Articles | Volume 11, issue 8
Research article
09 Aug 2018
Research article |  | 09 Aug 2018

A neural network approach to estimating a posteriori distributions of Bayesian retrieval problems

Simon Pfreundschuh, Patrick Eriksson, David Duncan, Bengt Rydberg, Nina Håkansson, and Anke Thoss

Related authors

GPROF V7 and beyond: assessment of current and potential future versions of the GPROF passive microwave precipitation retrievals against ground radar measurements over the continental US and the Pacific Ocean
Simon Pfreundschuh, Clément Guilloteau, Paula J. Brown, Christian D. Kummerow, and Patrick Eriksson
Atmos. Meas. Tech., 17, 515–538,,, 2024
Short summary
The Chalmers Cloud Ice Climatology: Retrieval implementation and validation
Adrià Amell, Simon Pfreundschuh, and Patrick Eriksson
EGUsphere,,, 2023
Short summary
An improved near-real-time precipitation retrieval for Brazil
Simon Pfreundschuh, Ingrid Ingemarsson, Patrick Eriksson, Daniel A. Vila, and Alan J. P. Calheiros
Atmos. Meas. Tech., 15, 6907–6933,,, 2022
Short summary
Ice water path retrievals from Meteosat-9 using quantile regression neural networks
Adrià Amell, Patrick Eriksson, and Simon Pfreundschuh
Atmos. Meas. Tech., 15, 5701–5717,,, 2022
Short summary
GPROF-NN: a neural-network-based implementation of the Goddard Profiling Algorithm
Simon Pfreundschuh, Paula J. Brown, Christian D. Kummerow, Patrick Eriksson, and Teodor Norrestad​​​​​​​
Atmos. Meas. Tech., 15, 5033–5060,,, 2022
Short summary

Related subject area

Subject: Clouds | Technique: Remote Sensing | Topic: Data Processing and Information Retrieval
Deriving cloud droplet number concentration from surface-based remote sensors with an emphasis on lidar measurements
Gerald G. Mace
Atmos. Meas. Tech., 17, 3679–3695,,, 2024
Short summary
A random forest algorithm for the prediction of cloud liquid water content from combined CloudSat–CALIPSO observations
Richard M. Schulte, Matthew D. Lebsock, John M. Haynes, and Yongxiang Hu
Atmos. Meas. Tech., 17, 3583–3596,,, 2024
Short summary
Identification of ice-over-water multilayer clouds using multispectral satellite data in an artificial neural network
Sunny Sun-Mack, Patrick Minnis, Yan Chen, Gang Hong, and William L. Smith Jr.
Atmos. Meas. Tech., 17, 3323–3346,,, 2024
Short summary
A new approach to crystal habit retrieval from far-infrared spectral radiance measurements
Gianluca Di Natale, Marco Ridolfi, and Luca Palchetti
Atmos. Meas. Tech., 17, 3171–3186,,, 2024
Short summary
Multiple-scattering effects on single-wavelength lidar sounding of multi-layered clouds
Valery Shcherbakov, Frédéric Szczap, Guillaume Mioche, and Céline Cornet
Atmos. Meas. Tech., 17, 3011–3028,,, 2024
Short summary

Cited articles

Aires, F., Prigent, C., and Rossow, W. B.: Neural network uncertainty assessment using Bayesian statistics with application to remote sensing: 2. Output errors, J. Geophys. Res., 109, d10304,, 2004. a
Bishop, C. M.: Pattern Recognition and Machine Learning, Springer-Verlag New York, 2006. a
Brath, M., Fox, S., Eriksson, P., Harlow, R. C., Burgdorf, M., and Buehler, S. A.: Retrieval of an ice water path over the ocean from ISMAR and MARSS millimeter and submillimeter brightness temperatures, Atmos. Meas. Tech., 11, 611–632,, 2018. a
Buehler, S. A., Mendrok, J., Eriksson, P., Perrin, A., Larsson, R., and Lemke, O.: ARTS, the Atmospheric Radiative Transfer Simulator – version 2.2, the planetary toolbox edition, Geosci. Model Dev., 11, 1537–1556,, 2018. a
Cannon, A. J.: Quantile regression neural networks: Implementation in R and application to precipitation downscaling, Comput. Geosci., 37, 1277–1284,, 2011. a
Short summary
A novel neural-network-based retrieval method is proposed that combines the flexibility and computational efficiency of machine learning retrievals with the consistent treatment of uncertainties of Bayesian methods. Numerical experiments are presented that show the consistency of the proposed method with the Bayesian formulation as well as its ability to represent non-Gaussian retrieval errors. With this, the proposed method overcomes important limitations of traditional methods.