 |
| Titel |
An integrated view of Southern Ocean cyclones from WindSat |
| VerfasserIn |
Adrian McDonald, Jack Coggins |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250108782
|
| Publikation (Nr.) |
 EGU/EGU2015-8552.pdf |
|
|
|
|
|
| Zusammenfassung |
| In this study, we use the WindSat satellite dataset to form an integrated view of the
characteristics of extra-tropical cyclones over the Southern Ocean. WindSat polarimetric
microwave radiometer measurements of brightness temperature at a variety of frequencies
and polarisations can be used to derive sea-surface temperature, near-surface horizontal wind
velocities, total column water vapour, cloud liquid water content and rain rate over the
ocean surface and are not presently assimilated into reanalyses. Thus, this dataset
provides a powerful way to examine the strengths of reanalyses’ representations of
cyclones.
To examine cyclones over the Southern Ocean statistically, we transform the satellite data
into a cyclone centred coordinate system by forming composites of data based on cyclone
positions ascertained from the ERA-Interim reanalysis surface pressure field; composites
are derived using all cyclone centres between 40 and 60°S for the period 2007 to
2012. The composites are shown to match well with composites formed from other
satellite datasets, namely AMSR-E and NVAP-M, and confirm the quality of the
WindSat dataset. We then compare the WindSat composites with those derived from
the ERA-Interim output to examine the similarities and differences between these
composites. Our results suggest that the mean cyclone composite horizontal wind
field observed by WindSat and ERA-Interim matches rather well, providing an
independent validation of the ERA-Interim output. However, inspection shows that the
two datasets’ velocity distributions are quite different in some quadrants of the
cyclone composite, suggesting that a simple analysis of mean patterns around the
cyclone can hide relevant detail. Comparisons between the two composites for the
total column water vapour show larger differences, with the largest differences
(roughly 10%) being observed near the cold frontal region. The correspondence
becomes still poorer when the cloud liquid water content and rain rate fields are
intercompared, with relative differences being as large as 30%. This provides an independent
analysis suggesting that the cloud field related to cyclones in the reanalysis contain
significant flaws over the Southern Ocean, as might be expected from previous studies.
Examination of the differences between the ERA-Interim and WindSat composites are also
shown to increase as the cyclone evolves suggesting errors accumulate as the system
develops. |
|
|
|
|
|
|