AT2 Strategy Document

John Burrows and Peter Borrell

 AT2 Steering Group

John Burrows

IUP, University of Bremen, D

Peter Borrell

P&PMB Consultants,
Newcastle-under-Lyme, UK

Martin Dameris

DLR, Oberpfaffenhofen, D

Jean Marie Flaud

LISA, Uni. Paris-XII, F

Maria Kanakidou

University of Crete, GR

Gerrit de Leeuw

TNO, The Hague, NL

Anki Piters

KNMI, de Bilt, NL

Ulrich Platt

IUP, University of Heidelberg, D

Thomas Wagner

IUP, University of Heidelberg, D

 ACCENT Secretariat
Urbino, Italy

May 2005 



 Full Report as PDF file, 1.6 MB 


Table of Contents

  Part 1: An overview of AT2.


1.       Introduction.

2.       ACCENT-TROPOSAT-2 (AT2)

3.       Task Group 1: Algorithm development

4.       Task Group 2: The synergistic use of models and observations.

5.       Task Group 3: Strategies for the validation of tropospheric products from satellites.

6.       E-learning.

7.       Availability of Tropospheric Satellite Data.

8.       AT2 Web Page.

9.       Organisation and Principal Investigators.

9.1     Coordinator and Steering Group.

9.2     Principal Investigators.

Part 2: Contributions from Principal Investigators.

Contributions from Task Group 1.

Contributions from Task Group 2.

Contributions from Task Group 3.

Part 1: An overview of AT2

1.     Introduction

1.1     Atmospheric Environmental problems

The second half of the twentieth century was marked by the realisation that air pollution was not only of local and regional importance but that it was indeed a global phenomenon, with local actions giving rise to the depletion of stratospheric ozone, the inter-continental transport of pollutants and the appearance of pollutants in hitherto pristine parts of the world.

The atmosphere, like the other components of the Earth system, is affected by the continuous increase in human population and activity, which have resulted in a variety of remarkable changes since the industrial revolution of the 19th century. Among these are:

-        the global decrease in stratospheric ozone and the attendant increase in surface ultraviolet radiation, emphasised by the ozone hole appearing over the Antarctic;

-        the occurrence of summer smog over most cities in the world, including the developing countries, and the increased ozone background in the northern troposphere;

-        the increase in greenhouse gases and aerosols in the atmosphere and associated climate change;

-        acid rain and the eutrophication of surface waters and other natural ecosystems by atmospheric deposition;

-        enhanced aerosol and photo-oxidant levels due to biomass burning and other agricultural activity;

-        the increase in fine particles in regions of industrial development and population growth with an attendant reduction in visibility and an increase in human health effects; and

-        the long range transport of air pollution to regions far from the industrial activity.

Many of these changes have socio-economic consequences through adverse effects on human and ecosystem health, on water supply and quality, and on crop growth. A variety of abatement measures have been introduced, or considered, to reduce the effects. However, continued growth in human activities, to expand economies and to alleviate poverty, will ensure that these effects continue to be important for the foreseeable future.

The physical, chemical and biological processes, which determine the composition of the atmosphere, the conditions at the Earth’s surface and its climate, comprise a complex system having many non-linear interactions and much feedback. To assess accurately our current knowledge of the Earth-atmosphere system, detailed global information about the amounts and distributions of key atmospheric constituents and parameters is required.

1.2     Global Remote Sensing using satellite instrumentation

Using remote sensing instrumentation on-board orbiting satellites it is now possible to obtain global information for both the stratosphere and, thanks principally to recent European efforts, for the troposphere. In particular, passive remote sensing in the ultraviolet, visible and infrared spectral regions has provided, for the first time, information about the tropospheric column amounts of trace gases and aerosols, enabling comprehensive global views of the atmosphere to be built up.

A variety of data is now becoming available from the new generation of satellite instruments, such as ESA-ESR2/GOME and ATSR-2, NASA-TERRA/MOPITT and MODIS,
NASA-AQUA/MODIS, ESA-ENVISAT/SCIAMACHY, AATSR, MIPAS and MERIS, NASA-AURA/TES and OMI, EUMETSAT-ESA-Metop/GOME-2, SEVIRI and IASI, and ADEOS/POLDER which allow the determination of two- and three- dimensional distributions and time series of trace constituents, aerosols and pollutants in the troposphere and stratosphere.

The principal research groups have already developed algorithms for retrieving tropospheric information from the complex spectroscopic data streams, and have produced the first global maps of the tropospheric distributions of NO2, O3, SO2, HCHO, SO2, BrO and aerosols.

These activities have demonstrated how such data might be used, either alone or in combination with information from other satellites or from ground stations, to determine:

-        the distributions of NO2, O3 and SO2 both globally, and regionally;

-        regional source strengths;

-        the effects of forest fires and biomass burning;

-        photochemical oxidation in areas of fossil fuel combustion and biomass burning using plumes of formaldehyde as an indicator.

-        the intercontinental transport and transformation of NO2 and aerosol;

-        the spatial distribution of PM2.5 over Europe;

-        the presence of BrO around the Arctic and Antarctic sea-ice in spring (which corresponds to events having nearly zero surface ozone concentrations); and

-        the plumes of noxious gases resulting from volcanic eruptions.

Although it is essential that the data are fully appreciated and used by the scientific and environmental policy communities, the exploitation of satellite data has unfortunately not been the most important driver in the development of space-based instrumentation It is for this reason that the present project, AT2, has been set up. AT2 now has funds to undertake some coordination and its facilitating activities are aimed at achieving the optimal exploitation of data.


Figure 1.   Global tropospheric NO2. A yearly average of observations taken by SCIAMACHY in 2003. The industrial sources are clearly visible. [Image courtesy of H. Kelder, KNMI]

The remainder of this document details the overall aims and organisation of AT2, the detailed aims of the three task groups and the two additional activities, provides lists of the steering group, task group leaders and principal investigators and gives the proposed contributions by the principal investigators


2.1     ACCENT and AT2

ACCENT, Atmospheric Composition Change: a European network, is an EU Network of Excellence with the goal of promoting a common European strategy for research on atmospheric composition sustainability. ACCENT-TROPOSAT-2 (AT2) is an ACCENT integration task which addresses the need for global information about trace atmospheric constituents. AT2 will ensure that the utility of such data is fully appreciated and will facilitate the provision of data to the communities that wish and need to use them. AT2 is building upon on the efforts initiated within TROPOSAT, a subproject of EUROTRAC-2, to facilitate the generation of tropospheric data products, to encourage their use for research in the medium term, and indicate their potential use in the development of environmental policy.

The field is still in its initial phase and many scientific problems remain to be solved, particularly those associated with the retrieval of tropospheric distributions of chemically active species, and the generation of profiles of their concentrations in the atmosphere. These will be facilitated by the use of limb sounding as well as nadir observations on some instruments (SCIAMACHY for example), and the utilisation of the daily observations to generate a continuous picture of concentrations of trace species in space and time. A substantial effort is required in the near future to ensure that the potential of the data is fully realised.

2.2     Objectives of AT2

The objectives for AT2 given in the original ACCENT proposal are as follows.

-        The co-ordination and optimisation of the efforts of European scientists in the retrieval of the data products for tropospheric research from the measurements by instrumentation aboard orbiting satellite platforms.

-        The establishment of the remote sensing scientific team and the development of a strategy to sustain this activity within ACCENT.

-        The provision of added value to the national programmes exploiting remote sensing data within the European Research Area by promotion of discussion and exchange.

-        The definition of the spectroscopic data base needs and initiation of the data base

-        The provision of global tropospheric data products of trace tropospheric constituents (gases, aerosol and clouds) using remote sensing from space.

-        The exploitation of remote sensing data from space-based instrumentation for tropospheric research within the European Research Area.

-        The provision of an interpretative interface to address the existing and developing European and international environmental policy and the role of remote sensing data from space.


The programme is ambitious but it should be possible to attain these during the proposed time scale of ACCENT.

2.3     Deliverables from AT2

The following deliverables were specified in the original ACCENT proposal.

-        A web-based listing of data sets for research and policy support.

-        Research Tools comprising the description of retrieval algorithms

-        Documentation: Workshop reports and a Final report.

-        Scientific work published in the refereed literature.

-        The preparation for a meeting with a resulting peer-reviewed book to document the progress in the activities, later in the project.

2.4     Organisation of AT2

A steering committee was appointed by ACCENT from among the partners and augmented by adding those task group leaders who were not already members.

Principal investigators were invited to apply for membership of AT2 by submitting a short proposal indicating how they would contribute scientifically to AT2. Many of these were active in the former TROPOSAT project; others were new to the group.

The first AT2 workshop organised the project into three groups, with the following tasks.

1.       The development and improvement of algorithms for the retrieval of tropospheric data. Task Group 1 is divided into three groups, specialising in aerosols, infra-red measurements and UV/Visible measurements.

2.       The synergistic use of models and observations to improve our understanding of tropospheric chemistry and dynamics.

3.       The development of validation strategies for tropospheric satellite data products using existing data.

Each task group consists of a number of principal investigators each of whom has agreed to contribute to the work of the integration task.

In addition, two further activities will be undertaken by groups of the principal investigators:

-        to develop an updatable source list of tropospheric data products for use by the wider community; and

-        to contribute to the outreach activities of ACCENT by developing an educational package on Remote Sensing from Space for use in Universities.

The aims of the project and its constituent parts are being pursued through regular workshops and meetings, inter-laboratory exchanges and the publication of regular reports.

The results from AT2 together with information about tropospheric satellite data available and a variety of images derived from satellite data are displayed on the AT2 web page,


3.     Task Group 1: Algorithm development

Thomas Wagner, University of Heidelberg, Task Group Leader

3.1     Summary of main objectives

The aim of task group 1 is to distribute information on data products and algorithms between the different groups involved and also to a wider community of data users.

During recent years a large variety of methods for the analysis of tropospheric data products from satellite sensors have been developed; it was possible to retrieve global maps of trace species (e.g. BrO, NO2, HCHO, SO2, H2O, CO) and aerosol properties from novel instruments like, for example, GOME, ATSR, MERIS, MOPITT, SCIAMACHY, POLDER, MODIS, MISR and SeaWIFFS. Most of the retrieval methods have been developed by individual groups as scientific algorithms and are still subject to further refinement and improvement. A first very successful attempt to coordinate and encourage these activities was carried out within the framework of the EUROTRAC-2 project TROPOSAT An overview on algorithms and data products can be found at the AT2 website, http://troposat.iup.uni-heidelberg.del.

Since the launch of ENVISAT (and recently also AURA), improved and new sensors have become available, in particular including new instruments for aerosol properties and for absorbers in the IR spectral range. Because of these new opportunities and also due to an increased demand on high quality satellite data sets (e.g. for initialisation of and comparison with global CTMs), increased efforts need to be undertaken for the coordination, homogenisation and distribution of satellite algorithms for tropospheric trace gases and aerosols.

After the ending of TROPOSAT a new forum for these activities has been established within the framework of ACCENT. As a response to the increased number of sensors for IR products and aerosol properties task group 1 was subdivided into three parts:

·        Trace gases derived from UV/visible sensors

·        Trace gases derived from IR sensors

·        Aerosol and cloud products

For all of these sub-groups specific challenges exist and special strategies have to be applied. Detailed algorithm work and exchange of specific information can best be carried within specialised sub-groups. For the specific strategies and aims of these sub-groups please refer to the particular sections below.

Besides the activities within the sub-groups, an important part of task group 1 will be the information exchange between the different sub-groups. Such information exchange is not only important because several algorithmic procedures are common between the different sub-groups, but also because algorithms for different products might depend on each other (e.g. tropospheric trace gas data depend on good cloud and aerosol corrections). In addition, and probably most important, such information exchange will stimulate the application and adaptation of strategies and algorithms which have been originally developed within other sub-groups. The communication and transfer of information will be ensured with different strategies.

·        The activities of the different sub-groups will be reported to a general audience at regular meetings.

·        Special workshops on algorithm development including all sub-groups will be held.

·        The reports of sub-group workshops will be distributed.

·        Research visits for scientists between different institutions will be initiated.

·        Products which are available from different sub-groups (e.g. O3, NO2, H2O and aerosol properties) will be compared.

Besides these activities, the long term aims of task group 1 also include the distribution of data products and documentation to the public. As in TROPOSAT, a list of the available data products and the respective PIs will be collected and posted on the project's web-page. The documentation distributed will include individual algorithm descriptions, validation results and reports on workshops and meetings.

3.2     Trace gases; UV/visible

Andreas Richter, University of Bremen, subgroup leader

Retrieval of tropospheric trace species from UV/visible measurements of scattered light started more than two decades ago with analysis of the TOMS data for tropospheric O3 and SO2. As data from the GOME instrument became available in 1995, retrieval of the tropospheric columns of NO2, BrO, H2O, and HCHO also became possible. Since then, several groups have developed independent retrieval algorithms that in most cases are based on the Differential Optical Absorption Spectroscopy (DOAS) method, but differ in the approaches taken for the compensation of stratospheric contributions and the correction for the variation in vertical sensitivity.

Tropospheric products derived from GOME measurements have been used in a multitude of case studies dealing, amongst others, with tropospheric bromine explosions, biomass burning, anthropogenic NOx emissions and long range transport of pollution. They have also been compared intensively with model results and, in an iterative process, lead to substantial improvements in the quality of both model and satellite results. More recently, first attempts have been made to validate tropospheric column products, and this turned out to be a difficult task as averaging over the large ground pixel size and the intrinsic vertical integration are difficult to achieve with other measurement techniques.

After the first phase of demonstration and exploitation of tropospheric measurements from space, the second phase of satellite data use has now started with a focus on new instruments that provide better spatial resolution and coverage, improved algorithms that provide higher accuracy, and more detailed error assessments that provide more quantitative results. Also, the synergistic use of data products from different instruments (fire counts, aerosol information, different trace species) and an integrated retrieval using measurements different instruments (e.g. IR and UV/visible) will be investigated in the future.

Within TG1, data from four UV/visible satellites will be used : GOME-1, SCIAMACHY, OMI, and GOME-2. Several PI contributions deal with different approaches to retrieve tropospheric O3, arguably the most difficult retrieval in particular outside the tropics. A second focus is on NO2 retrieval with BrO, HCHO, SO2, and H2O also covered. Another development is the study of long-term changes in tropospheric composition, which after 8 years of GOME operation and nearly two years of SCIAMACHY measurements now is becoming possible. In view of the planned GOME-2 series on Metop, observation of tropospheric change will probably be one of the most relevant contributions of satellite based observations in the years to come.

The contributing principal investigators to the UV/visible subgroup are listed in section 9 and their individual contributions given in section 10.


3.3     Trace gases; infra-red

Johannes Orphal, LISA, Uni. Paris-XII, F; IR subgroup leader

The infrared region is useful for tropospheric remote-sensing from space because the molecular absorption bands are relatively well separated and the individual lines provide vertical information due to their pressure-dependent profiles and temperature-dependent intensities. In the mid-infrared region (MIR, about 3-15 mm), information on tropospheric traces gases can be obtained using the thermal emission of the surface and lowest atmospheric layers. It is relatively difficult to extract trace gas concentrations within the planetary boundary layer in this spectral region, however, due to the small thermal contrast, the MIR weighting functions peak in the lower troposphere, therefore making this spectral region highly complementary to the UV/visible and near-infrared (NIR, about 0.8-3.0 mm) regions. In the latter region, the photons are mainly due to surface-reflected sunlight. This, together with the fact that in the NIR atmospheric scattering is much weaker than in the UV/visible region, makes the NIR highly interesting for tropospheric observations.

Retrievals of atmospheric species from MIR Nadir spectra have been demonstrated in the past using data from the IMG instrument onboard ADEOS. Limb-spectra (e.g. from MIPAS onboard ENVISAT) have the intrinsic problem of clouds occurring within the line-of-sight, but provide the important advantage of very high vertical resolution in the upper troposphere. Successful NIR instruments are MOPITT (on board EOS-Aqua) and SCIAMACHY (on board ENVISAT) who have clearly demonstrated the sensitivity of the NIR region for important tropospheric species, such as CO and CH4.

In the frame of the ACCENT-AT2 project, several PI contributions focus on the use of recent or near-future infrared satellite instruments in Nadir viewing geometry. Recent Nadir-viewing instruments comprise SCIAMACHY (covering the near-infrared region up to 2.4 microns, where absorptions of CO, CH4, N2O and CO2 can be used), MOPITT (measuring CO and CH4 in the mid- and near-infrared using a gas correlation technique), AIRS (also on board EOS-Aqua, a multi-channel spectrometer in the 3.7-15.4 micron region, capable of observing O3, CO, and many other species), and TES (on board EOS-Aura), a high-resolution Fourier-transform spectrometer covering the entire mid-infrared region, that will provide data for the retrieval of tropospheric O3, CO, CH4 and other trace gases. Within the time-frame of AT2, the launch of MetOp in late 2005 will provide data from another Nadir-viewing instrument (IASI, a Fourier transform spectrometer with medium spectral resolution, covering the range of 2-14 microns). Limb-viewing instruments such as MIPAS onboard ENVISAT and ACE-FTS onboard SCISAT can provide vertical profiles below the tropopause and in the upper troposphere down to 6 km (in cloud free cases).

The main target species are tropospheric columns, and possibly a few tropospheric layers, of O3, CO, CH4, N2O, H2O and CO2 and vertical temperature profiles, but in addition there is a much longer list of other species that will be observed and/or investigated, such as the most important CFCs and HCFs, several VOCs (like H2CO or C2H6), as well as HNO3, PAN and SO2 (the latter mainly in volcanic regions).

The main activities of the different groups within the infrared part of TG-1 will be the development of algorithms to extract tropospheric information from satellite data, with contributions from stratospheric sensors, data assimilation including coupling of the retrieval or retrieved concentrations with chemical transport models, and high-resolution molecular spectroscopy for species where the accuracy is limited by the knowledge of the molecular parameters.

The development of combined retrievals using infrared and UV/visible data will demonstrate the capability of current and future atmospheric sounders for monitoring tropospheric trace gases.

The principal investigators contributing to the infra-red subgroup are listed in section 9 and their individual contributions are given in section 10.

3.4     Aerosol and clouds

Gerrit de Leeuw, TNO, The Hague, NL; aerosol subgroup leader

Satellite remote sensing can be a cost-effective method providing the spatial and temporal resolution to monitor the highly variable aerosol fields on regional to global scales. Until recently it was thought that the retrieval of aerosol properties from satellite data was only possible over dark surfaces, such as the ocean. However, novel satellite sensors allow for the accurate retrieval of aerosol properties over brighter surfaces, in particular over land. Satellite-based instruments that are used for retrieval of aerosol properties were not designed for this purpose. Examples are TOMS, AVHRR, OCTS, ATSR-2, and SeaWIFS. TOMS, initially developed to calculate the total ozone concentration, is especially sensitive to absorbing aerosols both over land and sea. Since 1978, a long time series of the Aerosol Index (AI) over water and over land is available from this instrument. AI is a measure of the wavelength-dependent reduction of Rayleigh scattered radiance by aerosol absorption relative to a pure Rayleigh atmosphere. AVHRR aboard NOAA satellites, designed to determine the sea surface temperature, has been used to retrieve aerosol information over the ocean since 1981. Together these two instruments provide a long time series that, even when not ideally suited for aerosol retrievals (e.g. the AVHRR has calibration problems, TOMS delivers only the AI, i.e. for absorbing aerosols), provides an excellent reference for global aerosol studies. The newer instruments include OCTS (on board ADEOS) and SeaWIFS, both primarily designed for retrieval of ocean properties, which are applied to retrieve aerosol properties over the ocean. ATSR-2 on board ERS-2 was designed to derive sea surface temperature, but it has been successfully used to retrieve aerosol properties over the ocean, and taking advantage of the dual view and the multiple wave bands, also over land. The ATSR series has been continued with AATSR on ENVISAT and is planned for the Earth Watch mission which ensures long-term data availability. POLDER was the first sensor specifically developed for aerosol retrieval over land and ocean by utilizing both polarization and multiple-angle viewing. Two instruments were launched but unfortunately they were operational only during short time periods due to platform problems. Multiple angle viewing is also used on MISR. The instrument that currently delivers most information on aerosol and cloud properties from space is MODIS, which was uniquely designed (wide spectral range, high spatial resolution, and near daily global coverage) to observe and monitor these and other Earth changes.

Within TG-1 data will be used from ATSR-2/AATSR, GOME, SCIAMACHY, MISR and OMI, and possibly from other instruments as well. Activities are underway to expand the aerosol and cloud section with PI contributions from other ACCENT participants active in these areas. Scientific questions that will be addressed are:

·        what aerosol information can be retrieved from instruments that were not designed for this purpose (AOD, Ångström coefficient, speciation, modal sizes, PM2.5, ssa, optical properties, vertical distribution), and with which accuracy?

·        what are the improvements that can be expected from dedicated aerosol instruments:

o     Lidar (CALIPSO)

o     Polarization (POLDER1&2, PARASOL)

o     Multiple angle viewing (MISR, ATSR, CHRIS-PROBA) and

o     Wide spectral range, including IR: e.g. MODIS?

·        what are the improvements from Synergistic use of different instruments (e.g. SYNAER)? and

·        what are the improvements in the retrieval from synergistic use of satellite instruments and chemical transport models (data assimilation)?

Furthermore the question arises what the requirements are as regards the accuracy of the retrieval products for scientific users, for operational users and for policy support.

The principal investigators contributing to the aerosol subgroup are listed in section 9 and their individual contributions given in section 10.

3.5     List of achievables in the first half of the project

A number of the activities listed below cannot be completed within the first half of the project. Nevertheless, substantial progress can be expected.

·        Improvement of algorithms:

o          trace gases (BrO, NO2, HCHO, SO2, H2O, CO), e.g. better cloud and aerosol correction;

o          cloud and aerosol properties from GOME and SCIAMACHY;

o          aerosol products from AATSR, OMI, other sensors;

o          synergistic use of SCIAMACHY and AATSR for aerosol speciation;

o          demonstration of quasi-operational aerosol retrieval algorithm for ATSR-2, by
application to data over Europe for the year 2000;

o          extension of aerosol algorithms to areas with complicated aerosol composition; and

o          demonstration of feasibility to deliver NRT AOD data from AATSR.

·        Development of new algorithms:

o          CO, CO2, CO, O3 from SCIAMACHY;

o          synergistic use of AATSR and MSG-SEVIRI to provide accurate aerosol information  with high resolution in both time and space ;

o          test on feasibility to derive regional PM2.5 directly from AOD data; and

o          PM2.5 maps based on EO data assimilated in CTM.

·        Long term data sets and trends:

o          BrO, NO2, HCHO, SO2 from GOME

·        Synergy and integration:

o          workshop on retrieval methods;

o          workshop on aerosol methods;

o          exchange of information; and

o          provide a list of products and descriptions.

·        Definition of future aims:

o          define realistic long term aims;

o          long term aim for aerosol is to provide reliable, accurate and sustainable products meeting requirements of various user segments: AOD, speciation, PM2.5, ssa, etc.; and

o          define distribution channels for products: GSE PROMOTE


3.6     Encouraging cooperation

As well as the internal communication within task group 1, the external communication with the other task groups (and also outside AT2) is of crucial importance to the success of the project. This is not only because the members of task group 1 must respond to the user needs of the tropospheric data sets derived from satellite sensors. Moreover, the data users must know not only the data product itself but also all necessary information associated with it, e.g. the sensitivity of the measurement under various conditions.

To ensure this information flow in both directions, several activities are foreseen.

·      Besides the data products themselves necessary documentation should also be made available at the web-site.

·      Reports on workshops (whole task group 1, or sub-groups) will be distributed.

·      Overview presentations will be given at project meetings.

·      Feedback from other task groups and projects will be distributed.

     Scientist visits at different institutions will be initiated.

3.7     Performance Indicators

The following indicators can be employed to estimate the progress of AT2 Task Group 1.

·                  Number of 'achievables' delivered and progress toward accomplishment of this list.

·                  Organisation of workshops and distribution of summary reports.

·                  Establishment of a data base for data products and documentation for different algorithms  developed within task group 1.

·                  Improved use of tropospheric data products derived from satellite measurements.

·                  Harmonisation of similar algorithms from different groups.

·                  Number of scientific exchanges between different institutions.

3.8     Task Group 1 principal investigators

The contributing principal investigators to task group 1 are listed in section 9 and their individual contributions are given in section 10.


4.     Task Group 2: The synergistic use of models and observations

Martin Dameris, DLR, Oberpfaffenhofen, Germany,
Task Group Leader

4.1     Summary of main objectives

Task group 2 aims to demonstrate how data products of the troposphere derived from satellite and non-satellite measurements can be evaluated and employed for scientific applications. The synergistic usage of these data together with results of different model systems is planned in order to improve the qualitative and quantitative interpretation and the understanding of dynamical, physical and chemical processes in the troposphere and stratosphere. Satellite data are increasingly suited to supply initialisation, boundary conditions, and test data for chemical-transport models (CTMs) on regional and global scales, and interactively coupled global chemistry-climate models (CCMs). In this sense, the activities of this AT2 task group includes:

·      development of methods for using satellite data from the troposphere as part of model validation strategy;

·      investigation of physical, dynamical, and chemical processes in the troposphere and stratosphere, especially in the upper troposphere / lower stratosphere (UTLS);

·      use the combination of model results, satellite observations, ground based and airborne measurements in a synergistic way to improve our knowledge about individual tropospheric processes, such as, source attribution and impact assessment of gaseous and particulate pollutants; cloud occurrence and the hydrological cycle; and

·      the use of model results to help bridge the gap between point measurements and the satellite view footprint for evaluating satellite retrievals (see example in Figure 2).

4.2     Aim of the research

Within the EUROTRAC-2 project TROPOSAT it was demonstrated that tropospheric data products derived from satellite measurements can be evaluated and employed for scientific applications. Within AT2, scientific investigations will be intensified, in particular the synergistic use of satellite and non-satellite (ground based and airborne data) together with results derived from a hierarchy of different numerical models describing atmospheric processes. Special attention will be given to data products derived from the satellite ENVISAT, in particular those from the instruments MIPAS and SCIAMACHY. ENVISAT is unique due to its combination of a wide range of multiple instruments on one platform. The synergistic use of these scientific data products will help to improve the knowledge about tropospheric and stratospheric dynamical, physical, and chemical processes and their interaction. The studies will address the following scientific points:

evaluation of data products derived from satellite instruments (e.g. GOME, SCIAMACHY, MIPAS, MOPITT); synergistic use of satellite data with ground-based and airborne measurements; scientific interpretation of measurements, among others together with model results;

·      characterisation of the atmosphere and its change in composition in time employing data assimilation techniques:

·      investigation of long-range transport of tropospheric trace gases;

·      quantification of natural and anthropogenic effects on dynamics and chemical composition of the UTLS;

·      explanation of  changes in the UTLS observed during recent decades;

·      improvement of modelling and forecasting of tropospheric chemistry and air quality;

·      evaluation of hydrological cycle and ozone budget in models.

4.3     List of achievables within the first half of the project

·      Evaluation of model results employing data products derived from satellites. Key molecules include NO2 tropospheric columns, ozone, CO, and water vapour.

·      Initial studies of pollution transport mechanisms observable from space.

4.4     List of achievables for the complete project

·      Quality assurance of tropospheric measurements from space.

·      Preparation of climatologies of GOME O3, NO2, BrO, etc. (e.g. tropospheric columns) for the years 1995 to 2004; statistics regarding seasonal and inter-annual variability, description of uncertainties regarding estimates derived from other methods (ground based, aircraft, balloon). It is most interesting to investigate the consistency and relationships between the various gases that are measured simultaneously.

·      Provide consistent data sets (GOME plus MIPAS and SCIAMACHY data products).

·      Monitoring of global and regional changes; continuation of climatologies.

·      The further development of techniques to detect tropospheric composition change (the 'chemical climate') from space.

·      Modelling studies to determine uncertainties of different retrieval algorithms,
e.g. modelling of tropospheric NO2 columns with CTMs.

·      Characterisation of the 'chemical weather' of the troposphere, that is, its short-term change in chemical composition; detailed analysis of episodes; use of satellite data in chemical weather forecasting (e.g. assimilation of GOME/ SCIAMACHY near-real-time O3).

·      Identification and evaluation of pollution transport mechanisms observable from space;

·      Inter-comparison of model results with climatologies derived from GOME data to determine general discrepancies, uncertainties and deficiencies. The investigations will focus on NO2 tropospheric columns, in particular over industrial areas (Europe, north-eastern North America), where major discrepancies are most obvious.

4.5     Encouraging cooperation

There is a need for 'cross-cutting' activities within this task group. First, the synergistic use of observations (here mainly from space) and results derived from model studies will enhance our knowledge with regards to the importance of individual physical, dynamical and chemical processes in the troposphere and stratosphere, as well as the interaction mechanisms (mutual effects) of these processes. This will be the basis for further model developments and improvements. Assessments of possible future developments of tropospheric composition and climate, which are often based on model calculations will be set on a more reliable basis.

Another cross-cutting activity of the task groups is required for inspecting the processing chain with regards to error analysis of the end products. The common effort will identify which processing steps are most critical, which scales are most difficult, which physical and chemical processes are most important, for which kind of application.

Part of the work planned within Task group 2 is a common activity with the ACCENT Modelling task, i.e. the use of data assimilation methods in combining observations and modelling to obtain a complete and continuous as possible picture of the atmosphere.

4.7     Performance Indicators

The following indicators can be employed to estimate the progress of AT2 Task group 2:

·      improved use of tropospheric data products derived from satellite measurements;

·      establishment of a joint data base available inside and outside of AT2;

·      development, harmonisation and joint use of modelling tools;

·      number of exchanges, joint  activities and publications within task group 2; and

·      a number of exchanges carried out outside AT2.

4.8     Task Group 2 principal investigators

The contributing principal investigators to task group 2 are listed in section 9 and their individual contributions given in section 10.


5.     Task Group 3: Strategies for the validation of tropospheric products from satellites

Ankie Piters, KNMI, de Bilt, the Netherlands. Task Group Leader

5.1     Introduction

The tropospheric products retrieved from satellites require a thorough validation in order to assess the usefulness of these products for tropospheric research. The satellite product accuracies should be investigated by comparison with other independent observations of the same products, using ground-based, air-borne and other satellite measurements. Special attention should be paid to methods for the validation of the tropospheric part of molecules that have a substantial abundance in the stratosphere as well (e.g. ozone, NO2), since the retrieval of both the satellite product and the correlative product will be influenced by this stratospheric component.

The relevant current and future atmospheric chemistry satellite missions are: ERS, ENVISAT, EOS-Aura and METOP. These missions deliver a wealth of tropospheric products, many of them generated within the framework of task group 1. Dedicated validation projects exist for every instrument onboard these satellites. However, these dedicated validation projects are typically limited in time and resources, and a lot of the validation is actually being performed by the scientists using the data in a later stage. These individual validation efforts are usually not brought together, and several independent studies often show inconsistencies or conclusions which are not comparable on data accuracy. In addition some specific tropospheric products are developed after the dedicated validation projects, or as non-official satellite products by scientific institutes rather than by the responsible space agencies. Most of these products undergo a first limited validation by the product developers themselves. This situation is not ideal, therefore this task group has the following long-term objectives.

5.2     Long-term objectives (general)

The aim of task group 3 (TG3) is to develop an international open network of collaborating validation scientists, in which the quality of the available tropospheric satellite products is thoroughly assessed. More specifically, the general long-term objectives are:

·      to encourage validation of tropospheric products retrieved from current and future atmospheric chemistry satellite missions from ESA, NASA and EUMETSAT, like ERS, ENVISAT, EOS-Aura, and METOP;

·      to collect validation results of tropospheric products;

·      to communicate knowledge about product quality;

·      to encourage interaction with retrieval groups, in particular with scientists from task group 1, in order to improve the tropospheric products more efficiently;

·      to integrate the activities performed within this task group into existing large-scale validation projects, in particular those for SCIAMACHY, OMI, and GOME-2 (PI contributions: Kroon and Piters)

·      to encourage activities in special regions in new EU member states;

·      to encourage interaction between data providers, validators and modellers;

·      to exploit the use of existing networks (e.g. NDSC);

·      to encourage the participation in instrument intercomparison and validation campaigns; and

·      to encourage mobility of instruments and scientists.

5.3     Long-term objectives (scientific)

TG3 involves scientists working individually on the validation of one or a number of tropospheric satellite products, most of them on national funding. Therefore the content of the research done in this task group very much depends on what individual scientists can do within their own national science programmes. The long-term scientific objectives, listed below, serve as a guideline for possible future nationally funded research, and more importantly for possible future cooperation projects within the task group.

The general scientific aim of TG3 is to assess the quality of tropospheric satellite products, more specifically the usability of those products in tropospheric research or monitoring of pollution or climate change. More specific long-term objectives are listed below.

1)      Develop validation strategies to account for differences in retrieval methods and differences in representation of the satellite product and the correlative instrument product or model, for example differences in spatial resolution and sampling (see Figure 3), averaging kernels, and viewing geometries (PI contributions: Galle and Mellqvist)

2)      Develop techniques to assess the accuracy of a product from a combination of information from different validation methods, such as comparisons to ground-based, models, and satellites (PI contributions: Gloudemans and Van der Broek)


Figure 3.   Illustration of the difference in representation of a satellite measurement, which is the NO2 column over a large rectangle, and the ground measurement, which gives the NO2 column over one specific location (picture provided by A. Petritoli, ISAC).

3)      Collect, analyse, and improve correlative measurements in special areas of interest, such as megacities and volcanoes (PI contributions: Galle and Mellqvist)

4)      Collect, analyse, improve, and make available correlative measurements from long-term continuous measurement stations and networks (PI contributions: Mahieu, Notholt and Sussmann)

5)      Perform comparisons between tropospheric satellite products and independent correlative measurements (PI contributions: Galle/Mellqvist, Gloudemans, Van der Broek, Monks/Leigh and Petritoli)

6)      Perform comparisons between tropospheric satellite products and models (PI contributions: Gloudemans and Van der Broek)

7)      Assess the usability of tropospheric satellite products for specific tropospheric research areas, e.g. radiation budget, long-term evolution of chemical composition, air quality (PI contributions: Mahieu, Monks/Leigh, Orsolini and Petritoli)

8)      Establish operational ground-truthing stations especially suited for validation of tropospheric satellite products (PI contributions: Sussmann and Ubelis)

5.4     Encouraging cooperation

·      Participation in this Task group will be encouraged by inviting as many as possible colleagues in the field of validation to actively participate (which means, in first instance, to write a PI contribution). A special effort will be made in encouraging validators of cloud and aerosol products.

·      The participating projects will be fully integrated in the existing large-scale validation projects for SCIAMACHY and OMI. This means that the scientists participating will have early access to data, will be informed on instrument and data issues, will be invited to validation meetings, and can participate in joint validation publications.

·      A special campaign will be defined for comparison of SCIAMACHY CO total columns (from TG1) with collocated ground-based measurements, where cooperation between retrieval groups, instrument teams, and validation scientists is exploited.

·      Validation scientists will be brought together at least twice a year for dedicated workshops on the validation of tropospheric products from satellites, mostly combined with other interesting workshops for validation, provided that funding is available to cover the corresponding expenses.

5.5     Specific Comparisons envisaged for the first half of the project

The new satellite measurements of tropospheric species and the algorithms used for their retrieval require an extensive validation. Various methods are envisaged to achieve this. These include intercomparisons of independent satellite measurements, comparisons of satellite measurements with ground-based measurements and/or balloon experiments, comparisons of satellite measurements with results from chemistry/transport model calculations. The products to be validated during the first 18 months are tropospheric O3, NO2, SO2, CO, CH4, and CO2 from GOME, SCIAMACHY, MOPITT, and OMI. In addition to these products, TG3 also generates correlative measurements of (partial) columns of BrO, H2O, N2O, C2H6, HCFC-22 (+ many other FTIR species), and profiles of O3, H2O, T/p, and aerosol extinction.

O3, NO2

o       Report-Assessment on O3 observations during the European heat wave of August 2003.

o       Chemiluminescence monitors in 8 separate locations in Leicester provide hourly data on boundary layer ozone and NO2 over an area of several kilometres. The use of such a network for the validation of tropospheric measurements from GOME, OMI and SCIAMACHY will be explored.

o       Data analysis from ground-based stations to retrieve NO2 column in the PBL (by means of DOAS observations) and in situ concentration at representative sites.

o       Validation studies at the Zugspitze (2964 m) FTIR: SCIAMACHY total columns of O3 and NO2.

o       Training of personnel from Latvia for MAXDOAS validation.

o       Planning and execution of a joint validation campaign in Latvia using existing mobile validation stations.

o       General concept and design of stationary MAXDOAS validation station in Latvia.

CO, CH4, CO2

For these molecules only a few ground-based stations are available, most of which are at elevated locations. A detailed validation of satellite measurements of these species therefore requires not only comparisons with ground-based measurements, but also intercomparisons of independent satellite measurements and comparisons with model calculations. In addition, comparisons with measurements from a mobile solar FTIR instrument allow for good collocations in both time and space.  To achieve this, the following comparisons/validation studies are proposed for the first 18 months.

o       Comparisons of SCIAMACHY CO and CH4 with several NDSC FTIR ground-stations. In particular, Harestua, Jungfraujoch, Zugspitze (2964 m) and the new Garmisch (734 m) station. The build-up and first measurements of the latter station will allow a characterization of the boundary layer and free troposphere and thereby complement measurements at the nearby Zugspitze station.

o       Comparison of SCIAMACHY and MOPITT CO with mobile solar FTIR measurements from the ground at selected places: Mexico City, April 2003 and Milan/Po Valley, Sept. 2003 and Aug. 2004. This allows for better collocation in both time and space than fixed FTIR stations.

o       Intercomparison of SCIAMACHY and MOPITT CO measurements, allowing validation to be done on a global scale, instead of at selected sites.

o       Comparison of SCIAMACHY CO/CH4/CO2 and MOPITT CO with atmospheric chemistry models, allowing validation to be done on a global scale also for species for which no additional satellite measurements yet exist.


o       Comparison between SCIAMACHY SO2 and MAX-DOAS measurements around a few selected volcanoes (Etna: Sept. 2004, Nyirangongo, Kongo: Jan.-June, 2004).


o       Start of operation and first measurements of new Zugspitze/Schneefernerhaus (2600 m) water vapour lidar: Expected profiles up to 12 km altitude.


o       Routine measurements with the Garmisch stratospheric aerosol lidar made available for the validation of the AURA mission.

o       Upgrade of Garmisch aerosol lidar with the 'high-resolution' option: thereby independent retrieval of extinction profiles. Airborne measurements of aerosol (sahara dust) above Lampedusa for satellite validation.

o       Start of permanent in-situ measurements at the Zugspitze station of aerosol optical parameters, size distributions, scattering und backscatter coefficients, and wavelength dependent absorption.

5.6     Performance Indicators

TG3 expects to have a significant contribution, especially to the following performance indicators from the Description of Work.

1.       Efficiency of the different levels of management within the NoE in running NoE activities. Integration within NoE: co-ordination and activities of scientific community at large. Establishment of action plan. Efficient solution of problems. Regularly updated activity planning.

2.       A number of multilateral research activities within the subject of the NoE planned and executed by the Partners and with others outside the NoE. A number of scientific publications within the Network.

8.       Establishment of joint databases available also outside the NoE, establishment of a data management procedure within the NoE. Improved use of satellite data.

The planning of co-ordinated multilateral activities, such as dedicated experimental campaigns will contribute to performance indicators 1 and 2, and the definition of a quality flag or the development of a method for a better exploitation of satellite data (for example involving the environmental services) will contribute to performance indicator 8.

5.7     Task Group 3 principal investigators

The contributing principal investigators to task group 3 are listed in section 9 and their individual contributions given in section 10.


6.     E-learning

Maria Kanakidou, University of Heraklion, Crete.

AT2 set up an e-learning group at its first workshop. The members are Joern Bleck-Neuhaus (Bremen), Annette Ladstätter-Weißenmeyer (Bremen), Gerrit de Leeuw, Rob Mackenzie, Thomas Wagner and Peter Borrell.

A pilot e-learning project will focus on the understanding of the central role of NO2 in atmospheric chemistry and the retrieval and use of NO2 observations from space.

The e-learning course aims to introduce the student to the topic of atmospheric chemistry and spectroscopy and provide references and links for further reading. The course will continue with a step-by-step use of the satellite observations to calculate NO2 distributions and interpret them. This itinerary will be enriched with examples and exercises that will allow the student to evaluate of his or her understanding and comprehension at each step.

The provisional outline of NO2 module is as follows.

a.       Introduction: atmospheric chemistry and physics of the troposphere and stratosphere (including examples for case studies). This part will be NO2 oriented (chemistry/sources). At a latter stage of this effort in ACCENT, a complete introduction to atmospheric chemistry will be built - perhaps in collaboration with another ACCENT activity.

b.       Remote sensing measurements to observe NO2:

-        why satellite measurements?

-        why this species (chemical importance)?

-        introduction in spectroscopy (different wavelength regions = different instruments) and in addition the DOAS method for UV/visible spectroscopy.

c.       DOAS: satellite – ground based measurements (technique), temporal and spatial resolution of the instruments.

d.       Retrieval methods to obtain column amounts of the different trace gases from the measured spectra.

e.       Comparison of satellite based measurements with ground based and in-situ air borne data for total column and tropospheric column amount of NO2.

i.        what can we learn from? What precautions are to be taken?

ii.       Examples.

f.        Comparison of satellite derived total column and tropospheric column amount NO2 with model simulations.

i.        what can we learn from? What precautions are to be taken (references).

ii.       Examples.

It is expected that AT2 will provide the scientific material and will then seek professional help to produce the e-learning module.



7.     Availability of Tropospheric Satellite Data

An aim of AT2 is the provision of global tropospheric data products of trace tropospheric constituents (gases, aerosol and clouds) using remote sensing from space. Although some data for O3 is available from the space agencies, data for the other chemical species is difficult to produce. It generally requires an expert assessment and processing of the lower level data, produced by the satellite instrumentation, sometimes examining data pixel by pixel to retrieve the total column densities of the desired chemical species.

Thus the majority of data of interest are research products, and are made available to interested individuals on a person to person basis with a resulting co-authorship of 'producer' and 'consumer' on scientific papers produced from the work. Furthermore, the results can change as algorithms are improved and the details of the instrumentation and the necessary corrections for atmospheric conditions are better understood.

The situation is improving and it is expected that, during the life of the project, substantial steps will have been made towards implementing automatic retrieval algorithms for the more common species.

The first step to be taken in AT2 will be to list the data available on the web site, to alert the community to its existence and to invite collaboration to exploit it. This will follow on from the start made in TROPOSAT:
Authors will be contacted and asked to provide the detailed information about the data to assist prospective users in choosing among what is available. It is also intended to set up a web-based questionnaire for this purpose to facilitate regular updating.

Consideration may be given later in the project to establishing a data base for data itself, although this may require too many resources to be worthwhile.


8.     AT2 Web Page

A web page for AT2 has been developed from the earlier TROPOSAT page and is situated at the University of Heidelberg: It is used to disseminate information about AT2 and will be used to collect and display information about the satellite data available for the troposphere; see section 7. Some AT2 pages are also available on the ACCENT web site itself, These are used to provide preliminary information together with notices of meetings and events. The two are connected by click-through links.


 Return to the top of the page
 AT2 Home Page