Développement et validation d'un modèle statistique de la surface de la mer pour la télédétection aux hyperfréquences / Development and validation of a sea surface statistical model for microwave remote sensing
(2005)
Files
ThCh_Chapitre_0_in_palatino_pdf.pdf
Open Access - Adobe PDF
- 106.96 KB
ThCh_Chapitre_1_in_palatino_pdf.pdf
Open Access - Adobe PDF
- 197.69 KB
ThCh__Titre_in_palatino_pdf.pdf
Open Access - Adobe PDF
- 51.2 KB
ThCh_Chapitre_7_in_palatino_pdf.pdf
Open Access - Adobe PDF
- 506.55 KB
ThCh_Chapitre_TdM_in_palatino_pdf.pdf
Open Access - Adobe PDF
- 88.13 KB
Details
- Authors
- Supervisors
- Abstract
- (en) The Telecommunications and Remote Sensing Laboratory of UCL has developed a method to retrieve the sea surface state parameters, from remote sensing measurements collected by a combination of microwave satellite payloads. This approach has required to construct a rough sea surface statistical model (wave displacement spectrum, long wave slope probability density function, ... ) as a function of a limited number of relevant sea state parameters, to develop an electromagnetic scattering model suitable for the rough sea surface model, to simulate radar and radiometric measurements for a given set of sea states and to develop a computational inversion method to retrieve the sea state parameters from satellite data. The objective of this thesis has been to analyze, develop and implement the statistical modeling of the rough sea surface and to validate this model (called UCL-3) on real data. In this research, four parameters have been chosen as the minimum set required to provide a suitable enough description of the sea state for microwave remote sensing purposes. The small wave structure of the rough sea depends on the instantaneous local wind speed, therefore it may be described by this parameter only. On the other hand, the large sea wave structure which contains the wave history needs more degrees of freedom; therefore, we propose to describe these waves not only as a function of the wind speed but also of the wave significant slope, of the fetch and of the wavenumber at the peak of the surface displacement spectrum. With respect to previous researches made at UCL we have introduced an additional peak enhancement in the large sea wave spectrum. The quality of the resulting sea surface model has been confronted with "ground truth data" in particular those from the BEATRICE buoy (located near Ouessant Island, West of France). In a second step, the surface model, along with the electromagnetic scattering one, has then been validated on satellite data. In this thesis a particular attention has been paid to the TANDEM mission synergistically combining data from different sensors borne on two different satellites ERS1 and ERS2. The obtained geophysical retrievals have been compared with simultaneous "in situ" buoy measurements from a set of NOAA buoys located in the Atlantic Ocean, the Pacific Ocean and the Gulf of Alaska. This comparison allowed us to improve models previously derived at UCL and to reduce the percentage of retrieval failures i.e. the amounts of data points for which the retrieval procedure in the previous researches was failing. These sea surface results from inverted radar data agree with those derived from buoy data.
- Affiliations
Citations
Baufays - Gaublomme, C. (2005). Développement et validation d’un modèle statistique de la surface de la mer pour la télédétection aux hyperfréquences / Development and validation of a sea surface statistical model for microwave remote sensing. https://hdl.handle.net/2078.5/129191