 |
| Titel |
New developments in spatial interpolation methods of Sea-Level Anomalies in the Mediterranean Sea |
| VerfasserIn |
Charles Troupin, Alexander Barth, Jean-Marie Beckers, Ananda Pascual |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250095945
|
| Publikation (Nr.) |
 EGU/EGU2014-11423.pdf |
|
|
|
|
|
| Zusammenfassung |
| The gridding of along-track Sea-Level Anomalies (SLA) measured by a constellation of
satellites has numerous applications in oceanography, such as model validation, data
assimilation or eddy tracking. Optimal Interpolation (OI) is often the preferred method for
this task, as it leads to the lowest expected error and provides an error field associated to the
analysed field. However, the numerical cost of the method may limit its utilization in
situations where the number of data points is significant. Furthermore, the separation of
non-adjacent regions with OI requires adaptation of the code, leading to a further increase of
the numerical cost.
To solve these issues, the Data-Interpolating Variational Analysis (DIVA), a technique
designed to produce gridded from sparse in situ measurements, is applied on SLA data in the
Mediterranean Sea. DIVA and OI have been shown to be equivalent (provided some
assumptions on the covariances are made). The main difference lies in the covariance
function, which is not explicitly formulated in DIVA.
The particular spatial and temporal distributions of measurements required adaptation in
the Software tool (data format, parameter determinations, ...). These adaptation are presented
in the poster.
The daily analysed and error fields obtained with this technique are compared with
available products such as the gridded field from the Archiving, Validation and
Interpretation of Satellite Oceanographic data (AVISO) data server. The comparison
reveals an overall good agreement between the products. The time evolution of the
mean error field evidences the need of a large number of simultaneous altimetry
satellites: in period during which 4 satellites are available, the mean error is on
the order of 17.5%, while when only 2 satellites are available, the error exceeds
25%.
Finally, we propose the use sea currents to improve the results of the interpolation,
especially in the coastal area. These currents can be constructed from the bathymetry or
extracted from a HF radar located in the Balearic Sea. |
|
|
|
|
|
|