The GOME Water Vapor Climatology Project (GWC) V1.0


Create your own plot:Map
Time Serie
 
Download of NetCDF data (please sent a request to the authors).
 
PLEASE NOTE -
- to contact the authors if you want to use the results for publication! -
 
Updated: 24/06/2007
Data availability: 08/1995 to 04/2003
Data version: V1.3
Author: R. Lang
Contact: ruediger.lang *at* eumetsat.int
Co-Authors: S. Casadio, A. N. Maurellis
 J. Steinkamp, M. G. Lawrence

General Product Description


The GOME Water Vapor Climatology (GWC) project V1.0 provides water vapor (WV) total column amounts (WVC) derived from the Global Ozone Monitoring Experiment (GOME) onboard ESA's ERS-2 satellite platform. WVC are derived using a technique called IGAM developed by Casadio et al. (2000) at the Institute for Geophysics Astrophysics and Metereology, University of Graz. The technique is based on Beer's law which assumes a relationship between the natural logarithm of the ratio of the measured intensity of two GOME earthshine detector pixels to the slant column of the absorber of interest. GOME-IGAM currently employs WV absorption signature pairs around 730 nm for retrieving WV slant column amounts and O2 absorption signatures around 760 nm for retrieving the amount of traversed Air Mass (Air Mass Factor (AMF) retrieval). The pairs are chosen such that the differential contributions of other absorbers, as well as that of the solar irradiance spectrum, are negligible and divide out. The AMF is used to convert slant column to vertical column amounts, assuming that the AMF is the same for oxygen as for water vapor. The retrieval technique does not account for the well-known non-linear 'saturation' effects of line-absorbers like WV and therefore requires some sort of independent adjustment for retrievals involving large optical depths (i.e. high WVC amounts) (Lang et al., 2002, Lang et al., 2004, Lang et al., 2007). The independent adjustment is performed using the accurate but considerably slower Spectral Sampling Parameterization (SSP) technique developed by Maurellis et al., 2000a, and Lang et al., 2003 for a limited GOME measurement set. Both methods rely on pressure and temperature profiles currently taken from a climatology. Pixel averaged cloud top pressure (CTP) is derived for each pixel from the O2 AMFs. An empirical conversion from CTP to cloud fraction (CF) is performed by employing a representative set of CTP values derived from the Fast Retrieval Scheme for Cloud from the Oxygen A-band (FRESCO) Algorithm (Koelemeijer et al., 2001) and relating it to FRESCO-derived CF. All GOME-IGAM WVC retrievals with a CF of more than 15% are rejected.

Gome WV data retrievals have consistently demonstrated over the last five years that they can deliver good results over all surface types and in cloud-free situations (Lang et al. 2003, Lang et al. 2004). GOME data is not assimilated in either ERA or NRA reanalysis data. GOME data also provides a data record starting in August 1995 and running until 2003. Combining both aspects makes the GWC database v1.0 a unique resource for model evaluation.

The database provides monthly averaged WVC values for the whole globe. For regions with unsatisfactory statistics (e.g., frequent cloudiness along the ITCZ) or large retrieval errors (e.g., high Solar Zenith Angle (SZA) corresponding to high latitudes as well as instrument failure or calibration windows) values are either set to zero (field1) or values of the Special Sensor Microwave Imager (SSM/I) are used instead (field2). The absolute error on the monhthly averaged pixel as a result of retrieval error and statistics is also delivered (field3).

For a detailed description of the retrieval method and the retrieval performance we refer to Casadio et al 2000, and Lang et al, 2004. The publications can be downloaded in pdf format via the URLs given in the following section.

A detailed evaluation of the database emplying comparisons to radiosonde, re-analysis and microwave sensor data has been recently accepted for publication in JGR-Atmopsheres.

Please read carefully the copyright and responsibility section provided below before downloading or using the data. Before publishing of parts of the data please notifiy the authors via the given contact details. Usage of the data should be acknowledged.


Basic literature and important information for data users


Casadio (2000) (PDF, 1.2 MB)
Lang (2004) (PDF, 480 KB)
Technical summary Data format Known problems
Acknowledgements and usage of data Scientific background Important References

Contact Information


Page last updated: Mon Feb 14 13:10:18 MET 2005