Latest Newsletter





The GlobVapour project has been successfully ended on the 31st January 2012 .

It successfully passed its review




Project Manager:

Dr. Marc Schröder (DWD)


Technical Officer:

Dr. Bojan R. Bojkov (ESA)

User Help Desk

Data Request & Technical Support:

Dr. Marc Schröder (DWD)

News Archive


September 2011

Successful ESA DUE GlobVapour User Consultation Meeting in Oslo

by Dr. Marc Schröder (ESA DUE GlobVapour Project Manager)


The ESA DUE GlobVapour project carried out its second User Consultation Meeting in Oslo, Norway on 8 September 2011, in parallel to the EUMETSAT Meteorological Satellite Conference. EUMETSAT kindly offered facilities and beverages, what is acknowledged by the project.


The major objective of the meeting was to gather feedback on the GlobVapour project's products and approaches related to the retrieval, validation and assessment of atmospheric water vapour. The meeting offered a great opportunity for intense discussions as it brings together the producers and users of water vapour data sets.


In total, 22 scientists from various communities attended the meeting. The project received valuable feedback from the user community, with presentations of e.g. Ralf Bennartz (University of Wisconsin), Christine Rademacher (Max Planck Institute for Meteorology), Thomas August (EUMETSAT) and Estelle Obligis (Collecte Localisation Satellites).


Together with the outcome from stimulating discussions the presentations led to refinements of the project's activities and might lead to a continuation of GlobVapour activities beyond the projects end. Topics of major decisions were to initiate a dialogue with the Coupled Model Intercomparison Project Phase 5 (CMIP5) community with the aim to provide the GlobVapour SSM/I+MERIS data set for CMIP5 evaluation and to get visibility in the 5th Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR5). Also, a guidebook containing methods, lessons learned and results from the IASI assessment would be valuable for the community and in particular for the water vapour and temperature profile assessment of the GEWEX Radiation Panel (GRP).


All participant and in particular the users are encouraged to continuously provide the project with feedback and results from first applications. The final meeting of the project is planned for 17+18 January 2012 at ESRIN, Frascati, Italy.



Participants of the GlobVapour User Consultation Meeting 2 in the Oslo Kongressenter Folkets Hus in Norway.




please click on figure to enlarge 


July 2011

Test Products online & Announcement of ESA DUE GlobVapour User Consultation Meeting

by Marc Schröder


Dear GlobVapour Community and Friends,


the European Space Agency (ESA) Data User Element (DUE) project series aims at bridging the gap between research projects and the sustainable provision of Earth Observation (EO) climate data products at an information level that fully responds to the operational needs of user communities. The ultimate objective of GlobVapour aims at providing validated multi-annual global water vapour data sets with improved error estimates and spatio-temporal sampling.


In July 2011, the test products (2006-2008) have been released by the project and are freely available through this webpage. Please register here.


On the afternoon of the 8th September 2011, the second ESA DUE GlobVapour User Consultation Meeting will take place in Oslo during the EUMETSAT Meteorological Satellite Conference. The workshop will be carried out in parallel to the EUMETSAT Users Workshop, and EUMETSAT kindly offered a room at the Kongresscenter. The major objective of the meeting is to gather feedback on the GlobVapour project's products and approaches related to the retrieval, validation and assessment of atmospheric water vapour. In general, the meeting is a great opportunity for intense discussions as it brings together the producers and users of water vapour data sets. The agenda has been finalised.


The meeting room offers limited seats only. Therefore, you are invited to submit an email to Nadine Schneider if you plan to attend the meeting. Please provide relevant personal information, that is, your name and affiliation and potentially the title of your contribution. Abstract submission is not required. Participants are expected to cover their own expenses to attend the meeting.


The attendance of the User Consultation Meeting does not require a registration at the EUMETSAT Users Conference. But please make sure to be registered for the User Consultation Meeting until the 12th of August 2011 so that access to the Kongresscenter can be organised in time.

Independently from the meeting and if not a user so far, you are encouraged to also register as a user on the web page.


We are looking forward to your active participation in what we are certain will be a stimulating and productive meeting. Your feedback is highly welcome and will impact the progress of the project.


Kind regards


Marc Schröder - Project Manager


 Agenda of User Consultation Meeting 2 


January 2011

Retrieval of total precipitable water from GOME-1, SCIAMACHY and GOME-2

by Thomas Wagner (MPI-C), Steffen Beirle (MPI-C), Diego Loyola (DLR), Kornelia Mies (MPI-C), Sander Slijkhuis (DLR)


UV/VIS satellite instruments like GOME-1, SCIAMACHY and GOME-2 observe spectra of backscattered sunlight. From such spectra the atmospheric absorptions of water vapor and O2 in the red spectral range are analysed using the DOAS technique. In our algorithm, from the O2 absorption the length of the atmospheric absorption path is estimated, which is used for the conversion of the measured H2O absorption into the total column precipitable water (vertical column density). Because of the different profile shapes of water vapor and oxygen, additional corrections are performed, depending on the viewing angle, solar zenith angle and the surface albedo. For that purpose pre-calculated look-up tables are used. It should be noted that the correction for surface albedo effects was only recently introduced (e.g. EUMETSAT, 2010) and was not part of the original version of our analysis (Wagner et al., 2006, 2007). In contrast to other satellite retrievals of water vapor, our analysis can be applied over both oceans and continents. Since clouds shield part of the atmospheric water vapor profile, the accuracy of the retrieved total precipitable water depends on the cloud cover, and thus only mainly cloud-free observations are used. The selection is done using a threshold for the O2 absorption. In Fig. 1 the global distribution of the atmospheric water vapor column measured from GOME-2 for the year 2009 is shown.


Our water vapor product was developed with the aim to study time series and temporal trends of the atmospheric water vapor content with high accuracy. Thus no a-priori information or information from external data sets is included in our algorithm. On the other hand, this approach has also some drawbacks. In particular, the errors for individual measurements can become rather large (up to > 100%), especially in the presence of partial cloud cover, and if the true atmospheric water vapor profile deviates strongly from the assumed profile. For mainly cloud-free conditions, the mean relative error (compared to radio sonde observations) is typically < 20%. Another limitation is that the spatial resolution of individual measurements is rather coarse (e.g. 40 x 80 km² for GOME-2).


So far total precipitable water for the period 1996 to 2010 was analysed from all three sensors. Fortunately, between the individual time series, large overlap periods exist, which allow the determination of trends with small uncertainties. From the individual satellite observations global monthly mean maps are calculated with a spatial resolution of 0.5° x 0.5° degree.


In Fig. 2 time series of monthly mean H2O columns for two selected regions are shown. Over Europe (40° - 60°N, 10°W - 30°E), as expected, large seasonal variations are found. After removal of the seasonal cycle (lower part), a positive trend of about 10% from 1996 - 2010 was found. Over the eastern central Pacific (10°S - 10°N, 135° - 85°W), much smaller seasonal variations were found. The de-seasonalised data (lower part) show almost no trend, but are correlated with the ENSO index (taken from



-  EUMETSAT, Algorithm Theoretical Basis Document for GOME-2 Total Column Products of Ozone, NO2, tropospheric NO2, BrO, SO2, H2O, HCHO, OClO and Cloud Properties (GDP 4.4 for O3M-SAF OTO and NTO,, 2010.

-  Wagner T., S. Beirle, M. Grzegorski, U. Platt, Global Trends (1996–2003) of Total Column Precipitable Water Observed by Global Ozone Monitoring Experiment (GOME) on ERS-2 and Their Relation to Near-surface Temperature, J. Geophys. Res., 111, D12102, doi:10.1029/2005JD006523, 2006.

-  Wagner, T., S. Beirle, M. Grzegorski and U. Platt, Satellite Monitoring of Different Vegetation Types by Differential Optical Absorption Spectroscopy (DOAS) in the Red Spectral Range, Atmos. Chem. Phys., 7, 69-79, 2007.

Figure 1. Global mean water vapor column for 2009 derived from GOME-2 measurements.


Figure 2. Time series of the global water vapor column for two selected regions (top: Europe, bottom: eastern central Pacific). The time series of monthly anomalies over Europe (gray line) is fitted by a linear trend (magenta line); Over the east central Pacific the time series of monthly anomalies (gray line) are compared to the ENSO index (magenta line). The small figure shows the correlation between the monthly anomalies of the water vapor column and the ENSO index.


please click on figure to enlarge 


September 2010

First prototype month of the combined SSM/I and MERIS product was determined

by Rasmus Lindstrot (FUB) and Nadine Schneider (DWD)


Right in time before the next consortium meeting in Mainz in November, the first prototype data set of the combined SSM/I-MERIS product was processed within the GlobVapour project (see figure 1 for the daily composite). When completed, this product will be a global, multi-annual, total column water vapour (TCWV) time series from a combination of MERIS (Medium Resolution Imaging Spectrometer) onboard ENVISAT and SSM/I (Special Sensor Microwave Imager) onboard the DMSP satellites. The first prototype data set comprises the monthly mean of July 2007.


Since the MERIS retrieval algorithm is based on measurements in the near infrared region, problems occur where the ocean is dark (very little reflection). Hence it can only be used for water vapour retrievals above land surfaces during daytime. On the contrary, the SSM/I retrieval uses passive microwave measurements, allowing accurate water vapour retrievals over ocean, where the emissivity of the surface is known accurately, independent of the time of day. Due to these retrieval characteristics, the final product combines the excellent performances of MERIS and SSM/I water vapour retrievals over land and ocean, respectively.


In order to assess and correct the bias between both data sets, the MERIS algorithm is applied to sun glint cases over ocean. Here the direct reflection of the incoming solar radiation at the water surface can be used for a TCWV retrieval with MERIS. The monthly mean for the combined product is pictured in figure 2 for July 2007. In this case, the water vapour values derived from SSM/I measurements were 1.5 kg/m2 higher than those derived from MERIS observations, the estimated retrieval uncertainty of the combined product is displayed in figure 3.


Due to its relatively large footprints and the significant degradation of the retrieval quality when land surface contaminates these footprint, SSM/I can not be used for the TCWV retrieval in coastal regions. At these grid points, MERIS measurements were used to fill the gaps, accepting a higher uncertainty of the final product. Similar applies to sea ice where MERIS will be used.


The spatial resolution of the daily composites and monthly means is 0.05° above land an 0.5° above ocean.


Within the next weeks, the SSM/I and MERIS time series will be validated against ground based, airborne and other satellite based TCWV estimates.

Figure 1. Daily composite of the combined product of total columnar water vapour (TCWV) from SSM/I (DMSP-13) and MERIS for the 15th July 2007.


Figure 2. Monthly mean for July 2007of the combined SSM/I+MERIS product of total columnar water vapour (TCWV).


Figure 3. Monthly mean of the estimated retrieval uncertainty of the combined SSMI/I+MERIS product. The accuracy above land was set constantly to 1.5 kg/m2.

please click on figure to enlarge 

May 2010

Initial results of the application of a 1D-Var system to SSM/I measurements

by Martin Stengel (DWD)

As a first mile stone of the GlobVapour project, a one-dimensional variational 1D-Var retrieval (NWP-SAF User’s Guide, 2001) system has be implemented and initially been tested using the Special Sensor Microwave/Imager (SSM/I) on-board the Defense Meteorological Satellite Program (DMSP) (find details at:, and
In the current set-up, the 1D-Var system uses data extracted from forecasts of the ERA-Interim archive (Berrisford et al., 2009) as a first guess of the atmospheric state which is the iteratively modified to fit the co-located measurements.

Exemplarily, pixel-based total columnar water vapour (TCWV) values, calculated from the analysed atmospheric state, is shown in Figure 1. for a sample located over the Atlantic Ocean. It can be seen that despite the similarities between ERA-Interim and the retrieval results some regions can be identified with significant differences. These differences raise from the additional information content taken from SSM/I observations. On a global and long-term scale these results will provide a highly accurate water vapour analysis.

Another advantage of the system used is the implicit provision of a measure of the retrieval uncertainty, which is separately available at each satellite footprint, and subsequently for each grid box of the global, long-term water vapour analysis. For the case mentioned above the retrieval uncertainty estimate is shown in Figure 2. The retrieval uncertainty is generally lower than 10% of the retrieved absolute TCWV value in each footprint. The highest uncertainty values are located in regions with strong convective activity, which can be seen in the geostationary RGB image (Figure 2, left).

The next step will be to validate instantaneous as well as Level 3 data (e.g. monthly means) with high quality reference observation to estimate the quality and possible systematic errors. This will be followed by a checking the 1D-Var system in terms of consistency and stability for the production of global long-term data sets.


- Berrisford, P., Dee, D., Fielding, K., Fuentes, M., Kållberg, P., Kobayashi, S., and Uppala A., 2009, The ERA-Interim archive – Version 1.0, ERA reports. Available here.
- NWP-SAF User’s Guide, 2001: Standalone 1D-var scheme for the SSM/I, SSMIS and AMSU. G. Deblonde, NWPSAF-MO-UD-001 Version 1.0, 22 August 2001.

Figure 1. Total column water vapour as calculated from ERA-Interim (left) data and as derived with the SSM/I 1D-Var scheme (right) for DMSP-13 over the Atlantic Ocean.

Figure 2. Left: Pseudo-RGB Image derived from the Spinning Enhanced Visible and InfraRed Imager (SEVIRI) measurements. Right: Estimate measure of 1D-Var retrieval uncertainty for the same scene as in Figure 1.

Contact • Links • Imprint
Project Manager: Marc Schröder (DWD)   ·   Technical Officer: Bojan R. Bojkov (ESA)   ·   The web site is hosted at Brockmann Consult