Articles | Volume 14, issue 7
Geosci. Model Dev., 14, 4641–4654, 2021
https://doi.org/10.5194/gmd-14-4641-2021
Geosci. Model Dev., 14, 4641–4654, 2021
https://doi.org/10.5194/gmd-14-4641-2021

Model description paper 28 Jul 2021

Model description paper | 28 Jul 2021

Exploring deep learning for air pollutant emission estimation

Lin Huang et al.

Download

Interactive discussion

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2021-80', Anonymous Referee #1, 30 Apr 2021
    • AC1: 'Reply on RC1', J. X. XING, 14 May 2021
  • RC2: 'Comment on gmd-2021-80', Anonymous Referee #2, 10 May 2021
    • AC2: 'Reply on RC2', J. X. XING, 25 May 2021

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision
AR by J. X. XING on behalf of the Authors (25 May 2021)  Author's response    Author's tracked changes    Manuscript
ED: Referee Nomination & Report Request started (11 Jun 2021) by Samuel Remy
RR by Anonymous Referee #2 (17 Jun 2021)
RR by Anonymous Referee #1 (25 Jun 2021)
ED: Publish as is (01 Jul 2021) by Samuel Remy
AR by J. X. XING on behalf of the Authors (02 Jul 2021)  Author's response    Manuscript
Download
Short summary
Accurate estimation of emissions is a prerequisite for effectively controlling air pollution, but current methods lack either sufficient data or a representation of nonlinearity. Here, we proposed a novel deep learning method to model the dual relationship between emissions and pollutant concentrations. Emissions can be updated by back-propagating the gradient of the loss function measuring the deviation between simulations and observations, resulting in better model performance.