Abstract

Two methods for performing clear-air temperature retrievals from simulated radiances for the Atmospheric Infrared Sounder are investigated. Neural networks are compared with a well-known linear method in which regression is performed after a change of bases. With large channel sets, both methods can rapidly perform clear-air retrievals over a variety of climactic conditions with an overall RMS error of less than 1 K. The Jacobian of the neural network is compared with the Jacobian (the regression coefficients) of the linear method, revealing a more fine-scale variation than expected from the underlying physics, particularly for the neural net. Some pragmatic information concerning the application of neural nets to retrieval problems is also included.

© 1995 Optical Society of America

Fast radiative-transfer model based on the correlated k-distribution method for a high-resolution satellite sounder

Yuzo Mano and Hiroshi Ishimoto
Appl. Opt. 43(34) 6304-6312 (2004)

Neural network pattern recognition of thermal-signature spectra for chemical defense

Arthur H. Carrieri and Pascal I. Lim
Appl. Opt. 34(15) 2623-2635 (1995)

Potential use of spaceborne lidar measurements to improve atmospheric temperature retrievals from passive sensors

Patrick Chazette, Gérard Mégie, and Jacques Pelon
Appl. Opt. 37(33) 7670-7679 (1998)

Neural network scheme for the retrieval of total ozone from Global Ozone Monitoring Experiment data

Martin D. Müller, Anton Kaifel, Mark Weber, and John P. Burrows
Appl. Opt. 41(24) 5051-5058 (2002)

Thermodynamic product retrieval methodology and validation for NAST-I

Daniel K. Zhou, William L. Smith, Jun Li, Hugh B. Howell, Greg W. Cantwell, Allen M. Larar, Robert O. Knuteson, David C. Tobin, Henry E. Revercomb, and Stephen A. Mango
Appl. Opt. 41(33) 6957-6967 (2002)