The PROMOTE 3D ozone record service: Overview and first evaluation of stratospheric ozone reanalyzes based on satellite observations between 1992 and 2004

Data assimilation of atmospheric constituents is becoming more and more important as the number and quality of satellite observations increases. New assimilation methods allow us to derive continuous four-dimensional depictions of the atmos-phere’s chemical state. Within the European ESA/GMES project PROMOTE three CTM-based assimilation systems have been applied to derive long-term records of stratospheric ozone and related species from UARS/MLS, ERS2-GOME/NNORSY and ENVISAT/MIPAS ozone observations covering a time period from 1992 to 2004. Using ROSE/DLR (sequential), SACADA and BASCOE (both 4D-VAR) daily chemi-cal analyzes and error statistics have been generated. We present first validation re-sults using independent ground-based and satellite observations. With respect to cli-mate research the benefit of data assimilation is sometimes called into question: e.g., forcing the model with observations may lead to inconsistencies in derived parame-ters, e.g. chemical ozone depletion rates. By means of cross-comparison we there-fore discuss the possible model influence on final analysis results. We show that in most cases results are highly consistent showing errors well within expected error bars. However, under certain conditions and/or if observational coverage is poor, e.g., at the polar vortex edge during spring, specific model limitations have to be ad-dressed and taken into account. Concerning stratospheric ozone, we discuss the ap-plicability and potential of the PROMOTE data record for the SPARC community