Abonnement à la biblothèque: Guest
Page d'accueil Archives Thermal Letter Responsables Réunions à venir Assembly for International Heat Transfer Conferences
International Heat Transfer Conference 13

ISSN: 2377-424X (online)
ISSN: 2377-4371 (flashdrive)

A BAYESIAN ALGORITHM FOR THE RETRIEVAL OF GEOPHYSICAL PARAMETER IN THE ATMOSPHERE

M. Deiveegan
Indian Institute of Technology Madras, Chennai, India

V. Swaminathan
Thermal specialist, Flomerics, Bangalore, India

Chakravarthy Balaji
Heat Transfer and Thermal Power Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai-600036, India

S. P. Venkateshan
Department of Mechanical Engineering, IIT Madras, Chennai 600036, India

DOI: 10.1615/IHTC13.p23.20
12 pages

Résumé

Radiation reflected and transmitted by a planetary atmosphere contains information about particles and molecules in the atmosphere. Therefore, accurate modeling of the radiation field may be used to retrieve information on atmospheric precipitations. In this paper, a simple two-layer model for a vertically inhomogeneous atmosphere is implemented by using the doubling-adding method for a plane-parallel atmosphere. Additionally, sea surface albedo, rain rate and ice content for two layers are estimated simultaneously by inverse analysis, from the knowledge of measured brightness temperatures based on vertical and horizontal polarizations. Measured brightness temperatures are simulated by adding random errors to the brightness temperatures computed from the forward problem. To assess the accuracy of predictions, a statistical analysis is made to establish error bar for inverse solution. The inverse methodology to retrieve the parameters using Bayesian retrieval algorithm is presented. To check the performance and accuracy obtained in the retrieval, a comparison is presented between two retrieval methods, viz. Bayesian algorithm and artificial neural network. The results show that the Bayesian retrieval algorithm and artificial neural network are robust and yield accurate estimation of parameters even when noise is present in the observation. Both artificial neural network and Bayesian retrieval algorithms yield estimation of parameters in real time with good accuracy. However they require time consuming data base creation.

Achat $25.00 Check subscription Télécharger l'archive MARC Déontologie de l'édition scientifique et abus Recommander à mon bibliothécaire Ajouter cette page aux favoris
 

Bibliothèque numérique IHTC

Measurement of fluid temperature with an arrangement of three thermocouples