the requirements for the inverse solution are needed, which introduces additional errors in the calculations; the high diffusivity of atmospheric transport and mixing quickly erases the small-scale emissions structure so that measurements at remote monitoring stations only capture the large-scale variability of the emissions.  However,  the continued development of inverse modeling will be facilitated by the development of denser monitoring networks and the massive amounts of data that are becoming available from satellite observations.

Emissions for the TROTREP project 

J.J.M. Berdowski¹, J.P.J. Bloos² and A.J.H. Visschedijk^1
1 TNO-MEP, Apeldoorn, the Netherlands,
 ² IMAU, Utrecht University, Utrecht, the Netherlands

TROTREP emissions work package:

1.            Anthropogenic emissions of CO, NMVOC, NO_x, (CH₄, SO_x and NH₃), for 1970 ñ 1995

2.            Temporal variation (monthly, weekly, daily)

3.            NMVOC-profiles

4.   Spatial distribution (point sources)

Ad 1. Emissions of several components show a decrease after environmental policy changes in the 70ís, 80ís and 90ís. The start of the emission decrease varies per country and per component. Another factor contributing to changes in emissions is the collapse of the centrally planned (Eastern European) economies in the 90ís.

Ad 2. Sensitivity analyses were performed to see which of the three possible temporal variations was the most critical. Three model sets (flat, step and sinusoidal profile) were used and later combinations were made to define the importance of monthly, weekly and daily variations.

Ad 3. Due to environmental measures the composition of e.g. solvents is changing. One of the most prominent changes is the phase-out of chlorinated hydrocarbons.

Ad 4. The use of point sources improves the spatial distribution in three dimensions. The use of point sources, of course, increases the reliability of spatial distribution in the latitudinal and longitudinal plane, but they can also be used for differentiation in the (effective stack) height at which emissions are released.

Quantification of carbon dioxide, methane, nitrous oxide, and
chloroform emissions over Ireland from atmospheric observations at Mace Head

 Sébastien BIRAUD
Laboratoire des Sciences du Climat et de l'Environnement (LSCE), C.E. de l'Orme des Merisiers,
91191 Gif-sur-Yvette Cedex, France

Flux estimates of CO₂, CH₄, N₂O, and CHCl₃ over Ireland are inferred from continuous atmospheric records of these species. We use radon-222 (²²²Rn) as a reference compound to estimate unknown sources of other species. The correlation between each species and ²²²Rn is calculated for a suite of diurnal events that have been selected in the Mace Head record over the period 1995-1997 to represent air masses exposed to sources over Ireland. We established data selection based on criteria using ²²²Rn and ²¹²Pb thresholds. We estimated flux densities of 12-15 10³ kg CH₄ km^-2 yr^-1, 680-830 kg N₂O km^-2 yr^-1, and 20-30 kg CHCl₃ km^-2 yr^-1 for CH₄, N₂O, and CHCl₃, respectively. We also inferred flux densities of 250-310 10³ kg C km^-2 yr^-1 for CO₂ during wintertime, of 760-950 10³ kg C km^-2 yr^-1 for CO₂ during summer nighttime. Our CH₄ inferred flux compare well with the CORINAIR90 and CORNAIR94 inventories for Ireland. The N₂O emission flux we inferred is close to the inventory value by CORINAIR90, but twice the inventory value by CORINAIR94 and EDGAR 2.0. This discrepancy may have been caused by the use of the revised 1996 IPCC guidelines for national greenhouse gas inventories in 1994, which include a new methodology for N₂O emissions from agriculture. We carried out a first estimation of CHCl₃ emission fluxes over Ireland, where emissions from natural sources were about 4 times larger than those measured close to the Mace Head station over peatlands. Our CHCl₃ emission fluxes estimate is consistent with the interpretation of the same data by  , who obtained, using a Lagrangian atmospheric transport model, CHCl₃ fluxes of 24 ± 7 kg CHCl₃ km^-2 yr^-1. Our estimates of CO₂ emission fluxes during summer nighttime and wintertime are close to those estimated from inventories and to one biogeochemical model of heterotrophic respiration.

Using isotope information to better know
the emissions of CO

Carl Brenninkmeijer, Valerie Gros, Thomas Roeckmann and Patrick Joeckel
Max Planck Institute for Chemistry, Mainz, Germany

By combining a sufficient number of atmospheric observations of CO with model calculations, it is in principle possible to estimate the distribution, strength and variations for the various sources of CO. Various research groups tackle the problem of better understanding the global atmospheric CO cycle, yet a picture of detail to that degree that does justice to the atmospheric importance of the gas is far away indeed. Now, besides measuring the mixing ration of CO, one can measure 4 isotopic properties that are rather independent of each other. These isotopic characteristics pertain to the ¹⁴C, ¹³C, ¹⁷O and ¹⁸O content. Different sources have different isotope ratios, for instance CO from combustion sources is most enriched in ¹⁸O. Also the removal of CO in the air through mainly the oxidation by OH inflicts isotopic changes. Over the years good progress has been made in applying the isotope signatures. The CO budget estimates for the SH have been scrutinized using inverse modeling of the ¹³C content, and 3D inverse modeling has been used subsequently. It seems that biomass burning constitutes a sources of CO that is difficult to quantify.

Isotope analysis have always been elaborate. This however is changing markedly because of analytical advances, and consequently, it is probable that much more use will be made of isotope analysis. Accordingly models should be amended to incorporate the isotopic information. We show here the extremely interesting, clear and large isotopic changes that occur in atmospheric CO, and try to use these to draw conclusions about emissions. Examples will be given from the INDOEX campaign, from observation sites in the NH showing biomass burning influences, from the CARIBIC intercontinental measurement flights, and from polluted surface air in Europe.  

Application of an improved technique to evaluate controls on N₂O emissions
from grassland soils.

L.M. Cardenas, J.M.B. Hawkins, D. Chadwick and D. Scholefield
Institute of grassland and environmental research
Notrh Wyke, Okehampton, Devon, UK

It is well known that N₂O plays a key role in ozone destruction in the stratosphere. It is also a greenhouse gas 200 times more effective than CO₂ on a per molecular basis. The total N₂O emission from the UK in 1990 was 136 kt N. Soils are the main source of N₂O with a total contribution from agricultural land in the UK in 1990 of 63.9 kt N of which 30 to 50% is derived from grasslands. The source of N₂O from grassland is mainly the process of denitrification, which is typically associated with wet, poorly drained soils. Published N₂O data are mainly based on field measurements using static chambers, N isotopes and micrometeorological methods. Laboratory studies use microbial inhibitors such as acetylene, but this could have other effects, as it may be a source of carbon too.

This work describes an improved laboratory technique that allows direct and simultaneous measurements of N₂O and N₂ produced during denitrification. Intact soil cores are flushed with He/O₂ and the impact on emissions of N₂O, N₂ and other gases such as CO₂ and NH₃ after fertiliser application is measured. Continuous automated measurements of N₂O and N₂ and other gases can be carried out at high sensitivity and sampling frequency. Controls can be evaluated with the aim of constructing a model that will provide management guidelines for reduced emissions of N₂O from grazed, poorly drained pasture.

Seasonal Trends of n-Alkanes, PAH and n-Alkanoic Acids
in the Atmosphere of a Green Park in Rome

 A. Cecinato¹, C. Balducci¹, P. Di Filippo²
(1)   Istituto sullíInquinamento Atmosferico CNR, CP 10,
00016 - Monterotondo Scalo RM, Italy
(2)ISPESL, Via di Fontana Candida, 00040-Monteporzio Catone RM, Italy.

Concentrations of n-alkanes, PAH and n-alkanoic acids have been determined in the atmosphere of Villa Ada, the largest green park in Rome. Analytes have been investigated in both condensed and vapour phase, by sampling size-segregated aerosols (i.e., PM_2.5, PM₁₀ fractions and total suspended particles) as well as semi-volatile species (by means of polyurethane foam cartridges).

Samplings have been performed in all seasons during 2000 and the winter campaign has been repeated in 2001, in order to investigate qualitative and quantitative variations of contents in the atmosphere for all classes of compounds investigated, and also to assess the relative impact of emission sources onto the ambient quality. By looking to n-alkanes and acids, natural emission influenced over all the composition of large particles. By contrast, anthropic sources characterised the nature of aerosols affecting human health (i.e. fine particles: see PAH distribution among size-segregated fractions of aerosols). The common trend of anthropogenic pollutants along the year presented maximum concentrations in winter and minimum in summer. Instead, concentration maxima of naturally-emitted components were shifted in the mild season (spring and summer). Thanks to low levels reached in summer, average annual concentration of benzo(a)pyrene is below 1.0 ng m^-3, which is guideline for air quality established by Italian Ministry of the Environment (Decree November 22, 1994 No. 191) to represent urban aerosol carcinogenicity.

Spatial Analysis of Metals in Total Suspended Particulates 

Chandra Sekhar. M
Water & Environment Division, Regional Engineering College,
Warangal - 506 004, INDIA

Particulates are generally the most significant air pollutants that can be observed in urban areas. Several activities industrial, transportation and other associated urban activities are responsible for these emissions.  Assessment of the degree of harm or nuisance caused by inhalable metallic particulates either when they are still airborne or after removal and deposition on objects on ground introduces the problem of quantification. Spatial analysis of metallic particulates based upon the predominant land use activities in an urban ecosystem can be of great importance to impose land use controls. Such legislative controls in urban planning are helpful to reduce air pollution and its effects on the human population.

Tirumala - Tirupathi region, one of the most celebrated pilgrim centers of Hindu religion is located in the southern part of India. The region covers several hill ranges and the valley portion down the hill. As such, the wind profiles and movement are very different and hence, the spatial analysis of metallic particulates becomes much more significant. The climate in this region is dry with a short wet spell due to the south-west monsoon. Being an important pilgrim center it attracts 100,000 - 150,000 pilgrims every day from several parts of the country and world. Apart from the religious sanctity attached to the place, it is an important educational and industrial center. The growth of the place is influences by the floating population and hence, the observed land use activities are limited and most often mixed. In the present work an attempt is made to study the spatial analysis of selected pollutants in the study region. The results obtained in the study can form as a base to compare the ambient air quality.

The results of the study indicate that the urban activities are responsible for high concentrations of inhalable metallic particulates at some of the sampling stations. Sampling stations located at no-activity zone gives an indication of probable background levels, that were to be present if the contributions from urban activities are zero/negligible. High concentrations of lead and zinc (100 and 60 micograms/m³) are observed at the locations where transportation is predominant activity. The contributions of sodium, calcium and magnesium are attributed to urban activities such as building construction, land excavations, road construction, etc.,. The industrial emissions with respect to sodium. Calcium and magnesium are also considerable especially at stations near the industries manufacturing bricks, ceramics, tiles, etc.,. The results of the study explain only spatial variations and an exhaustive study covering temporal variations is essential for developing air quality models.

Estimates of Global Oceanic Dimethylsulfide Emission Fluxes
Using Model-Derived Wind Speeds

 E. G. Chapman, R. C. Easter, W. J. Shaw, X. Bian, and S. J. Ghan
Pacific Northwest National Laboratory, PO Box 999, Richland, WA 99352 USA

The flux of dimethylsulfide (DMS) from the worldís oceans is the largest known source of biogenically derived reduced sulfur compounds to the atmosphere.  Its impact on atmospheric chemistry and radiative transfer is an active area of scientific research, and DMS is routinely included in three dimensional global climate change and chemical transport models.  In such models DMS fluxes typically are based on global sea surface DMS concentrations and wind-speed-dependent parameterizations of the mass transfer coefficient.  Emission estimates may be generated off-line by pre-calculating monthly average fluxes using monthly average winds, or calculated on-line using short-term winds (e.g., 6-hourly winds from a re-analysis data set or 15-60 minute winds from a global climate model).  Most mass transfer coefficient parameterizations, and thus the computed fluxes, are non-linearly dependent on wind speed.  This paper reviews four parameterizations and discusses their suitability for use with monthly average vs. short-term winds.  We demonstrate that, because of the non-linear functional dependence, the use of short-term winds leads to significantly greater (>20%) estimates of global annual and monthly DMS fluxes compared to fluxes calculated using monthly average winds.

Tropospheric Measurements from Nadir-Viewing Infrared Instruments

 C. Clerbaux, J. Hadji-Lazaro, S.Turquety, D.Hauglustaine, C. Granier, G.MÈgie
Service d'AÈronomie, BP102, 4 Place Jussieu, 75252 Paris

Satellite data currently represent the majority of the observations available to improve our understanding of tropospheric chemistry. Nadir-viewing instruments dedicated to the sounding of the troposphere using the thermal reemission of Earth were launched on board of polar-orbiting satellites (IMG, MOPITT), and several other remote sensors are planned to be launched in the coming years. The analyze of the data provides global scale information on atmospheric contents of CO, O₃, and CH₄. In order to maximize the scientific return to be expected from these missions, efforts are being dedicated towards the development of fast forward radiative codes, efficient inversion methods for the retrieval of trace gases, and data assimilation techniques in chemical models. The paper will present measurements currently available and work undertaken in this framework.

Improving regional estimates of trace gas emissions by ²²²Rn tracing 

Franz Conen
Institute of Ecology and Resource Management, University of Edinburgh,
West Mains Road, Edinburgh EH9 3JG, UK.

Using the relatively uniformly emitted ²²²Rn as a tracer, regional estimates (10² - 10⁵ km²) of CO₂, CH₄, N₂O and other species have been made in Europe, Japan, and Canada. In this context, the emission (E) of the species of interest (i.e. CH₄) is calculated as:

E_CH4 = ( D C_CH4 / D C_222Rn) _* E_222Rn

where D C_CH4 and D C_222Rn are concentration differences in CH₄ and ²²²Rn in the atmosphere at different times during a synoptic event. This part of the equation can, for example, be obtained by analysing data from a Global Atmosphere Watch (GAW) station, if there is also a parallel monitoring of atmospheric ²²²Rn. The emission rate of ²²²Rn (E_222Rn ) is usually obtained by direct measurements with flux-chambers but on large land surfaces also by aircraft soundings of ²²²Rn concentrations across the entire depth of the boundary layer.

While there are numerous published spot measurements of E_222Rn, there are only very few published systematic studies on a larger spatial or temporal scale. In the northern hemisphere, E_222Rn seems to decrease with increasing latitude. On a temporal scale, emissions are sometimes found to be lower during the winter than during the summer months.

Estimates of regional scale trace gas fluxes derived from data collected at atmospheric monitoring stations could be improved, if a spatially and seasonaly more detailed characterisation of ²²²Rn emission was available. This issue could be addressed in the following ways:

1.) Collection, analysis and distribution of all published and unpublished data on ²²²Rn emission in one data base.

2.) Identification of priority areas for further ²²²Rn emission measurements (probably within the radius of a few 100 km around atmospheric measurement stations).

3.) Systematic measurements of ²²²Rn emissions over one full year in these regions.

Points 1 and 2 are hoped to be discussed at the workshop. Concerning point 3, we have devised a set-up based on commercially available components that allows sampling and off-line analysis of about 10-15 samples per day and have started monthly monitoring at two sites with soil types for which there is no published data (peaty gley, peat) in Scotland. A third site (loamy sand) is monitored with higher frequency. First results will be presented.

Dimethylsulfide (DMS) Oceanic Emissions around Antarctica
in LMDz Atmospheric General Circulation Model

 E. Cosme, C. Genthon and P. Martinerie
Laboratoire de Glaciologie et Géophysique de líEnvironnement
54, rue Molière, Domaine Universitaire BP 96
38402 Saint-Martin-d'Hères Cedex,France

LMDz (Laboratoire de MÈtÈorologie Dynamique ´zoomª) Atmospheric General Circulation Model is used to study sulfur cycle in Antarctica. In these remote regions, concentrations of sulfur species are very sensitive to oceanic DMS flux, particularly on the coasts where, moreover,  all the Antarctic monitoring stations are located. We are confronted with the strong lack of data concerning DMS flux or DMS oceanic concentrations in the Southern ocean, where measurements are quite sparse. Also, mechanisms leading to DMS production in the water are complex and still not well understood. The most recent global database of sea surface DMS measurements has been provided by Kettle et al. (1999) under the form of a climatology. These data are processed ´on-lineª in our model with the method of Liss and Merlivat (1986) to obtain sea-to-air fluxes. Some parameterizations of the impact of sea ice are tested. We hope to obtain both a better representation of sulfur cycle near the Antarctic coasts and a concrete interannual variability. 

Study of the reaction rate H₂ + O₃ using  isotopic form ¹⁸O₃
and FTIR technique.

 M-R. De Backer -Barilly, A. Barbe, Yu. N. Ponomarev*, V.V. Zuev*
Groupe de Spectroscopie Moléculaire et Atmosphérique, E.S.A Q6089, Université de Reims,
Faculté des Sciences, Moulin de la Housse, B.P. 1039, 51687 REIMS cedex 2, France
*Institut of Atmospheric Optics, 1 Akademicheskii Av., TOMSK 634055, Russie.

Experiments on diagnostics of the H₂¹⁸O molecule, generated in the gas phase reaction between molecular hydrogen and ¹⁸O₃ were conducted using the high resolution IR Fourier Spectrometer. As a result, we observed that the H₂¹⁸O molecules are generated in the binary mixture H₂ + ¹⁸O₃ at a rate roughly equal to 0.06 mTorr per hour when starting at an initial pressure of 1 Torr.

Interannual and Seasonal Variability of Biomass Burning
Emissions Constrained by Remote-Sensed Observations

 Bryan Duncan, Randall Martin, Amanda Staudt, Rose Yevich, and Jennifer Logan
G3B Pierce Hall, 29 Oxford St., Department of Earth and Planetary Sciences,
Harvard University, Cambridge, Massachusetts, 02138, USA

We present a method for estimating seasonal and interannual variations of biomass burning on a global scale using satellite observations. We used the TOMS Aerosol Index (AI) as a surrogate to estimate interannual variability for 20 years (1979-1999) for specific regions, such as Southeast Asia and Indonesia. Where possible, the data gap for mid-1993 to mid-1996 was filled with area burned estimates reported in the literature. Our estimates of interannual variability in biomass burning were evaluated with national statistics such as area burned and with ATSR fire-counts. Our estimate of interannual variability was applied to a biomass burning climatology and we estimated annual emissions for carbon monoxide (CO). No trend is apparent in global biomass burning emissions of CO over the last two decades, but there is significant interannual variability, particularly due to catastrophic fires in Indonesia. We implemented our estimate of interannual variation of biomass burning into GEOS-CHEM, a global model of tropospheric chemistry driven by assimilated meteorology from NASA DAO. Preliminary model results for CO for 10 years (1988-1997) are presented. Concentrations and trends of CO were evaluated with observational data from the CMDL network.

Global Tropospheric Chemistry Modelling and Preliminary Assimilation of
Column Carbon Monoxide Observation from MOPITT

 S. Edouard, J.W. Kaminski, J.C. McConnell, J.R. Drummond.
R. Menard, and P. Gauthier
Environment Canada, 2121 route Transcanadienne, Dorval, Quebec, Canada, H9P 1J3

Inversion of the CO data from the MOPITT (Measurement of Pollution in the Troposphere) instrument on board the Terra platform provide CO column and profiles with an implicit resolution of 2-3 km. We will describe the chemical module implemented on-line in the Global Environmental Multiscale (GEM) forecast model and the changes to the physics software. The assimilation component (3D-Var) used in the assimilation cycle will also be described with a particular emphasis on the representation of the background-error covariances and of the observation operator.

We will present preliminary results from column CO assimilation and analysis of the impact of CO data from MOPITT on modelled tropospheric oxidant chemistry. We will present results and discuss the methods to identify and study CO source regions, validate CO emission inventories, and contribute to air quality monitoring of large anthropogenic sources. In addition, using high resolution  (~22km) CO measurements and high resolution model simulations (GEM in regional configuration) we plan to validate convective transport of atmospheric tracers in the GEM model.

Satellite Tropospheric Trace Gas Remote Sensing  

D. P. Edwards
National Center for Atmospheric Research
PO Box 3000, Boulder CO, USA

Satellite remote sensing offers the best opportunity of making global measurements of tropospheric trace gases over extended periods of time. However, ´seeingª the troposphere presents major challenges that are only just being met. Until now, studies have relied on field campaigns, regular groundbased and aircraft measurements from specific sites, and an important input from chemical-transport modeling. Satellite instruments provide a new and exciting opportunity to extend our knowledge of global tropospheric chemistry and assess whether these measurements are consistent with our current understanding. This paper will review the current and planned US instruments with particular emphasis on the Terra platform Measurement of Pollution in the Troposphere (MOPITT) instrument status and early science results one year after launch.

Observations of Enhanced CO Concentrations from Biomass Burning in
Africa and South America as Measured by Terra/MOPITT

D. P. Edwards, J. -L. Attie, J. -F. Lamarque, J. C. Gille, and J. R. Drummond
National Center for Atmospheric Research
PO Box 3000, Boulder CO, USA

An important scientific goal of the Terra MOPITT instrument is to identify anthropogenic sources of carbon monoxide (CO) and to investigate the transport of pollution within the troposphere. In this paper we present observations of enhanced CO concentrations retrieved from MOPITT measurements over Africa, the Atlantic and South America. The seasonal variation of the CO plume origins is seen to correlate well with the burning regions as detected by the TRMM VIRS instrument, and the CO sources and resulting atmospheric distributions are compared with the predictions of the MOZART2 chemical transport model. We examine the CO convection and transport over the Atlantic as indicated by the MOPITT profile measurements and examine trajectories to study the eventual fate of the CO emitted by burning once it reaches the free troposphere. The impact on the production of tropospheric ozone is investigated using MOZART2 and compared to tropospheric ozone products derived from satellite observations.

On-Road Fleet Average Measurement of Motor Vehicle Emissions in Australia  and 
            South  East  Asia  and Comparison  with  Estimates  from Edgar v2.                                                                        

Ian E. Galbally, Ian A. Weeks, Simon T. Bentley
CSIRO Atmospheric Research Private Bag No. 1, Aspendale Vic 3195,Australia

Motor vehicles make up a substantial fraction of global anthropogenic emissions of CO₂, CO, NO_x and VOCs. However current estimates are based heavily on vehicle emission rates from Europe and North America due to the absence of testing programs in much of the rest of the world. A new method is presented, based on in-traffic or roadway sampling, that measures the integrated emissions from multiple vehicles using a particular roadway at the time of sampling. The method has been used to determine vehicle emissions of CO, NO_x, and VOCs, in Perth and Melbourne, Australia and Hongkong and Bangkok, Thailand and to prepare vehicle emissions inventories. The results will be presented and compared with estimates of vehicle emissions from these regions using the Edgar V2 database of RIVM. The new measurements presented provide a more reliable basis for estimating motor vehicle emissions in Oceania and Asia.

A Model of the Production of Methanol by higher Plants for
Global Emissions Studies

 Ian E. Galbally, Wayne Kirstine
CSIRO Division of Atmospheric Research
PMB 1 Aspendale, VIC 3195, Australia

Recently, methanol has been recognised as an important constituent of the background atmosphere, but little is known about its overall cycle in the biosphere/atmosphere system. A quantitative model is proposed for the production and emission to the atmosphere of methanol by higher plants based on plant structure and metabolic properties. The model is based on plant biochemical observations and is verified by comparison with observed methanol emission rates from plants that are independent of any information used in the model. A global rate of release of methanol from plants to the atmosphere of 103 Tg y^-1 is calculated using this model. This represents the largest single global source of atmospheric methanol. The model is in a form suitable for inclusion in global atmospheric models with an interactive biosphere. One consequence of the connection between plant growth and methanol emissions is that regional estimates of plant uptakes of carbon could be constrained by observations of the regional emissions of methanol.

 A Comparison of the regionally averaged surface exchange rates of
trace gases made through atmospheric concentration observations

I.E. Galbally, W. Zaharowski, C. P. Meyer, S. Whittlestone
CSIRO Atmospheric Research
P.B. 1 Aspendale, Victoria, Australia, 3195

A key issue in trace gas exchange studies is how to integrate from point measurements to larger scales. Here we present a methodology to compare point measurements with regional (10³ to 10⁴ km²) exchange rates based on atmospheric concentration observations. We present a numerical model of trace gas exchange in the night-time turbulent boundary layer beneath a nocturnal inversion under specific advective conditions. The ratio of the regional (10³ to 10⁴ km²) exchange rates of the two gases at the underlying land surface is uniquely related to the ratio of the atmospheric concentrations of the two gases. We have used this model to interpret simultaneous observations of ozone and radon in surface air at the coastal site at Cape Grim, Tasmania, where the land surface exchange rates of ozone and radon are approximately known from experiment. We are able to determine the best estimates of the regional land surface exchange rates of these gases and their seasonal variation. In this case of ozone and radon, the summer exchange rates are well defined and well represented by existing knowledge. In winter when the vegetation surfaces are continually wet and the underlying soil is waterlogged, the exchange rates determined from this method are very variable and poorly known from other means. This study serves to highlight these deficiencies in our knowledge and provides pointers for further research. The model is applicable to other gases and can be used either to determine an unknown exchange rate or verify an exchange rate that has been determined by means other than this approach.

Global soil-biogenic NO_x emissions and the role of canopy processes

Laurens Ganzeveld, Jos Lelieveld, Frank Dentener, Maarten Krol,
Lex Bouwman, and Geert-Jan Roelofs
Max-Planck Institute for Chemistry
Joh.-Joachim-Becher-Weg 27, P.O. Box 3060, 55020 Mainz, Germany         

Soils are an important source of oxidized nitrogen to the atmosphere. Global inventories of the soil biogenic NO_x emissions show a large range between 4 and 21 Tg N/yr. One of the uncertainties in the emission inventories is the role of the canopy interactions between emissions, dry deposition, turbulence and chemistry. Previous studies that only considered the role of dry deposition in terms of a Canopy Reduction Factor (CRF) indicate a reduction of about 50% of the globally emitted NO_x by soils. We have implemented a multi-layer trace gas exchange model in the chemistry-GCM ECHAM to explicitly calculate the role of canopy interactions in regulating the effective NO_x emissions to the atmosphere. Moreover, algorithms for the on-line calculations of soil-biogenic NO_x emissions and isoprene emissions by the vegetation have been implemented. Our new NO_x emission algorithm calculates a global soil emission flux of about 12 Tg N/yr. For comparison, we have also included a global soil NO_x emissions inventory of about 21 Tg N/yr. It appears that the enhancement of NO_x and O₃ concentrations in response to the soil emission flux is suppressed to the compensating effect of dry deposition. For sites that are exposed to relatively large emission fluxes, the multi-layer and the previously used big leaf model, which does not consider canopy interactions, calculate similar surface NO_x fluxes. This confirms the validity of the big leaf approach for most polluted regions at mid-latitudes. However, for relatively pristine sites in the subtropics and tropics, there are distinct differences between the multi-layer and big leaf NO_x surface fluxes. Use of the CRF to account for the reduction of the annual soil NO emission flux on a global scale seems feasible. However, comparison of the NO_x surface fluxes calculated with the multi-layer and the big leaf model, including the CRF, shows that there are distinct differences between the surface fluxes for specific ecosystems, e.g., tropical forest.

Trace Gases Concentrations in Amazonia

L.V. Gatti, A.M. Cordova, P. Artaxo, A. Yamazaki,
A. Guenther, J. Greenberg, F. Meixner
Instituto de Pesquisas Energeticas e Nucleares (IPEN), Travessa R, 400,
Cidade Universitaria,  Sao Paulo, Brazil, CEP: 05508-900 

As part of the LBA project, several intensive atmospheric sampling campaigns were accomplished, in AmazÙnia. During March and April 1998, a wet season, in Balbina, North of Manaus. In 1999, a wet season experiment and a dry season experiment were run in RondÙnia, at the peak of the biomass burning season. Measurements were also done in a wet season campaign in SantarÈm during Feb. 2000. A number of trace gases were measured (CO, NO, NO₂ and VOCs), simultaneously with aerosol mass, organic carbon, light scattering and absorption. Several meteorological parameters such as solar radiation, PAR, temperature, humidity, wind speed and direction were also measured. The O₃ concentration during the wet season at daytime was 9.5 - 13 ppb, while at the dry season was 41 ppb. Nighttime O₃ was measured at 5.5-9.8 ppb in the wet season, and 15 ppb in the dry season. The concentrations of NO₂ during the wet season at daytime were 0.34 ppb. At the dry season was 2.1 ppb. Nighttime NO₂ was measured at 1.1 ppb in the wet season, and 3.3 ppb in the dry season. The concentrations of CO during the wet season at daytime were 150 - 300 ppb. At the dry season was 823 ppb. Nighttime CO was measured at 230 - 350 ppb in the wet season, and 450 ppb in the dry season.

New Inventory of Global Dust Sources derived from the TOMS
Aerosol Index used in a global transport model

 Paul Ginoux¹, J. Prospero², Omar Torres³ and Sharon Nicholson⁴
(1) Georgia Institute of Technology, NASA GSFC Code 916, Greenbelt, MD-2077, 
(2)   University of Miami, 4600 Rickenbacker Causeway, Miami, FL-33149,
(3)  University of Maryland, Baltimore County, NASA GSFC Code 916, Greenbelt, MD-20771,
(4) Department of Meteorology, Florida State UniversityTallahassee, FL 32306-3034

The global distribution of absorbing aerosols index (AI) derived from Nimbus 7 Total Ozone Mapping Spectrometer (TOMS) measurements has been used to identify the major sources of atmospheric dust particles. The sources are identified on the basis of the appearance of a persistent spatial-temporal pattern in the TOMS AI distribution. The location of the source coupled with knowledge of local environments (e.g. oil characteristics, climate, etc.) enables to identify those characteristics that are most important for dust generation. These informations have been used to build a global distribution of dust sources which has been included in a transport model of mineral dust. The comparison between the model results and satellite data indicate that most dust plumes can be correctly reproduced with our dust sources inventories and that the contribution of anthropogenic disturbed sources is much less significant than previously thought. These results have been the object of three submitted papers.

In this presentation we will show the methodology followed to establish this new inventory of global dust sources. We will then show that the TOMS AI can be further used to analyze seasonal and interannual variations of dust emission. After describing the parameterization of dust uplifting and removal in a transport model, the results of several years of simulation will be analyzed by focusing on the relative contribution of the different sources to the atmospheric dust load.

Advances in Methods for Predicting
Global Emissions of Biogenic VOC

 A. Guenther
National Center for Atmospheric Research
P.O. Box 3000, Boulder CO USA 80307-3000

Research efforts and technological developments during the past decade have improved our ability to characterize regional and global scale biogenic volatile organic compound (VOC) emissions.  These advancements allow a better speciation of these emissions (i.e., estimates for individual compounds rather than lumped categories) and higher resolution (1 km2 rather than 1 degree x 1 degree).  The improved emission factors, emission algorithms and driving variables are described as well as methods for implementation in global and regional models.  Results from the new model and comparison with previous estimates are presented. Significant improvements and current uncertainties are described and future challenges and research priorities are discussed.

Development of Emissions Inventory from Biomass Burning in
Real-Time Using MODIS Data of the NASA Terra Satellite

 W.M. Hao¹, Y.J. Kaufman², D. Herring², C.O. Justice³
(1)USDA Forest Service, Fire Sciences Laboratory, P.O. Box 8089, Missoula, MT 59807, U.S.A.,
(2),NASA Goddard Space Flight Center, Code 913, Greenbelt, MD 20771, U.S.A.,
(3) University of Maryland, Department of Geography, College Park, MD 20742, U.S.A

Emissions of trace gases and aerosol particles from biomass burning have a significant impact on air quality, tropospheric and stratospheric chemistry, and global climate. Real-time information on the magnitude of fires and the amount of pollutants emitted is critical to assess quantitatively various sources of pollutants, to forecast air quality, and to model the ozone chemistry in the troposphere. We are currently developing a new method of quantifying in real-time an emissions inventory of trace gases and particulate from biomass burning in the United States. A satellite receiving station will be installed in Missoula, Mont. to retrieve MODIS data for the western U.S. from the NASA Terra satellite. Within two hours after the satellite overpass, we will be able to derive the locations of active fires, the sizes of burned areas, and fire intensities with a spatial resolution of 1 km x 1 km. We will then calculate the emissions inventory based on the burned area, the percentage of aboveground biomass burned, combustion efficiency, and the emission factor of each compound. We believe this new methodology could be applied to other regions of the world. There are currently 29 receiving stations to retrieve MODIS data worldwide. By linking the 29 receiving stations, it is feasible to derive a daily global emissions inventory from biomass burning.

Optimisation of Airborne Multi AXis DOAS (AMAX) geometries for
boundary layer measurements using radiative transfer model calculations

 K.-P. Heue, F. Finocchi, C. v. Friedeburg, U. Platt, I. Pundt, R.T. Rollenbeck, T. Wagner
Institut f¸r Umweltphysik, Universitet Heidelberg, INF 229, D - 69120 Heidelberg, Germany

The usability of the novel Airborne Multi AXis (AMAX-) DOAS technique is investigated for measurements of trace gas distributions in the planetary boundary layer. AMAX-DOAS measures the reflected and scattered sunlight from up to 10 different directions below and above the aircraft. The DOAS spectral analysis is used to obtain differential slant column densities (DSCDs) for different viewing geometries. Radiative transfer modelling serves to convert these DSCDs into vertical profiles. The quality of the profiles depends especially on the chosen viewing angles and the observed species. The instrument, which will be used for SCIAMACHY validation flights has been optimised for a flying altitude of 10000 m. Here we discuss the capabilities of the new technique for low flight altitudes for planetary boundary layer measurements with respect to BrO and HCHO emissions. For example using the above instrument at a 2000 m flight altitude, it would be possible to derive HCHO distributions with vertical resolutions of 700 m and detection limits of 1 pptv. The detection limits of the species and the spatial resolutions of the profiles will be presented for different viewing geometries, numbers of telescopes and flight altitudes.

Vehicle emissions derived by three dimensional DOAS 
tomography measurements near a motorway

  M. Hofmann, J. Loesch, J. S. Lee, P. Xie., K.U. Mettendorf, U. Platt and I. Pundt  
Institute for Environmental Physics, University of Heidelberg, Germany

We present here the geometrical setup and preliminary results from the first tomographic long-path DOAS experiment taking place in April 2001 next to a motorway. The setup includes four coaxial telescopes emitting light onto 10 retroreflecting arrays along and across the motorway. The arrays are mounted on two cranes positioned on both sides of the motorway 800 meters away from the telescopes. On each crane five arrays are placed at different altitudes: 10, 20, 30, 40 and 50 meters. They are scanned sequently by the telescopes. The retroreflectors are composed of three specular reflecting surfaces right angled oriented like a corner cube with a diameter of 60 mm. Their main property is to redirect the light in the direction of its source (in our case back to the telescopes). Thereby they compensate atmospheric refraction processes. The arrays used here consist of 10 to 20 individual retroreflectors.

Our measurements will provide vertical distributions of O₃, NO₂ and other species from 0 to 50 m altitude on both sides of the motorway. The use of 20 lightpaths will allow the calculation of three dimensional concentration fields of the measured species and hence the investigation of the vehicle emissions.

Emission of Volatile Organic Compounds (VOCs) into  
the Atmosphere by Leaves Litter

  V. Isidorov and M. Jdanova  
Department of Chemistry, St.Petersburg State University, Universitetsky pr.2, 
198904 Stary Petergof, Russia

On the global scale, phytogenic VOCs emission is considered to be 1100-1500 Tg/yr. For comparison, anthropogenic is evaluates as about 140Tg/yr. For land ecosystems, the litter mass is estimated to be about 50*10¹⁵ g/yr. That is why the leaves litter can be a potentially  important sour of VOCs in the atmosphere. The components of VOCs emission by leaves litter were identificated by GC-MS method.  This VOCs contain about 50-70 components from different classes of organic compounds (alkans, alkens, alcohols, aliphatic esters, esters of aromatic alcohols, terpens and terpenoids ). Some of this compounds are present in the essential oils of leaves, collected in the vegetation period. But most of them are not present in it and can be obtained only in the VOCs of leaves litter. For our opinion, this fact is a result of decomposed leaves litter by soil microorganisms in the autumn period.

Anthropogenic Methane Fluxes in West Siberia:  
Measurements and Modeling

Jagovkina S.V.¹, Karol I. L.¹, Zubov V. A.¹, Lagun V.E.², Reshetnikov A. I.³, Zinchenko A.V.³, 
Paramonova N.N.³, Privalov V.I.³, Rozanov E.V.⁴  
1. Main Geophysical Observatory, 7 Karbyshev Str St.-Petersburg, 194021, Russia;  
2. Arctic and Antarctic Research Institute, St.-Petersburg, 199397, Russia;  
3. Research Center for Atmospheric Remote Sensing, St.-Petersburg, 194021, Russia;  
4. Physical-Meteorological Observatory and World Radiation Center, CH-7260 Davos  &  
Institute for Climate Research ETH, CH-8057, Z?rich, Switzerland

The West Siberia is one of the main gas extracting area in the world and it can be considered as one of significant contributors of anthropogenic methane. To evaluate the intensity of West Siberian anthropogenic sources several measurement and modeling experiments were carried out. Few scenarios of anthropogenic methane fluxes were developed for West Siberia, based on the data about natural methane sources geographical distribution, about location of gas deposits and gas leakage from gas extracting objects. For testing of scenarios the 3D regional chemical transport model was applied, developed for description of methane distribution in West Siberia with resolution of 0.5⁰ (longitude) and 1⁰ (latitude) and with time step of 5 minutes. The measured near the gas deposits methane mixing ratios are compared with methane concentrations calculated for several methane emission scenarios and methane anthropogenic emission is estimated. For more accurate estimation the switched to the back trajectories calculation regime 3D model was used. On the base of summer 1999 and winter-spring 2000 measurement campaigns the methane leakage was estimated as 2.5 +/- 0.5 Mt per year. These results allowed to correct the input of this anthropogenic source into the global atmospheric methane budget.

Impact of Emissions on Modelling of Ozone Episodes with MC2AQ 

J.W. Kaminski, J.C. McConnell, L. Neary, D.A. Plummer,  
J. Struzewska and L. Lobocki  
Department of Earth and Atmospheric Science, York University, Toronto,  
Canada, M3J 1P3

A three-dimensional air quality model (MC2AQ) was developed for studying oxidant chemistry on regional to urban scales over North America and European. The model is based on the Canadian Mesoscale Compressible Community (MC2) Model, a non-hydrostatic meteorological model, to which modules permitting on-line calculations of chemical transformations, anthropogenic and biogenic emissions, and deposition have been integrated. The transport of chemical species is done on the same grid and with the same advection, convection, and diffusion schemes as are used for the meteorological fields. The developed model is highly flexible and was adapted to different scales by allowing for self-nesting.

We will present model results for several ozone episodes over North America and Europe from model integration at ~20 and 5 km resolutions. Model results where two different anthropogenic emission inventories were used for North America will be compared. Differences in predicted ozone distributions will be quantified and related to the distribution of ozone precursors.

The Tropospheric Hydroxyl Content Evolution as a Function 
of the Relation between NO_x and CO total emissions

  I.L. Karol and A.A. Kiselev  
Main Geophysical Observatory, 7 Karbyshev Str., St. Petersburg, 194021, Russia

It is well known that hydroxyl radical is the main tropospheric cleansing agent. Therefore the evolution of tropospheric gas composition and its impact on climate are significantly due to the OH content behavior there. According to many recent studies, forming of hydroxyl, carbon monoxide, ozone, methane and nitrogen oxides NOX = NO + NO2 in the troposphere is determined mainly by their photochemical interactions and by the intensity of CO, CH4 and NOX emissions into the atmosphere. The atmospheric pollution has been increasing intensively during the last decades. The atmospheric content change of OH as the main oxidant in the atmosphere has to be studied no less extensively as ozone and other radiatively important atmospheric minor constituents. The expert estimations show that the NOX anthropogenic emission rate is much higher than the CO one. At the same time the photochemical transformations mechanism depends critically on the level of tropospheric loading by NO.

            This report is devoted to the modeling of the response of the tropospheric hydroxyl content to evolution of the alignment between NOX and CO emissions. There are various scenarios of increase of the NOX and CO emissions intensity in the future with large enough range. We study the behavior of tropospheric hydroxyl content as a function of the NOX and CO emissions evolution. The special attention is paid to extreme variants where the relevant releases of NOX and CO lead to sharp decrease of tropospheric OH concentration. The causes of the such decrease are discussed and analyzed. The 30-60 deg. N latitudinal belt is under our examination. The interest to the northern temperate belt has the following motivations: I) it is the most polluted area of the Earth and ii) the highest level of pollution there decreases significantly the effect of model "noise". The MGO 2D (altitude-longitude) channel photochemical transport model is used. This model, like the 3D photochemical transport models, allows to compute the diurnal variations of the basic tropospheric components which obviously is necessary when the hydroxyl content evolution is studied. At the same time this model is more computer time saving and thus enables to carry out the numerous model runs.

Inverse Modeling of Sources of Chemically Active Gases ñ  
Problems and Progress

B. Khattatov, G. Pétron, J-F Lamarque, V. Yudin, L. Lyjak and J. Gille  
NCAR, ACD, P.O. Box 3000, Boulder CO, 80307

The MOPITT (Measurements Of Pollution In The Troposphere) instrument on board the NASA Terra satellite for the first time in history provides measurements of tropospheric carbon monoxide on a global scale. Measurements are made by intercepting the infrared radiation coming from the planet and then isolating the required signals. MOPITT is a nadir-sounding instrument since this geometry gives the maximal chance of avoiding cloud features. The field-of-view of MOPITT is 22x22km and is continuously scanned through a swath about 600km wide as the instrument moves along the orbit increasing the spatial coverage of the instrument.

MOPITT measurements provide a unique opportunity to better understand surface sources and sinks of carbon monoxide. Such task, however, is made difficult due to chemical interactions of carbon monoxide with OH and other atmospheric chemicals. These interactions make the relationship between local concentrations of CO and its surface emissions non-linear. We review challenges arising in inverse modeling of emissions of chemically active gases and present some results of our attempts to invert CO sources.

Cycling of Biogenic Sulfur Above the Eastern Mediterranean Sea.  

G. Kouvarakis¹, H. Bardouki¹, J. Sciare³ and N. Mihalopoulos¹  
1: Environmental Chemical Processes Laboratory, Department of Chemistry, 
University of Crete, P.O. Box 1470, 71409 Heraklion, Greece.  
2: LSCE, Orme des Merisiers, Bat 709, CE Saclay, 91191 Gif-sur-Yvette Cedex, France.

The spatio-temporal variation of DMS has been investigated in the Eastern Mediterranean area for the period spring 1997 to summer 1998. More than 200 DMS, data have been collected during eight oceanographic cruises. The above study consists the first approach to study the spatial and seasonal trend of DMS and its precursors in the Eastern Mediterranean sea.

Seawater DMS concentrations ranged from 1.0 to 12.7 nM with the lowest values observed during the winter (December, February and early March cruises) and the highest during spring and summer (the April, May, June and September cruises).

Annual mean seawater DMS concentration was calculated from the results of 204 samples collected all year round in the Cretan sea and averaged 3.8 nM. Our spring/summer mean value of 4.8nM is in good agreement with the value of 5.4nM reported by Simo et al. (1998) for the Western Mediterranean area and for the same period.

Despite low chlorophyll a levels recorded in the area, seawater DMS levels found in the E. Mediterranean sea are around a factor of 2 higher than what is usually reported for oceanic concentrations. An estimation of the emission of DMS from the Eastern Mediterranean sea to the atmosphere can be performed using seawater DMS data as well wind speed and sea temperature data from monthly and long-term meteorological observations. This biogenic sulfur emission is compared to the amount of sulfur deposited in the area through wet and dry deposition processes in the form of SO₂, SO₄^2- and MSA.

Western Siberia Wetlands as a Natural Source of Methane:  
3-D Regional Modeling

  Lagun V.E.¹, Jagovkina S.V.², Karol I.L.², Zubov V.A.²  
1. Arctic and Antarctic Research Institute, Saint-Petersburg, 199397, Russia  
2. Main Geophysical Observatory, 7 Karbyshev Street, Saint-Petersburg, 194021, Russia

The West Siberia region is considered as a large source of natural methane during summer season due to vast territories occupied by wetlands, river deltas, lakes, permafrost. At the same time there are very few measurements of methane fluxes carried out in separate points of this large region. These measurements cannot represent the total natural methane flux from West Siberia. To solve this problem the 3D regional photochemical transport model has been applied. The model consists of the transport (3D advection and diffusion), photochemical and planetary boundary layer parameterization modules. The developed model quantitatively describes the methane fluxes and atmospheric methane concentrations over expanded wetlands. The historical direct measurements of natural methane fluxes injected from different ecosystems of underlying surface obtained in separate parts of West Siberia are reviewed and assimilated for modelling study of methane distribution in the atmosphere over this region. On the base of comparison of modelling results and available measurements the natural methane fluxes distribution is mapped with resolution of regional scale. The results obtained have been compared with other estimations of methane fluxes for contemporary climate conditions. The share of Siberian wetlands in total natural methane flux is evaluated.

Considerations on Spaceborne Observations for Quantification of Emissions

  J. Langen  
ESA-ESTEC, POB 299, 2200 AG Noordwijk, The Netherlands

Space missions for determination of gaseous or particle emissions into the atmosphere are motivated by the need to quantify on a global level the abundances of sources for chemical transformation processes, i.e. to help understanding of atmospheric chemistry and building up predictive capabilities for future developments. They are also being proposed for the purpose of monitoring the compliance to international conventions. As vertical transport is rather slow compared to horizontal mixing the measurements need to be made in the lower troposphere ñ preferably in the boundary layer ñ and with high horizontal resolution. High background levels of many of the target species imply the need for very good accuracy. These requirements are challenging, in particular when considering cloud coverage and variability of surface emissivity which interfere in the measurement. Candidate measurement techniques will be reviewed.

Modeling Nitrogenous Gas Emissions from Terrestrial Ecosystems

  Changsheng Li  
Institute for the Study of Earth, Oceans and Space  
University of New Hampshire, Durham, NH 03824, USA

A biogeochemical model, Denitrification-Decomposition or DNDC, was developed with 10-year international efforts. DNDC predicts emissions of trace gases such as N₂O, NO, N₂, NH₃, CH₄ and CO₂ from terrestrial ecosystems (e.g., farmland, grassland and forests). DNDC simulates trace gas production and consumption by tracking coupled N and carbon (C) cycles in the plant-soil systems. DNDC consists of 6 sub-models for simulating soil climate, plant growth, decomposition, nitrification, denitrification and fermentation, respectively. Input parameters required by DNDC include climate (temperature, precipitation and N deposition), soil properties (texture, bulk density, organic matter content and pH), vegetation (crop, forest or grass types and rotations), and anthropogenic activities (fertilization, tillage, irrigation, grazing, deforestation et al.). DNDC has been validated against about forty data sets at site scale worldwide. Scaling up DNDC-modeled trace gas emissions to national scale has been conducted by the researchers of the U.S., China, the U.K., and Canada for their countries. Database construction and uncertainty analysis have been identified as two key factors for the scaling up practice.

Emission of SO₂ and NO_x from industrial area resulting from  
land use change

  Sangchan Limjirakan  
The Environmental Research Institute Chulalongkorn University  
Pathumwan  Bangkok 10330 THAILAND

Human activities have continuously induced changes on land use that consequently contribute some effects on the atmospheric level of greenhouse gases and other trace gases.  Industrial processes and agricultural activities generally release some greenhouse and trace gases to the atmosphere at some extents.  Conversion of land use from agriculture to industrial activities would release more GHG to the atmosphere.  This assumption was carried out in terms of SO₂ and NO_x emissions from this kind of land use change. The study was conducted using the LANDSAT TM imageries during the 10-year period (1987-1997) to render reliable information of land use change of the study area, an Industrial Estate in Thailand that use to be planted major economic cash crops of the nation.

The Industrial Source Complex Dispersion Model was used to study SO₂ and NO_x emissions from non-fuel combustible industrial processes in the Estate under different scenarios.  The results display maximum concentrations and the isopleths of their emissions at the community nearby, which were almost higher than the ambient air quality standard of Thailand, and the existing ambient air quality monitoring data at the study area.  This would cause some serious effects on human health and some nuisance to the people in the community nearby.

This study is a first step for an in-depth characterization of effects of land use change on greenhouse gas emissions focused on industrial activities.  The further study needs appropriate models and integrative tools for determining GHG emissions of both local and regional scales.

Black Carbon Particles Emission Estimates for the 1950 to 2100 
Period and Modeling of their Transport and Radiative Impact

  C. Liousse and H. Cachier  
Laboratoire díaÈrologie, UPS/CNRS/OMP, 14 Ave E. Belin, 31400 Toulouse, France,

Control of Black Carbon (BC) emissions emerges as a plausible tool for policy makers as part as control of global warming in the next 50 years. In this work, changes of fossil fuel BC emissions have been studied for the 1950-2100 period. From 1950 to present, BC inventory has been created following Cooke et al. (1999). Projections for 2020 and 2100 has been obtained by using fuel consumption data and population change proposed by IPCC scenarii. For any situation from the past to future, emission factors for carbonaceous aerosols were carefully adapted with attention to the nature of fuel, fuel useage and partition between transport, industrial and domestic sectors. The state of development of each country was also taken into account. These source inventories were then introduced in the European TM3 model, a 3D global off-line transport model with an horizontal resolution of 5 degrees by 3.75 degrees. The satisfactory vertical resolution allowed to prescribe adapted height injections. Global trends for modeled BC concentrations have been studied with focus on some relevant sites located in China, India and Europe. Modeling results stress the importance of the future scenario choice. At last, radiative forcing calculation has been performed for modeled fossil fuel BC for the different decades. An inventory of organic particles have been also tentatively created for the same period and the effect of these aerosols studied.

Effect of diesel formulation and engine condition on 
carbonyl exhaust emissions and on the distribution of polycyclic 
aromatic hydrocarbons (PAHs) in the exhaust particles

C. Maldonado, M. Pons  and  Klaus Wirtz  
Fundacion Centro de Estudios Ambientales del  
Mediterraneo, CEAM, C/Charles Darwin 14, 46980 Parque Tecnologico,  
Paterna (Valencia),  SPAIN

The aim of this work was the study of the effect of different Diesel formulations and engine conditions (i.e. torque (Nm) on the levels of carbonyl compounds and CO2 in the exhaust gas, and on the distribution of PAHs in Diesel exhaust partCarbonyl samples were collected from the diluted exhaust gas by using DNPH-cartridges with subsequent HPLC-UV/VIS detection of the formed hydrazones Formaldehyde, acetaldehyde, Acetone and propanal could be identified as major carbonyls emitted by the engine. The highest emissions of the carbonyls, normalized by the measured CO2 values, were found under Standby condition of the engine for all Diesel formulations investigated. With increasing load an engine revolution carbonyl Individual PAHs present in the exhaust particulates filters were analyzed by GCMS after extraction with dichloromethane and silica gel fractionation. Standard diesel showed under most conditions the highest particle mass emission. The collected particle mass normalized by the measured CO2 values showed under all operating conditions lower values for the Diesel fuel with the lowest aromatic content.

The Treatment of Long-Lived, Carbon-Containing Products in 
Inventories of Carbon Dioxide Emissions to the Atmosphere

Gregg Marland and Eric Marland  
Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6335, USA,
  and Department of Mathematical Sciences, Appalachian State University, Boone, NC 28608, USA

Estimating CO₂ emissions to the atmosphere is complicated by the fact that consumption of fossil fuels and harvesting of forests do not necessarily mean that the contained C has been released to the atmosphere as CO₂.  Some fraction of fossil fuels and harvested wood are embodied in products that have lifetimes ranging from months to centuries.  The IPCC methodology addresses fossil-fuel products by assuming that some prescribed fraction flows to long-lifetime applications while the remainder is oxidized instantly.  The IPCC methodology is attractive because it relies only on data for the current rate of production, data that should be readily available.  What we really want to know is the annual rate of oxidation of the pool of products, or the extent to which production exceeds oxidation (or vice versa), and this is the annual rate at which the pool of products increases.  The question posed here is whether the IPCC methodology can provide a reasonable estimate of what we really want to know, i.e. can the increase in stocks of long-lifetime products be reasonably represented as a simple fraction of the current rate of production? We explore the increase in stocks as a function of current production for a variety of production rates and oxidation rates, and show how these can be used in estimates of CO₂ emissions.

Contribution of Changing Sources and Sinks to the Growth Rate of  
Atmospheric Methane Concentrations for the Last Two Decades

E. Matthews¹, B. Walter², J. Bogner³, D. Sarma⁴, G. Portmey⁵  
(1) NASA Goddard Institute for Space Studies, 2880 Broadway, NY, NY 10025, USA  
(2) Columbia Univ. Center for Climate Systems Research, NASA  
Goddard Institute for Space Studies, 2880 Broadway, NY, NY 10025, USA  
(3) Landfills+, Inc., 1144 N. President St., Wheaton, IL, 60187, USA  
(4) New York University, NY, NY 10013, USA  
(5) College of Engin. and Mines, Dept. Mining and  
Geological Engineering, University of Arizona, Tucson, AZ 85721, USA

In situ measurements of atmospheric methane concentrations begun in the early 1980s show decadal trends, as well as large interannual  variations, in growth rate.  Recent research indicates that while wetlands can explain several of the large growth anomalies for individual years, the decadal trend may be the combined effect of  increasing sinks, due to increases in tropospheric OH, and  stabilizing sources.  We discuss new 20-year histories of annual, global source strengths for all major methane sources, i.e., natural wetlands, rice cultivation, ruminant animals, landfills, fossil fuels, and biomass burning.  We also present estimates of the temporal pattern of the sink required to reconcile these sources and atmospheric concentrations over this time period.  Analysis of the individual emission sources, together with model-derived estimates of  the OH sink strength, indicates that the growth rate of atmospheric methane observed over the last 20 years can only be explained by a combination of changes in source emissions and an increasing tropospheric sink.  Direct validation of the global sources and the terrestrial sink is not straightforward, in part because some sources/sinks are relatively small and diffuse (e.g., landfills and soil consumption), as well as because the atmospheric record integrates multiple and substantial sources and tropospheric sinks in regions such as the tropics.  We discuss ways to develop and test criteria for rejecting and/or accepting a suite of scenarios for the methane budget.

Acetone production from the atmospheric degradation  
of oxygenated  VOCs

A..Mellouki, I. Magneron and G. Le Bras  
LCSR/CNRS, 1C avenue de la recherche scientifique, 45071 OrlÈans-France  
mellouki@cnrs-orleans.fr

In addition to its anthropogenic and biogenic primary sources, acetone is produced in the atmosphere through the degradation of other volatile organic compounds (VOCs), including alcanes, alkenes and oxygenated compounds. Laboratory mechanistic studies of the atmospheric oxidation of several oxygenated VOCs, such as ethers, ketones, aldehydes, have shown that acetone is produced with significant yields. Injection to the atmosphere of large amounts of this type of oxygenated VOCs having a significant acetone forming potential can therefore contribute to the overall tropospheric budget of acetone and then influence the ozone concentration in the upper troposphere (UT). Based on recent laboratory studies, including our own studies, the potential importance of the production of acetone from the degradation of a number of anthropogenic oxygenated VOCs will be presented.

Comparison of Algorithms for the Reconstruction of Trace 
Gas Distributions Derived by Tom-DOAS
(Tomographic-Differential-Optical-Absorption-Spectroscopy)

K. U. Mettendorf, M. Hofmann, J. Lˆsch, U. Platt, and I. Pundt  
Institute for Environmental Physics, University of Heidelberg, Germany

Tomographic-Differential-Optical-Absorption-Spectroscopy (Tom-DOAS) is a new application of the DOAS method designed to map different trace gas concentration distributions. Longpath-DOAS-systems measure column densities, which are the integrated concentration along the light paths, between two fixed points. To reconstruct the  concentration distribution from its column densities is  the task of tomography. Compared to tomography in medicine the number of line integrals is much smaller. It is limited by the number of useable DOAS-systems. Therefore the measurement geometry and reconstruction method play an important role for the quality of the reconstruction.

In this work a comparison is made between different algorithms the Algebraic Reconstruction Technique (ART), ART with Weights (ARTW) and others methods, which will be used in the distribution reconstruction of trace gases like O₃, NO₂ and other species. They will be applied to the geometry, which will be used during a measurement campaign near a motorway in April/May 2001 .

A Spatially Explicit Inventory of Trace Gas Emissions from 
Wildfires and Controlled Burning over Australia

C.P. (Mick) Meyer, G. Cook, K.G. Tolhurst, I.E. Galbally and D. Graetz.  
CSIRO Atmospheric Research, Aspendale, Vic., Australia

Australia is the driest continent and its landscape is regularly burned by wildfires and controlled burning. We present a spatially explicit analysis of the extent and amount biomass burning in Australia and the resultant release of trace gases. This analysis utilises satellite-based remote sensing of fire scars, ground based records of smaller fires, primarily forest fires and controlled burns, direct observations and biological modelling of fuel loads and locally observed fuel composition and emission factors averaged within each of the 80 biogeographical regions of Australia. Ninety percent of the burning occurs in northern Australia. Approximately 5% of the Australian land surface is burned annually, and approximately 10% of the net primary productivity is consumed in these fires. The release of trace gases from biomass burning over Australia, for several major trace species including CO₂, CO, N₂O and total non-methane hydrocarbons, equals or exceeds that due to fossil fuel combustion. These emissions represent 10% of the total emissions from biomass burning in the southern hemisphere.

THE GEIA Center 

Paulette Middleton  
RAND Environment

The GEIA Center located at http://www.geiacenter.org is the virtual data distribution and communications hub for GEIA.  Details regarding GEIA operations and its plans for the future, all of the available data and project descriptions, and other news and outreach activities can be found at the Center.   As part of the expanded outreach strategy to enhance the exchange of information on GEIA and to encourage participation in GEIA, the GEIA Center recently developed the GEIA Brochure and GEIA Newsletter.  The brochure provides a summary of GEIA and is being updated on a regular basis as new projects and other information become available.  In addition GEIA is expanding its e-mail network which now includes over 500 users/developers, and establishing more two-way links to related web sites.  In order monitor and improve GEIA data use, the Center, since 1997, has been tracking the number of visitors to the web site, the number and type of downloads, and usersí comments regarding their use of the data.  Others are encouraged to join this international community and all are welcome to provide recommendations for future data and outreach activities.  The Centerís activities are supported by the National Science Foundation (NSF) and National Aeronautics and Space Administration (NASA).

TROTREP : Tropospheric Ozone and Precursors,  
Trends, Budgets and Policy

  Paul S. Monks, Michiel Roemer, Evi Sch¸pbach, Anne Lindskog , Jan Pieter Bloos, 
Sverre Solberg, Jan-Eiof Jonson, Andrew Rickard and Frank Dentener  
Department of Chemistry, University of Leicester, Leicester, UK: e-mail

The chemical composition of the atmosphere over Europe is changing.   Many different factors affect the composition of the troposphere over Europe on a range of scales. It is widely recognised that photochemical oxidant production is a problem on the local, regional and global scale. In order to address these problems, the European Union has put forward and implemented a number of different policies and legislation to improve air quality.       One of the fundamental questions at the heart of the air quality problem is ´how much has ozone changed by abatement measures in the past and how much more will it change under the current and future emission reduction plans?ª.   The prime objective of TROTREP is to evaluate, validate and predict the effectiveness of past and future EU air quality legislation with respect to ozone and its precursors.

In order to achieve the prime objective of TROTREP an integrated measurement and modelling approach is being developed.   Particular areas being investigated include the trend analyses of ozone and precursor concentrations over the last 100 years in relation to changes in emissions.  The shape, form and variability of seasonal cycles of ozone and its precursors as a proxy for changing atmospheric composition.   The interrelationships of photochemistry, emissions, atmospheric circulation and land-use in controlling ozone and precursor trends/seasonal cycles.   The role and sensitivity of photochemistry to changing atmospheric composition.

In order to investigate the role of changing emissions on ozone and precursors, a time series of emission data for CO, NO_x and NMVOC is being prepared based on present emission inventories and historical trends in activity statistics (1970 ‑ 1998). The basis for this is the global emission database EDGAR (Emissions Database for Global Atmospheric Research).    Applying historical trends in activity statistics for the period 1970 ‑ 1998 combined with the changes in emission factors due to technological developments has resulted in a source oriented series of emissions for CO, NO_x and NMVOC for 1970 ‑ 1998.   Historical time series of emissions of NO_x, VOC and CO are being constructed, back to 1870 with a time resolution of 10 years and on a somewhat coarser spatial resolution (1 _ x 1 _ ).  Emission factors and national consumption figures from OECD energy statistics for the period 1955-1985, together with estimates of increase in Eastern-Europe emissions and nationally reported emission numbers from EMEP will be used to reconstruct national European emission data for NO_x back to the 1950s.  

The work presented will detail the TROTREP project and particularly focus on the work being conducted on the generation of new emission inventories and their application in trend calculations and modelling studies.

An Evolutionary Strategy to estimate the Distribution of 
Emission Sources on a Regional Scale from Atmospheric Observations.

P. O'Brien, D. Corcoran, A. Roddy and D. Lowry  
Meteorology Section, Matin Ryan Marine Science Institute,  
National University of Ireland, Galway, Galway, Ireland.

An Evolutionary Strategy (ES), by which the regional distribution and strength of emission sources is estimated from a combination of atmospheric monitoring and back-trajectory analyses, is presented. The observed burden of a species above background is associated with a particular air mass, whose recent movements are estimated. The region is divided into an appropriate grid, and the observed burden equated to the sum of unknown emissions from each cell along the back trajectory. Thus a series of linear equations is generated. An ES has been developed to solve these equations for the spatial distribution of emissions. An initial population of random emission maps is generated with few a priori contraints on the distribution. Those members of the population which best reconstruct the observed signal are retained and pass their characteristics onto the next generation of solutions. The poorest solutions ìdieî, and are replaced by offspring ìbredî from the better solutions. Over many generations, the population tends towards solutions which result in increasingly more realistic emission maps. The ES is applied to artifical test distributions and to methane measurements from Mace Head, Ireland. The latter yields an estimate of the source distribution across western Europe and the North Altantic.

Estimates of Global and Regional Halocarbon Emissions from Measurements on the 
Cape Grim Air Archive and from the CARIBIC Aircraft Experiment

  D.E. Oram¹, C.E. Reeves¹, G.A. Sturrock¹, S.A. Penkett¹, P.J. Fraser²,  
C.A.M. Brenninkmeijer³ and A. Zahn³.  
¹ School of Environmental Sciences, University of East Anglia, Norwich, UK  
² CSIRO Atmospheric Research, Victoria, Australia  
³ Max Planck Institute for Chemistry, Mainz, Germany.

The Cape Grim archive, a unique collection of large volume air samples collected in Tasmania (41 _ S) since 1978, has been analysed by GC-MS for a wide range of halocarbons, including CFCs, HCFCs, halons, HFCs and PFCs. Collected under baseline conditions, the samples (~100) are representative of the background SH. The observed concentrations and trends are compared to those derived using 2-D global models based on industrial emission estimates. For some compounds, e.g. H-1211 and HCFC-142b, there are significant discrepancies that remain unresolved. Where industrial estimates are not available, the models can be used to derive independent emission scenarios based on the Cape Grim record. This is particularly relevant for many fluorinated Kyotoª gases (HFCs, PFCs), whose emission rates are poorly understood. In addition to ground-based sites like Cape Grim, regular data are now being obtained from the upper troposphere/lower stratosphere as part of the European project CARIBIC, by sampling from a commercial airliner flying between Germany and Asia/Africa. Early results have identified a number of different source regions/processes for a wide variety of chemical species.

Critical Evaluation by Aircraft Measurements of  
Emission Inventories for U.S. Point Sources

D.D. Parrish, D.K. Nicks Jr., T.B. Ryerson, J.S. Holloway, G.J. Frost,  
M. Trainer, G. Hubler, D.T. Sueper and F. C. Fehsenfeld  
Aeronomy Laboratory, National Oceanic and Atmospheric Administration, Boulder, CO 80305, USA

We have measured CO₂, nitrogen oxides (NO_x), sulfur dioxide (SO₂) and carbon monoxide (CO) from an aircraft in U.S. point source plumes.  The measurements provide two critical tests of emission inventories for particular point sources, and, by extension, U.S. point source emission inventories in general.  First, correlations between measured species can be directly compared with ratios of tabulated emissions.  If necessary, account is taken of differential loss rates; in this regard, correlations with CO₂ are especially useful, since CO₂ is not subject to photochemical loss processes.  Second, when the chemical measurements are combined with wind speed and direction information for the time of plume emission, absolute emission fluxes can be derived by integrating the species concentration above background across the plume transect.  In general we have found good agreement with emission inventories for power plants, with the exception of sometimes surprisingly large emissions of CO.

Decadal Change in Carbon Monoxide to Nitrogen Oxide  
Ratio in U.S. Vehicular Emissions

D.D. Parrish, M. Trainer, D. Hereid, E.J. Williams, K.J. Olszyna,  
R.A. Harley, J.F. Meagher, F.C. Fehsenfeld  
Aeronomy Laboratory, National Oceanic and Atmospheric Administration, Boulder, CO 80305, USA

Measured carbon monoxide (CO) and nitrogen oxide (NO_x) concentrations in U.S. urban areas show that the CO/NO_x vehicular emission ratio has decreased at an average annual rate of 7 to 9% from 1987 to 1999. The trend in maximum ambient CO levels in U.S. cities suggests a 5 to 6% average annual decrease in CO vehicular emissions, which implies a 2 to 3% annual increase in NO_x emissions from vehicles.  The U.S. Environmental Protection Agency (EPA) estimations of pollutant emissions indicate a factor of 2 to 3 smaller rate of decrease in CO emissions with no significant change in NO_x emissions.  In 1990 vehicle emissions of CO were 53 Tg of the estimated 89 Tg total U.S. emissions.  Thus over the decade of the 1990s U.S. CO annual emissions decreased by 22 Tg yr^-1.  This represents a decrease of 4 to 7% in the total global anthropogenic CO emissions of 300-550 Tg yr^-1, a decrease not accurately included in inventories based on U.S. EPA tabulations.  The VOC vehicular emission rate has decreased as much or more than for CO, implying that the VOC to NO_x ratio in the U.S. urban atmosphere has decreased by a factor of at least 3 over this period.

Emissions from Biofuel Combustion in India  

D.C. Parashar, Ranu Gadi, A.K. Sarkar, A.P. Mitra, and U.C. Kulshrestha^*  
National Physical Laboratory, New Delhi-110 0, India  
^*Indian Institute of Chemical Technology, Hyderabad-500 007, India

Concentrations of Organic and elemental carbon and SO₂ and NO_X in the atmosphere are strongly influenced by the emissions taking place from the combustion of biofuels. Measurements over the Indian Ocean during the INDOEX campaigns have shown signatures of biomass burning in the tropics. Major biomass burning is from rural cooking practices which in this region, come predominantly from wide use of fuelwood, dungcakes, agricultural residue and charcoal. Efforts have been made to study the emissions from biomass used for energy in India. An experimental setup has been designed to carry out controlled burning experiments and evaluate the emission factors for Organic carbon, Elemental carbon, SO₂ and NO_X from various biofuels commonly used in India.

The preliminary results obtained so far show the emissions of organic carbon, elemental carbon and SO₂ are maximum from dungcakes and minimum from charcoal and for NO_X they are maximum from fuelwood and minimum from agricultural residue. The budget estimate has a limitation as the information on the biofuel consumption is very scarce. The preliminary budget estimates for elemental carbon, organic carbon, SO₂ and NO_X are 1.2, 2.9, 0.7 and 0.6 Tg respectively.

Emissions of Carbon Monoxide and Carbonaceous Aerosols  
from 1950 to 1995

Joyce E. Penner and Brandon S. Preblich  
Department of Atmospheric, Oceanic, and Space Sciences  
University of Michigan

We have developed a system for defining emissions from fossil fuels over the time period 1950 through 1995. A data base for fuel use has been compiled by country and year for the time period from 1950 to 1995. The categories include transportation uses of fossil fuels, industrial uses, and residential uses. In addition, for emissions of CO, industrial production processes that contribute to CO emissions have been compiled. This information, together with emission factors by fuel use or industrial process for each country can be used to estimate emissions on a 1 x 1 degree grid for use by atmospheric models. Here, we present estimates for the annual emissions of CO and carbonaceous aerosols for developed and developing countries. The emissions for carbonaceous aerosols are compared with similar emissions for SO2. The implications for the net climate forcing from fossil fuel carbonaceous and sulfate aerosols are discussed.

Inverse modeling of CO emissions

G. Pétron^1,2, C. Granier^1,3,4, B. Khattatov², J.F. Lamarque²_,  V. Yudin²  
(1)   Service d'Aéronomie, Paris, France, (2) NCAR, Boulder, Colorado, USA  
(2)   NOAA/CIRES Aeronomy Laboratory, Boulder, USA  
(3)   Max Planck Institute for Meteorology, Hamburg, Germany

Inverse modeling techniques have been developed for a few decades to better assess long-lived species emissions like CO₂ and CH₄. These techniques are still being worked on and we now try to adapt them to short-lived components. Carbon monoxide has an average global lifetime of about two months and it is the principal sink for OH in the free troposphere. CO is a byproduct of fossil fuel and biomass incomplete combustion. The incomplete oxidation of hydrocarbons also produces substantial amounts of CO. The oxidation of methane probably contributes to up to 30% of total CO budget, yet the uncertainties on CO sources are still high (factor of 2 or 3). Therefore, to better estimate CO sources we also need to better know the emissions of methane. The MOPITT instrument (Measurements Of Pollution In The Troposphere) was launched on December 18, 1999, onboard the Terra platform. The retrieval of MOPITT radiance measurements provides us with concentration profiles of CO (resolution : 20km horizontal and 3km vertical) as well as column amounts of methane. A ´globalª coverage of the Earth is completed in 3 days and the mission is supposed to end in 200? which puts MOPITT at the first place as far as observation frequency and resolution and time/season coverage is concerned. At the National Center for Atmospheric Research, we develop tools to better quantify global emissions of CO and CH₄ using the 3D chemistry transport model MOZART combined with the MOPITT data. The techniques and the first results will be presented and discussed. 

Three dimensional measurement of  trace gas concentrations in the 
atmosphere by DOAS Tomography: Overview

I.Pundt, F. Finocchi, Ch.v. Friedeburg, K.P. Heue, M. Hofmann,
  G. Hˆnninger, J. Lˆsch, K.U. Mettendorf, U. Platt, R. T. Rollenbeck, T. Wagner, und P. Xie.  
Institute for Environmental Physics, University of Heidelberg, Germany

An overview will be given about the new Tomographic Differential Optical Absorption Spectroscopy technique, which will deliver two or three-dimensional concentration fields of different trace gases. From the tomographic observations it is possible to calculate the emissions of different trace gases, which then can be extrapolated to global emissions. The tomographic set-ups will comprise more than 10 light paths. Measurements are performed from aircraft and ground by the so-called MAX (Multi Axis) sunlight DOAS and at ground with long-path DOAS instruments. From aircraft the sunlight, which is scattered by the earth atmosphere and/or reflected on the ground is captured from about 10 different directions. The ground-based Long-path tomographic DOAS instruments use artificial light sources,  retroreflectors and possibly mirrors redirecting the light into the telescope. In the future it will be possible to use these set-ups over vegetation (e.g. forests or agriculture areas), urban or industrial areas for the measurement of natural or anthropogenic emissions. The quality of the data, the spatial resolution and detection limits will be discussed depending on the instrument performances, viewing angles, number of light paths and other parameters.

Measurements of the GOME Satellite Instrument

A. Richter, F. Wittrock, A. Ladstaetter-Weissenmayer, H. Nuess, L. Hild and J. P. Burrows  
Institute of Environmental Physics, University of Bremen  
P.O. Box 33 04 40, D-28334 Bremen

The GOME (Global Ozone Monitoring Experiment) instrument on board the European ERS-2 satellite launched in April 1995 provides global measurements of the radiance reflected from the earth and scattered back in the atmosphere. The instrument covers the UV and visible spectral range (240 ñ 700 nm) with moderate spectral resolution (0.2 ñ 0.4 nm). From these measurements, vertical columns of a number of species have been derived, including O₃, NO₂, HCHO, BrO, OClO, H₂O, O₄, and SO₂. For tropospheric absorbers, the GOME instrument provides for the first time the opportunity to derive global maps of tropospheric NO₂, HCHO and SO₂, and also of BrO. By comparison of these global data sets with model results, and possibly application of inverse modeling techniques, the GOME measurements should provide significant input for improvements of current emission inventories.

In this paper, global maps of the tropospheric columns of the above mentioned species will be presented, and the accuracies and possible retrieval problems be discussed in view of their application for emission studies.

Modelling the Global Sources and Sinks of Methane

N. J. Rolfe, K.S. Law, E.G. Nisbet, C. Bridgeman& J.A. Pyle  
Centre for Atmospheric Science, ChemistryDeptartment,  
University of Cambridge, Cambridge, UK.

As methane is both a greenhouse gas and an active constituent in tropospheric chemistry, it is important to understand the factors controlling methane levels in the atmosphere. Three-dimensional global modelling enables our current understanding of the global methane budget to be tested and can also serve as a basis for testing and verifying emission estimates. Methane concentrations within global models depend on sources, atmospheric and surface sinks and transport. By including the current best estimates for the atmospheric sinks of methane and the magnitude, location and seasonality of emissions, it is possible to quantify the methane budget by source, latitude and season. The accuracy of the model and our understanding of methane concentrations can then be tested by the comparison of modelled values with observations. A comparison of several independent three-dimensional global models is required in order to gain a true understanding of the global methane distribution and recently observed changes in the growth rate. In this work, a three-dimensional global atmospheric model, TOMCAT, has been used with a simple chemistry scheme to model the spatial and temporal behaviour of tropospheric methane. In order to test the performance of the model, TOMCAT was run for 19 years using meteorological analyses provided by ECMWF. Modelled concentrations of methane have been analysed and compare well with monthly mean data. To further test our understanding of the methane budget, experiments investigating the influence of meteorology on global methane concentrations and the contributions of individual methane sources to the total methane mixing ratio have been completed

Influence of the source formulation on modeling the 
atmospheric global distribution of sea-salt aerosol

M. Schulz, Y. Balkanski, W. Guelle  
Laboratoire des Sciences du Climat et de l'Environnement, CEA/CNRS  
F - 91191 Gif-sur Yvette Cedex

Three different sea salt generation functions are investigated for use in global 3D atmospheric models. Complementary observational data are used to validate an annual simulation of the whole size range (film, jet and spume droplet derived particles). Aerosol concentrations are corrected for humidity growth and sampler inlet characteristics. Data from the North American deposition network are corrected for mineral dust to derive sea salt wet fluxes. We find that sea salt transport to inner continental areas requires substantial mass in the jet droplet range, which is best reproduced with the source of Monahan et al (1986). The results from this source formulation also shows the best agreement with aerosol concentration seasonality and sea salt size distributions below 4 µm radius. Measured wind speed dependence of coarse particle occurrence suggests that above 4 µm the source from Smith and Harrison [1998] is most appropriate.

Mineral Dust Source Strength
derived by an Inversion Procedure Based on Satellite Observations

M. Schulz, C. Moulin, P. Peylin, Y. Balkanski, T. Claquin  
Laboratoire des Sciences du Climat et de l'Environnement, CEA/CNRS  
F - 91191 Gif-sur Yvette Cedex

This study utilizes satellite data of aerosol optical thickness to derive globally source strength of the mineral dust aerosol. Our transport simulations in the TM2/3 and two GCM's (ECHAM and LMDZ) have been developed recently to sufficient accuracy with respect to three dimensional dispersion patterns and aerosol size distribution, that inversion techniques can now be applied to refine source function parameters. These constitute the subgrid-fraction of a desert area acting as a dust source, the effective upward transport of surface mobilized dust, and any contribution from resuspended mineral dust. Meteosat satellite data were compiled to have a large set of daily observations of mineral dust for the North Atlantic and the Mediterranean Sea. We are also about to apply the method globally to invert aerosol source strength from Polder satellite data. The model resolution in the first runs is set to 2.5_*2.5_ with 9 layers in the vertical. The inversion is based on the prognostic aerosol optical depth in marine cloudfree ocean areas, derived from running the model in a forward mode for each of the grid boxes representing a desert area and thus a possible dust source. The results are interpreted by comparison with another global mineral dust source formulation which we developed recently on the basis of mapping globally the mineralogy of desert soils.

Generation of Source-Receptor Matrices with a Backward-Running 
Lagrangian Particle Dispersion Model

Petra Seibert  
Institute of Meteorology and Physics, University of Agricultural Sciences Vienna,
Tuerkenschanzstr. 18, A-1180 Wien, Austria.

Source-receptor matrices are an important intermediate product for the inversion of measurements with respect to emissions of trace substances, especially for linear source-receptor relationships. If the number of measurements is smaller than the number of resolved source elements, a backward/adjoint transport & dispersion simulation is computationally more efficient than a forward simulation. A new method is presented which allows to employ a Lagrangian particle model with its superior numerical properties in a backward-running mode to deduce source-receptor matrices. The method will be illustrated with the ETEX experiment (including the inversion of the matrix), and possibly other applications (Comprehensive Test Ban Treaty verification, sulfur emissions in Europe).

A sulphur dioxide and aerosol emissions inventory for India for implementation in the LMD-GCM  

M. Shekar Reddy^1,2, Chandra Venkataraman¹, Olivier Boucher^2,3, Mai Pham^{4  
1}Centre for Environmental Science and Engineering,  
Indian Institute of Technology, Bombay, Powai, Mumbai 400 076, India.  
²Laboratoire d'Optique Atmosphérique, Université de Lille, France  
³On sabbatical at the Max-Planck-Institute of Chemistry, Mainz, Germany  
⁴Service d'Aéronomie, Université P. et M. Curie, Paris, France

Significant uncertainties in the assessment of aerosol-climate interactions arise from uncertainties in estimated pollutant emissions or the source term used in climate models. The focus of this work is the development of an emissions inventory for India, with a spatial resolution of 0.25^o x 0.25^o (approximately 25 km x 25 km), suitable for regional-scale climate studies. Pollutants, of relevance to climate, included are sulphur dioxide (SO₂), and aerosol chemical constituents (fine particles smaller than 2.5 µm in diameter, black carbon, organic matter and "inorganic fraction") emitted from fossil fuel and biomass combustion in India.

            A fossil fuel consumption database has been constructed for 1996-97, including coal, lignite, petroleum and natural gas consumption in electric utilities, industrial, domestic and transportation sectors. SO₂ emission factors were calculated using India specific sulphur content of coal and petroleum fuels, and sector specific combustion technology features. Aerosol emission factors for stationary sources were derived from the U.S. EPA AP-42 compilation and for domestic and transportation sources, were compiled from literature, considering Indian source characteristics. Biomass burning activity data, including biofuel consumption (wood, crop waste and dung cake in cooking stoves) and forest burning were estimated. Fuel specific sulphur content was used to calculate SO₂ emissions from biomass combustion. Particulate matter emission factors compiled from studies of Indian cooking stoves, and best available data in literature for carbonaceous constituents have been used to estimate emissions.

            Estimated SO₂ emissions were 4.35 Tg SO₂ yr^-1, primarily from fossil fuel combustion (93%), with large industrial sources accounting 58% of total SO₂ emissions. Fossil fuel and biomass combustion contributed equally to estimated particulate matter emissions at 4.04 Tg yr^-1. Use of high ash coal (~39%) in electric utilities, and fuelwood and dung-cake combustion in cooking stoves were major contributors of particulate matter emissions. Fly-ash from coal combustion in electric utilities alone contributed 65% of total "inorganic fraction" emissions (2.49 Tg yr^-1). Biomass combustion was major source of carbonaceous aerosols in India, accounting 72% of the black carbon (total 0.35 Tg yr^-1) and 76% of the organic matter (total 1.21 Tg yr^-1) emissions. Other important sources of carbonaceous aerosol emissions were diesel use in road transport, and a region specific source of coal combustion in brick-kilns. Contribution of specific industrial sectors and individual fuels to regional pollutant emissions will be discussed. Results will be interpreted in light of existing emissions estimates for India and Asia. A GCM aerosol simulation corresponding to the INDOEX Intensive Field Phase 1999 using this emissions inventory will also be presented.

Biogenic Volatile Organic Compound (BVOC) Emission of 
European Forests under Future CO₂ Levels: Influence on Compound Composition 
and Source Strength (FUTURE-VOC)

R. Steinbrecher, R. Baraldi, M. Habram, N. Hewitt, Ch. Koerner, F. Loreto,
  F. Miglietta, S. Pepin, J.-P. Schnitzler, P. Scholefield,  
Fraunhofer-Institut f¸r Atmosph‰rische Umweltforschung, Kreuzeckbahnstr. 19, 
D-82467 Garmisch-Partenkirchen; steinbrecher@ifu.fhg.de

Forests are a major source of volatile organic compounds (VOC). Plant species differ in VOC emission and emissions are strongly co-determined by concurrent climate. Biogenic VOC play an important role in the oxidation processes of the lower troposphere contributing to high ozone and photochemical smog during periods of high radiation and temperature. Therefore, there is a pressing need to understand the effect of plant diversity and the effect of global change factors such as elevated CO₂ on the source strength and pattern of VOC emission of European forests for establishing effective photooxidant reduction strategies and for correctly assessing the carbon sinks in the biosphere.

The project uses existing infrastructure and data bases available in Switzerland (Swiss canopy crane project SSC near Basel), Italy (CO₂ springs in Tuscany) and the United Kingdom (controlled environment chambers in the Lancaster). Leaf-level and branch-level gas exchange measurements on key European plant species as well as biochemical studies related to the formation of isoprene and monoterpenes in the leaves are performed under ambient and elevated CO2 levels. These data are used to improve current models used for estimating future BVOC emission source strength of typical forested areas. First results of the investigations are presented.

System for observation of halogenated greenhouse gases
in Europe (SOGE): Verification of emissions

F. Stordal¹, N. Schmidbauer¹, P. Simmonds², A. McCulloch², S. Reimann³, 
M. Maione⁴, E. Mahieu⁵, J. Notholt⁶, I. Isaksen⁷, R.G. Derwent^{8  
1}Norwegian Institute for Air Research (NILU), P.O.Box 100, 2007 Kjeller, Norway.  
²University of Bristol, ³Swiss Federal Laboratories for Materials Testing and Research (EMPA), 
⁴University of Urbino, ⁵University of LiËge, ⁶Alfred Wegener Institute for Polar and Marine Research, 
⁷University of Oslo, ⁸UK Meteorological Office

SOGE is an integrated system for observation of halogenated greenhouse gases in Europe. A network of four atmospheric research stations is established, with full intercalibration of a range of CFCís, HCFCís and HFCís. For PFCís and SF₆ a technique for high-frequency measurements will be developed. These in situ measurements will be combined with vertical column measurements to derive trends in concentrations. The integrated system will be used to verify emissions in Europe down to a regional scale. The results can be used to assess compliance with the international protocols regulating the emissions, and they will be utilised to define criteria for future monitoring in Europe. This activity builds on a priori emission data provided by industry, regulatory bodies and on published data in the scientific literature. Verification of these data will be based on a dispersion model of a Lagrangian type, in which emissions are modelled by releasing large number of ëparticlesí into the ëmodel atmosphereí. In addition, global models will be used to estimate impacts of the observed compounds on climate change and the ozone layer.

Radiative Modeling of a Vegetation Fire 
for Remote Sensing Applications

A. Strulovici, B. Porterie , P. Simoneau, M. Schoenermark, M. Bittner and A. Roblin.  
MPSO /DOTA, ONERA, Chemin de la Hunière
F-91761 Palaiseau Cedex, France

Among means to estimate biomass burning emissions, satellite remote sensing is coming to maturity, enabling the building of global maps of emission, operational detection and monitoring of fires.

At the beginning of the presentation we will give an overview about methods currently applied in order to quantify biomass burning emissions using satellite data. We will then  give an outlook to methods proposed for satellite systems operating in the near future.

In this perspective, modeling the fire signal from ground to satellite is of interest as it takes into account the contribution of each part of the atmosphere-fire system: flame, burned vegetation, smoke  and clouds , thus enabling the finding of new detection algorithms (and their limits)  fitted for the diverse instruments that are to be launched in the next 5 years (BIRD, FOCUS, FUEGO).

 To simulate a vegetation fire, we use a fire propagation model, then use a radiative transfer code (line-by-line) to propagate the fire signal to the satellite, taking into account the soot particles and the possible convection clouds that can develop above the fire.

We will present the details of this method and the radiation characteristics for a dry savanna fire.

Modeling Volcanic Sulfur from Mount Erebus and Cerro Hudson  
in LMDz Atmospheric General Circulation Model

K. Sturm, E. Cosme, C. Genthon and R. Delmas  
Laboratoire de Glaciologie et Géophysique de l'Environnement CNRS/OSUG  
54, rue Molière, Domaine Universitaire BP 96  
38402 Saint-Martin-d'Hères Cedex,France

Volcanism, both with continuous and eruptive sources, contributes significantly to the Sulfur cycle at high latitudes of  the Southern Hemisphere. Due to the volcanoesí remote locations, measurements of SO2 emissions are sparse and often unreliable ; on the other hand, several  stations on the Antarctic coast monitor volcanic SO2, and remote ultra-violet sensing by the Total Ozone Mapping System enables to track large eruptive plumes.

We used the Laboratoire de Météorologie Dynamique zoomed AGCM, completed by a module for atmospheric sulfur cycle, to test the consistency between the measured SO2 emissions and the observations, both ground based and remotely sensed. Case studies were performed for Mount Erebus (Antarctica), a volcano with continuous fumarolic activity, and for the August 1991 Cerro Hudson (Chile) eruption. We show that such simulations enable both a better quantification of the volcanic emissions and improved regional impact assessments.

Dynamic Modeling of Anthropogenic Emissions
at a Regional Scale in Saxony, Germany

D.Theiss and E. Renner  
Institut fuer Troposphaerenforschung, Permoserstrasse 15  
D-04318  Leipzig

A modular emission inventory of Saxony, a federal state of Germany, was developed. The emissions of the anthropogenic air pollutants SO₂, NO_X, CO, NH₃, NMVOC, CO₂, CH₄, N₂O, PCDD/PCDF, TSP and heavy metal dust are dynamically estimated in types of point, line and area sources. The emission groups are represented by modules of bulk combustion plants, waste disposal sites, large industrial plants and large agricultural farms as point sources, as well as area sources of households, small commercial consumers, industrial and trade branches and total emission of agriculture. In addition there are two modules, one of other biogenic emissions and a mixed type one of transport and other mobile sources. The terms of emission groups are related to the categories of the official statistics in Germany. Therefore data for a topical estimation of emissions are usable. The local resolution of emission values, the relationship of emission values to geographical features or political territories, the inclusion of a high resolved digital street network, and the use of actual data concerning land use, density of population and density of build up areas are realized by means of the Geographic Information Systems ArcInfo and ArcView. The dynamic inventory can be used, alone or in conjunction with an atmospheric dispersion model, to assess trends or scenarios of air quality.

Uncertainties in Tropospheric Chemistry Simulations Related to 
Emission Inventories of Reactive Trace Gases.

K. Tsigaridis and M. Kanakidou  
ECPL, Dept of Chemistry, University of Crete, POBox 1470, 71409 Heraklion, Greece

Ozone and aerosols are known to affect human health, atmospheric chemistry and Earthís climate. Secondary organic aerosols (SOA) and ozone are photochemically formed in the troposphere from the oxidation of their precursors, including volatile organics. Their tropospheric distributions depend on the emissions of their precursors.

To study the sensitivity of the simulated formation of these secondary pollutants on the emissions of their precursors used, a global 3-dimensional chemistry/transport model has been applied. This model takes into account (i) temporal and spatial varying emissions of the precursors of ozone and SOA and (ii) volatile organic compounds chemistry including SOA formation from reactions initiated by O₃, OH and NO₃ radicals. To calculate uncertainties in the global annual formation of organic aerosols and ozone related to the emissions, 3-D simulations have been performed by using different emission inventories. The calculated uncertainties are presented and thoroughly discussed.

Primary Particulate Matter: Its emission and contribution  
to European PM levels.

  Tsyro, S.¹, Berdowski, J.², Visschedijk A.J.H.², Tarrason, L¹.  
1 Norwegian Meteorological Institute, P.O.Box 43, Blindern, N-0313, Oslo, Norway  
² Netherlands Organization for Applied Scientific Research (TNO), P.O. Box 6000, 
NL-2600 JA DEFLT,  the Netherlands

Particulate matter (PM) can either be directly emitted or form in the atmosphere from their gaseous precursors, namely emissions of SOx, NOx, NH3 and VOC. This implies that air concentrations of PM can be decreased by reducing either emissions of primary PM or emissions of gases-precursors of secondary aerosols. As the emissions of the gases have already been reduced as a consequence of the implementation of Protocols under the Convention on Long-Range Transboundary Air Pollution and daughter Directives, and further reduction is expected to be less significant and more costly. Therefore, emission abatement and control of primary PM could be a more efficient measure to decrease the total mass of anthropogenic aerosols in the atmosphere. The question is how significant the contribution of primary PM to the total aerosol levels in Europe is.

Given the recent research which has been undertaken in Europe in various countries on the emission of primary particulate and the need for high quality spatially distributed emission estimates, TNO (Netherlands Organization for Applied Scientific Research), has undertaken a new European emission inventory for the year 1995. This work can be regarded as an update of the previous inventory for the year 1990 and aims to incorporate relevant new information that has become available. The inventory covers the whole of Europe and distinguishes the emission of total suspended particles (TSP), PM₁₀ and PM_2.5. More than 300 source categories are addressed which are believed to cover all relevant sources of primary particulate. The emission estimates have been spatially distributed over a geographical grid for modelling purposes.

The results of this inventory will be discussed with regard to the following aspects:

-Maps of spatially distributed emissions of TSP, PM10 and PM2.5

-Source contributions by size fraction, detailed as well as aggregated overviews

-Country emissions and contributions

-Comparisons with existing national emission inventories for particulate

-Differences with the earlier developed inventory for 1990

Calculations on the dispersion of particulate matter emissions have been made with the Unified EMEP Eulerian model at the horizontal resolution of 50 x 50 km². Previous estimates using TNOís  inventory of PM₁₀ and PM_2.5 emissions in 1990 have shown that primary particles contribute to the total aerosol concentrations with 10-50 % in Western and Central Europe and with 50-80 % in Eastern Europe and Russian Federation. However, concentration levels and geographical distribution of primary particles depends on the national total PM₁₀ and PM_2.5 emissions, as well as their spatial distribution for each source category. This work discusses the results of recent calculations of air concentrations of primary particulate matter in Europe using the updated emissions in 1995 of PM₁₀ and PM_2.5 compiled by TNO, compared to the previous calculations. Changes in the European PM levels and the relative importance of primary PM across Europe resulting from differences in the emission inventories will be analysed.

Inventories of 1890-1990 historical anthropogenic  
trace gas emissions: emission inventory mania gone wild or  
a useful tool for atmospheric studies?

  J.A. van Aardenne, F.J. Dentener, J.G.J. Olivier, C.G.M. Klein Goldewijk and J. Lelieveld  
Environmental Systems Analysis Group, Wageningen University,
  P.O. Box 9101, 6700 HB Wageningen, The Netherlands

In this study a dataset is constructed with anthropogenic emissions of CO₂, CO, CH₄, NMVOC, SO₂, NO_x, N₂O and NH₃ in the period 1890-1990. The emissions have been calculated in 10-year steps based on historical activity statistics and selected emission factors derived from the 1990 data in the Emission Database for Global Atmospheric Research (EDGAR 2.0). The calculated emissions by country were allocated on a 1­⁰ x 1⁰ grid. The database focusses on emissions from energy/industrial and agricultural/waste sources and for completeness historical biomass burning estimates where added using a simple and transparent approach. This dataset can be used in trend studies of tropospheric trace gases  but can also be used for the analysis of historical contributions of regions and countries to environmental problems like the greenhouse effect and acidification.

Wind Direction dependent Differences between Model Calculations and Field Measurements as Indicator for Uncertainty in Emission Inventories.

J. A. van Aardenne, P.J.H. Builtjes, L. Hordijk, C. Kroeze and M.P.J. Pulles  
Environmental Systems Analysis Group, Wageningen University, P.O. Box 9101, 
6700 HB Wageningen, The Netherlands

When comparing results from atmospheric dispersion models with measurements it is difficult to trace the cause of the differences, because every emission inventory, atmospheric dispersion model, and atmospheric measurement contains uncertainties. To assess the uncertainties in an inventory of European SO₂ emission in 1994, we have plotted the calculated SO₂ concentrations from the LOTOS model together with the measured SO₂ concentrations from the EMEP network in wind direction sectors of 30 degrees. We argue that when the wind direction dependent differences at several measurement stations in different countries point to a specific region, the emission estimate for that specific region is the likely cause for the difference between model calculation and observation. In our presentation we will show the results from this analysis and discuss whether this type of uncertainty analysis could provide the emission inventory  community with an addition to the methods used so far for the uncertainty analysis of dispersion models.

Differences in the transboundary fluxes of ozone in Europe 
due to changes in emission distributions

By Vigdis Vestreng  
Norwegian Meteorological Institute (DNMI), Oslo, Norway

This study focuses on how changes in the spatial distribution of emissions alter the transboundary fluxes of ozone both outside Europe and between European countries.

The basis for this study is European emission data reported by the countries to the UNECE¹ under the EMEP² programme. Within this programme, spatial distributions of emissions are a necessary input to chemical transport model calculations. The spatial distributions available to EMEP modelling up to present were considered to be representative for the 1980s. Recently, gridded data of newer vintages has been made available, and this allows sensitivity studies where different spatial distributions can be used. Significant developments and socio-economical changes in Europe after 1990 have influenced the emission levels and the spatial distribution of emissions. These changes have consequences both for the analysis of trends and for the transboundary exchange of air pollution in Europe. The purpose of this work is to quantify and discuss the resulting influence on transboundary fluxes of ozone within and outside this region. It also analyses which factors are most important to determine these fluxes, and evaluates the influence of changes in emission levels and distribution, and the meteorological variability.

Work on emissions in general, and in particular an emission inventory based on submissions from several different institutions as in this case, faces the challenge of verification and quality assurance. The study provides indications of the data quality through comparisons of modelled ozone with observations. It is acknowledged that this approach does not assure the quality and comparability of the inventory, but only qualifies it. Efforts carried out within the CLRTAP³ in order to improve the verification procedures will be presented and discussed as a part of this study.

¹   Untied Nations Economic Commission for Europe

²  Co-operative programme for monitoring and evaluation of the long range transmission of air

    pollutants in Europe

³  Convention on Long Range Transboundary Air Pollution

Tropospheric BrO measured by the GOME

  T. Wagner, J. Hollwedel, M. Wenig, U. Platt  
Institut für Umweltphysik, University of Heidelberg, INF 229, D-69120 Heidelberg, Germany

The temporal and spatial distribution of enhanced boundary layer BrO concentrations in both hemispheres is presented using observations of the Global Ozone Monitoring Experiment (GOME) on board the European research satellite ERS-2. BrO concentrations (up to 50 ppt) are the major cause for catalytic ozone destruction typically observed during polar spring in both hemispheres. While autocatalytic mechanisms are most probably responsible for the release of the observed high concentrations of reactive bromine compounds uncertainties still remain with respect to the primary release mechanisms and whether the autocatalytic reactions are taking place on sea salt aerosol or the surface of sea ice. We find that enhanced boundary layer BrO concentrations correlate very well with ozone depletion events. Enhanced BrO concentrations are always found over or near to areas of frozen salt water (above sea ice or also above the frozen surface of the Caspian sea) confirming the assumption that such conditions are a prerequisite for the autocatalytic release of high BrO concentrations to the troposphere.

Development and evaluation of a canopy environment model for 
regional and global predictions of biogenic VOC emissions.

Sabine WALLENS, Jean-Francois MULLER et Alex GUENTHER.

Biogenic VOC emissions are very sensitive to leaf temperature and light intensity.  Thus an accurate description of canopy microclimate is a necessary component of a biogenic VOC emission model.  Predicting emissions at regional and global scales is challenging because of the need to specify general canopy characteristics for relatively broad vegetation types.  We present an improved canopy model that computes leaf temperature and light intensity for multiple vertical layers in a vegetation canopy.  Leaf temperature is calculated by energy balance and a detailed radiation transfer model is employed. The model is parameterized for seven basic growth forms.

We present fluxes estimates of isoprene, sensible heat, and latent heat for the seven basic growth forms and compare these measurements with relevant above canopy measurements from previous field studies.  In addition, we present a model sensitivity analysis and discuss priorities for future model refinement.

Development and verification of a spatially explicit  
inventory of methane emissions from Australia

Y. P. Wang and S. T. Bentley  
CSIRO Amospheric Research, Private Bag No 1, Aspendale, 3195, Australia

A spatially explicit inventory of methane emissions was compiled for Australia covering livestock, coal mining, gas distribution, fossil fuel combustion, waste and soil uptake. This inventory is constructed on a 75km x 75 km grid using data from local government areas and point source data. The methodology for anthropogenic sources is very closely comparable with that of the Australian National Greenhouse Gas Inventory (NGGI). A regional atmospheric model with a domain covering the continent of Australia and its surrounding oceans is used to estimate the time-varying influences of the spatially dependent emissions from the various regions of Australia upon surface methane concentrations continuously observed at Cape Grim, Tasmania (41 ° S, 147 ° E). These influences are used within a Bayesian synthesis inversion to refine the regional totals of surface emissions and uptake of methane as aggregated from the spatially explicit inventory. Based upon this inverse analysis, the NGGI is considered to overestimate net emissions from south-east Australia and Tasmania in 1997 by about 40%.

Investigation of two Imaging Spectrographs and one CCD array 
for Tomographic DOAS Measurements

  P.H. Xie, F.Finocchi, C.V. Friedeburg, G. Hoenninger, U.Platt, R. Rollenbeck,  
 T. Wagner, and I. Pundt  
Institute for Environmental Physics, University of Heidelberg, Germany

Differential optical absorption spectroscopy (DOAS) has been widely applied for trace gas emission monitoring. The method enables the average concentration measurements along the light path for various species of emission simultaneously (e.g. SO₂, NO_x,, HCHO, HONO, O₃.). With the specific viewing geometry of the novel tomographic DOAS technique, two or three dimensional concentration distributions can be obtained. The method can be used for Airborne Multi-Axis DOAS, ground-based Multi-Axis DOAS- and Long-path DOAS. The concentration fields will be useful for the validation of atmospheric Chemical-Transport-Models.

One possibility for DOAS-Tomography arrangements is the use of a two-dimensional CCD detector together with a so-called imaging spectrograph to measure the light coming from 5-10 light paths at the same time. In this work, the properties of one CCD (Type EEV CCD42-10 from Andor Technology) and two spectrometers (UV: model UFS 200 from Jobin Yvon, visible: model SpectraPro-300i from Acton Research Corporation) have been investigated. The results show that the CCD detector and both spectrographs are appropriate for the design criteria of tomographic DOAS systems. We will be able to measure a maximum number of 12 paths on one CCD.