 |
| Titel |
A simplified approach for generating GNSS radio occultation refractivity climatologies |
| VerfasserIn |
H. Gleisner, S. B. Healy |
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| ISSN |
1867-1381
|
| Digitales Dokument |
URL |
| Erschienen |
In: Atmospheric Measurement Techniques ; 6, no. 1 ; Nr. 6, no. 1 (2013-01-22), S.121-129 |
| Datensatznummer |
250017376
|
| Publikation (Nr.) |
 copernicus.org/amt-6-121-2013.pdf |
|
|
|
|
|
| Zusammenfassung |
| The possibility of simplifying the retrieval scheme required to produce GNSS radio
occultation refractivity climatologies is investigated. In a new, simplified retrieval
approach, the main statistical analysis is performed in bending angle space and an
estimate of the average bending angle profile is then propagated through an Abel transform.
The average is composed of means and medians of ionospheric corrected bending angles
up to 80 km. Above that, the observed profile is exponentially extrapolated to infinity
using a fixed a priori scale height.
The new approach circumvents the need to introduce a "statistical optimisation" processing
step in which individual bending angle profiles are merged with a priori data,
often taken from a climatology. This processing step can be complex, difficult to interpret,
and is generally recognised as a potential source of structural uncertainty.
The new scheme is compared with the more conventional approach of averaging individual
refractivity profiles, produced with the implementation of statistical optimisation
used in the EUMETSAT Radio Occultation Meteorology Satellite Application Facility (ROM SAF)
operational processing. It is shown that the two GNSS radio occultation
climatologies agree to within 0.1% from 5 km up to 35–40 km, for the three months
January, February, and March 2011. During this time period, the new approach also produces
slightly better agreement with ECMWF analyses between 40–50 km, which is encouraging.
The possible limitations of the new approach caused by mean residual ionospheric errors
and low observation numbers are discussed briefly, and areas for future work are suggested. |
| |
|
| Teil von |
|
|
|
|
|
|
|
|