Skip to Main content Skip to Navigation
Theses

Assimilation de données d'images télédétectées en météorologie

Abstract : Isolated vortices have been shown to emerge in two-dimensional and geostrophic turbulent flows. Coherent structures are apparent in three-dimensional atmospheric flows as well, and may share the property to determine the non-linear evolution and the predictability of the flow. Potential vorticity anomalies have indeed been shown to be of primary importance for cyclogenesis of mid-latitude storms. Their signature can be detected in satellite water vapour images and are known as dry intrusions. The first part of this work proceeds the tracking of dry intrusions on images. We use image processing tools based on thresholding algorithm, and develop additional filters to select relevant cells. One of this filters uses dynamical information on the jet from the background upper-level wind. We highlight the fact that conventional data assimilation does not handle coherent structures, and describe alternative procedures. We focus then on the assimilation of bogussed potential vorticity observations. This leads to a new vision of the relationship between water vapour and potential vorticity. The case study of the storms that has affected the Landes region in 2006 is then presented. We finally pinpoint ozone data as an alternative way to improve the initialization of upper-level potential vorticity
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00489243
Contributor : Abes Star :  Contact
Submitted on : Friday, June 4, 2010 - 11:51:14 AM
Last modification on : Friday, April 5, 2019 - 8:14:28 PM
Long-term archiving on: : Friday, September 17, 2010 - 1:08:59 PM

File

2008PEST0275_0_0.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-00489243, version 1

Collections

Citation

Yann Michel. Assimilation de données d'images télédétectées en météorologie. Sciences de la Terre. Université Paris-Est, 2008. Français. ⟨NNT : 2008PEST0275⟩. ⟨tel-00489243⟩

Share

Metrics

Record views

612

Files downloads

3015