ESA / ECSAT CCIWV.REP.022 Water Vapour Climate Change Initiative (WV_cci) - CCI+ Phase 2 D2.2 meteorological feature that determines the structure found in UTLS water vapour [RD-10–RD- 14], is chosen as the geophysical reference coordinate system in the vertical. In this new tropopause-based coordinate system, the biases in VRWV can be calculated between satellite instruments, and BBH observations including VRWV data obtained over a larger geographical domain. With the bias-correction applied to the VRWV data from satellite instruments, it is possible to merge together the VRWV profiles from both limb- and nadir-viewing geometries into a single data product.
It is highlighted here that the merging approach is assumed to mostly correct for the smoothing characteristics of the averaging kernels of the respective instruments. For this prototype CDR-4 product, there are still several limitations : 1. Biases between satellite data and BBH observations in the geophysical reference coordinate system are calculated within several broad latitude bands, thus the longitudinal variations of these biases may not be fully accounted for in this process. It should also be noted that in latitude bands where no BBH observations are available, the lack of a bias-correction may currently lead to discontinuities between adjacent latitude bands.
Tests on the bias-correction algorithm with ERAi reanalysis WV data on the quality of bias-correction to the satellite profiles. Thus, more frequent and denser show that the spatial and temporal coverage of the reference data have a notable impact BBH observations (or an alternative instrument reference) are needed across the globe for a quantitatively better bias-correction for the satellite observations. New tests introducing ACE-FTS profiles as add-on reference data in the bias-correction procedure suggest that the ACE-FTS profiles tends to be too dry over the tropical region in the upper troposphere.
Different reference profiles or a different combination (e.g., ACE in the extratropical lowermost stratosphere and tropical tropopause region down to 150 hPa and BBH in the tropical upper troposphere), should be explored. 2. The satellite instruments show a big difference in the amount of VRWV profile measurements across the UTLS. The proportional contribution of each satellite observations to the merged CDR-4 product is important to the final data quality. An appropriate weighting scheme to the satellite data based on the derived input quality and derived instrument-dependent biases is essential for improving the final product quality. 3.
The merging rules are likely to bring in a discontinuity in the merged CDR-4 product at 300 and 100 hPa. Assessment of this merging problem shows that there is no need for post-processing at 100 hPa due to the high quality of instrument WV values and 21