Comparison of H2O retrievals from GOME and GOME-2

Abstract

Climatologically relevant time series of total column water vapour may be derived from the family of GOME instruments. We compare the performance of two different algorithms for the retrieval of H2O total columns from the GOME and GOME-2 instruments: the AMC-DOAS method from University of Bremen, and a classical DOAS method from University of Heidelberg / M.P.I. Mainz. The data from these two algorithms are compared to each other, to SSM/I data, and to data of the global radiosonde network. We find a tight correlation between the retrieved columns from both algorithms. There is an offset difference of 0.2-0.3 g/cm2 and a difference in slope of 7-9%, depending on selection criteria. In the comparison to external data, the algorithms show slightly different behaviour, but overall we judge their performance to be on a comparable level.