Figure 5: Overview of the relative statistical distributions of
fulfilled criteria. The full synergy dataset is shown in the top panel,
the synergetic retrievals in MY 27 in the second, SPICAM/NIR is shown in
the third panel, while PFS/TIR is illustrated in the bottom panel.
Retrievals with the individual instruments were only conducted for MY
27. The y-axis counts the number of fulfilled criteria, and the x-axis
shows the number of retrievals per 10 000 which fulfill each criterium
configuration, sorted with the most likely configuration to the left,
and decreasingly likely configurations towards the right.
4 Results
The results presented below were derived using data from two time
intervals; Ls=334 of MY 26 until the end of MY 29, and from the
beginning of MY 32 to the end of MY 34. The spatial and temporal
coverage within each year is highly variable. All four selection
criteria described in section 3.2 were applied, with limits as shown in
Figure 5 (except for the assembly of the column abundance climatology in
section 4.1, which uses limits specifically adjusted for column
abundances). The first part of this section focuses on the retrieval of
total column abundances with the synergy compared to with single
spectral domain approaches and to the MCD. Then, results concerning the
vertical partitioning of water is shown, as seen in relation to the
total column and model predictions, where special attention is given to
the polar regions around the seasonal ice cap sublimation seasons.
4.1 Column abundance climatology
For the assembly of a complete and composite synergy climatology of
water vapor, the criteria described in Section 3 are applied to the
complete dataset shown in Figure 2, with an adjustment to the DOF and
error ratio limits. The DOF and error ratio requirements ensure
retrievals with sufficient vertical information to justify further
analysis of partitioning, and as there is no need for vertical
information content for a column abundance climatology, the DOF and
error ration limits are both set to 1, which is more than sufficient to
infer a highly reliable column abundance.
After the adjusted selection criteria are applied, the remaining water
vapor retrievals are illustrated as a function of season, latitude and
longitude in Figure 6. In order to account for topography, the total
water columns are normalized to an equivalent surface pressure of 610 Pa
using the MCD to constrain local pressure. However, this pressure
normalization is made on the assumption that water vapor is well mixed
with the ambient gases, which might not always be the case. The total
water columns are averaged in bins of 2° Ls, 2° latitude and 4°
longitude. Of all qualified retrievals, 31% are from MY 27, which for
the most part contributes to the coverage of the northern summer.