Preprints
https://doi.org/10.31223/X5197G
https://doi.org/10.31223/X5197G
01 Jul 2024
 | 01 Jul 2024
Status: this preprint is open for discussion.

FootNet v1.0: Development of a machine learning emulator of atmospheric transport

Tai-Long He, Nikhil Dadheech, Tammy M. Thompson, and Alexander J. Turner

Abstract. There has been a proliferation of dense observing systems to monitor greenhouse gas (GHG) concentrations over the past decade. Estimating emissions with these observations is often done using an atmospheric transport model to characterize the source-receptor relationship, which is commonly termed measurement "footprint". Computing and storing footprints using full-physics models is becoming expensive due to the requirement of simulating atmospheric transport at high resolution. We present the development of FootNet, a deep learning emulator of footprints at kilometer scale. We train and evaluate the emulator using footprints simulated using a Lagrangian particle dispersion model (LPDM). FootNet predicts the magnitudes and extents of footprints in near-real-time with high fidelity. We identify the relative importance of input variables of FootNet to improve the interpretability of the model. Surface winds and a precomputed Gaussian plume from the receptor are identified to be the most important variables for footprint emulation. The FootNet emulator developed here may help address the computational bottleneck of flux inversions using dense observations.

Tai-Long He, Nikhil Dadheech, Tammy M. Thompson, and Alexander J. Turner

Status: open (until 26 Aug 2024)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'No compliance with the policy of the journal', Juan Antonio Añel, 16 Jul 2024 reply
    • AC1: 'Reply on CEC1', Tai-Long He, 16 Jul 2024 reply
      • CEC2: 'Reply on AC1', Juan Antonio Añel, 19 Jul 2024 reply
        • AC2: 'Reply on CEC2', Tai-Long He, 24 Jul 2024 reply
Tai-Long He, Nikhil Dadheech, Tammy M. Thompson, and Alexander J. Turner
Tai-Long He, Nikhil Dadheech, Tammy M. Thompson, and Alexander J. Turner

Viewed

Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.

Total article views: 161 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
161 0 0 161 0 0
  • HTML: 161
  • PDF: 0
  • XML: 0
  • Total: 161
  • BibTeX: 0
  • EndNote: 0
Views and downloads (calculated since 01 Jul 2024)
Cumulative views and downloads (calculated since 01 Jul 2024)

Viewed (geographical distribution)

Since the preprint corresponding to this journal article was posted outside of Copernicus Publications, the preprint-related metrics are limited to HTML views.

Total article views: 158 (including HTML, PDF, and XML) Thereof 158 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 24 Jul 2024
Download
Short summary
It is computationally expensive to infer greenhouse gas (GHG) emissions using atmospheric observations. This is partly due to the detailed model used to represent atmospheric transport. We demonstrate how a machine learning (ML) model can be used to simulate high-resolution atmospheric transport. This type of ML model will help estimate GHG emissions using dense observations, which are becoming increasingly common with the proliferation of urban monitoring networks and geostationary satellites.