Bottleneck Channels Algorithm for Satellite Data Dimension Reduction: A Case Study for IASI

Victor Pellet (1) , Filipe Aires (1)

1 LERMA (UMR_8112) - Laboratoire d'Etude du Rayonnement et de la Matière en Astrophysique

The Infrared Atmospheric Sounder Interferometer (IASI) onboard the Metop-A and-B satellites is an essential instrument for numerical weather prediction centres. IASI observes surface and atmospheric geophysical variables twice a day around the world with its 8461 channels. This data volume will even increase in the future, stressing the need for practical and efficient dimension reduction techniques for storage, transmission, but also for inversion or assimilation. In a previous paper [Aires et al., 2015a], a new dimension reduction method, the so-called Bottleneck-Channel (BC), has been introduced. BCs are a good compromise between compression and selection techniques. In this paper , we introduce several new technical configurations of the BC approach, comparing their advantages/drawbacks. Comparisons are also made with two classical methodolo-gies widely used in the satellite community: Principal Component Analysis (PCA) compression and Entropy Reduction (ER) channel selection. The methodologies are tested on the temperature, water vapor and ozone retrieval application using IASI measurements. It is shown that the BCs present compression, de-noising and information content results analogue to the PCA, while preserving the physical meaning (i.e., actual channel TBs) of a channel selection approach. This has numerous advantages for inversion and assimilation: the mixing problem, contamination effects, minor constituants representation , or radiative transfer speed will be discussed.

Domaines: Sciences de l'environnement

Victor Pellet, Filipe Aires. Bottleneck Channels Algorithm for Satellite Data Dimension Reduction: A Case Study for IASI. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56 (10), pp.6069-6081. ⟨10.1109/TGRS.2018.2830123⟩. ⟨hal-02376956⟩