ESA / ECSAT CCIWV.REP.022 Water Vapour Climate Change Initiative (WV_cci) - CCI+ Phase 2 D2.2 data from ERA5 and local WV data from BBH profiles. This work is ongoing and will be updated in the next update of this document. 3.2 Bias-correction on water vapour profiles The original retrieved water vapour profiles consist of limb measurements from Aura-MLS and MIPAS instruments and the nadir data product IMS (which is based on a combination of IASI, MHS and AMSU satellite measurements). As mentioned in Section 3.1, the input VRWV profiles from different satellite instruments will be bias-corrected via a quantile-mapping technique in comparison to reference data from BBH observations.
Limited to the spatial and temporal coverage of the BBH observations, the comparisons are performed in a common geophysical reference coordinate system – here, the thermal tropopause height is chosen – to reduce geophysical noise. Figure 3-1 shows the comparison of VRWV profiles from different satellites to reference tropopause as reference). Due to limited spatial and temporal coverage, JULIA aircraft data datasets in a geophysical-based vertical coordinate system (here, using the thermal is not used in the bias-correction process and only BBH profile observations are chosen as reference data (see DARD [RD-8]).
The comparisons to BBH profile data clearly show that the limb WV data match well with hygrometer profiles above the tropopause and the nadir VRWV data from IMS agree better with the hygrometer profiles below the tropopause. Due to the limitation of the spatial and temporal coverage, the highly sparse ACE-FTS and ACE- MAESTRO are not included in the production of CDR-4 data. In Figure 3-1, all profiles from satellites and BBH are screened with quality control suggestions from each instrument. It is noteworthy to mention that the comparison is carried out in five latitude bands: 90°S–60°S, 60°S–30°S, 30°S–30°N, 30°N–60°N, and 60°N–90°N, due to the limitation of the sparse BBH observations.
The following bias-correction process will also be performed in each latitude band. As there is no BBH site in the latitude band 90°S–60°S, the original profiles from all satellite observations are used in the production of CDR-4 data in this band. The choice of these latitude bands leads to the problem with tropospheric/stratospheric intrusions around 30° in latitude, in the case of double thermal tropopause. In practice, only the lowest thermal tropopause is used in this study and the tropopause height of 14 km is chosen as the limit to deal with tropospheric/stratospheric intrusions.
The tropical profiles (30°S–30°N) with a tropopause height less than 14 km are treated as stratospheric intrusions with a tropopause height larger than 14 km are regarded as tropospheric intrusions from from mid-latitude and assigned into mid-latitude bands. Conversely, the mid-latitude profiles tropical regions and assigned into the tropical band. This latitude correction on the profiles 14