Homogeneous atmospheric temperatures derived from IASI satellite observations : retrieval, space-time variations and extreme events

• Main Author: Bouillon, Marie
• Other Authors: TROPO - LATMOS, Laboratoire Atmosphères, Milieux, Observations Spatiales (LATMOS), Université de Versailles Saint-Quentin-en-Yvelines (UVSQ)-Institut national des sciences de l'Univers (INSU - CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Université de Versailles Saint-Quentin-en-Yvelines (UVSQ)-Institut national des sciences de l'Univers (INSU - CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Sorbonne Université, Cathy Clerbaux, Sarah Safieddine
• Format: Doctoral or Postdoctoral Thesis
• Language: French
• Published: HAL CCSD 2021
• Subjects: IASI; Atmospheric temperature; Climate change; Extreme events; Trends; Remote sensing; Températures atmosphériques; Changement climatique; Événements extrêmes; Tendances; Télédétection; [SDU.STU.CL]Sciences of the Universe [physics]/Earth Sciences/Climatology; [SDE.MCG]Environmental Sciences/Global Changes; [PHYS.PHYS.PHYS-AO-PH]Physics [physics]/Physics [physics]/Atmospheric and Oceanic Physics [physics.ao-ph]; Arctic
• Online Access: https://theses.hal.science/tel-03688289; https://theses.hal.science/tel-03688289/document; https://theses.hal.science/tel-03688289/file/BOUILLON_Marie_2021.pdf

The IASI instruments, launched in 2006, 2012 and 2018, observe radiance spectra from the Earth-atmosphere system, allowing us to retrieve atmospheric temperatures. The EUropean organisation for the exploitation of METeorological SATellites (EUMETSAT) computes surface and atmospheric temperatures from IASI observations since 2007. To improve the quality of the retrieved temperatures, EUMETSAT has performed several updates on the temperature processing algorithm. In this work, we show how the EUMETSAT temperature record is inhomogeneous and therefore not adapted to the study of temperatures on the long term. Therefore, a new atmospheric temperature product was computed from IASI spectra with an artificial neural network, using IASI radiances as input and ERA5 reanalysis temperatures as output. The temperatures retrieved with the neural network were validated against ERA5 temperatures and radiosounding temperatures. The validation showed good agreement between the three datasets from 750 to 7 hPa. At 2 hPa, the quality of the temperature products decreases because of the low amount of information on temperature within the IASI spectra at this altitude. This new homogeneous temperature record was then used to study extreme events, in particular Sudden Stratospheric Warmings (SSW). This work studies several SSWs that have happened since 2007, and shows that IASI is perfectly suited to study both temperature and ozone changes during these events. The temperatures were also used to observe the cold and heat waves that happened in the past few years. Finally, the temperatures were used to compute trends over the 2008-2020 period. The troposphere is warming, especially in the upper equatorial troposphere and in the Arctic. The stratosphere north of 50°S is also warming due to the slow ozone hole recovery. In the rest of the stratosphere, the temperatures are decreasing. This thesis presents an homogenous and independent temperature data record from IASI. The methods developed can and will be applied to the rest of the ...