Combined neural network/Phillips–Tikhonov approach to aerosol retrievals over land from the NASA Research Scanning Polarimeter

Di Noia, Antonio; Hasekamp, Otto P.; Wu, Lianghai; van Diedenhoven, Bastiaan; Cairns, Brian; Yorks, John E.

doi:https://doi.org/10.5194/amt-10-4235-2017

Articles | Volume 10, issue 11

https://doi.org/10.5194/amt-10-4235-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/amt-10-4235-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 10, issue 11

Research article

|

08 Nov 2017

Research article |

| 08 Nov 2017

Combined neural network/Phillips–Tikhonov approach to aerosol retrievals over land from the NASA Research Scanning Polarimeter

Antonio Di Noia, Otto P. Hasekamp, Lianghai Wu, Bastiaan van Diedenhoven, Brian Cairns, and John E. Yorks

Download

Final revised paper (published on 08 Nov 2017)
Supplement to the final revised paper
Preprint (discussion started on 26 Jun 2017)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review', Anonymous Referee #2, 31 Jul 2017
- AC1: 'Reply', Antonio Di Noia, 18 Sep 2017
RC2: 'review', Alexei Lyapustin, 04 Aug 2017
- AC2: 'Reply', Antonio Di Noia, 18 Sep 2017

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Antonio Di Noia on behalf of the Authors (18 Sep 2017) Author's response Manuscript

ED: Publish as is (20 Sep 2017) by Alexander Kokhanovsky

AR by Antonio Di Noia on behalf of the Authors (22 Sep 2017)

Short summary

In this paper an algorithm for the retrieval of aerosol properties from NASA Research Scanning Polarimeter (RSP) data is presented. An artificial neural network is used to produce a first estimate of the aerosol properties, which is then improved using an iterative retrieval scheme based on Phillips–Tikhonov regularization. Using the neural network retrievals as a first guess for the Phillips–Tikhonov improved the retrieval convergence, confirming results previously found on ground-based data.