Preprints
https://doi.org/10.5194/gmd-2024-25
https://doi.org/10.5194/gmd-2024-25
Submitted as: development and technical paper
 | 
11 Mar 2024
Submitted as: development and technical paper |  | 11 Mar 2024
Status: a revised version of this preprint was accepted for the journal GMD and is expected to appear here in due course.

Deep Learning Driven Simulations of Boundary Layer Cloud over the US Southern Great Plains

Tianning Su and Yunyan Zhang

Abstract. This study developed a deep learning model to simulate the complex dynamics of boundary layer clouds (BLCs) over the US Southern Great Plains. Using over twenty years of extensive observations from the Atmospheric Radiation Measurement program for training and validation, the model diagnoses the BLCs from the perspective of cloud-land coupling. Morning meteorological profiles set as the initial conditions and then identifying triggers for BLCs formation from surface meteorology. The deep learning model offer accurate simulation of the convection initiation and cloud base of BLCs. In comparison with reanalysis data (i.e., ERA-5 and MERRA-2), it provides a notable improvement in the vertical structure of low clouds from a climatological perspective. The deep learning model can serve as the cloud parameterization and extend to analyzing stratiform and cumulus clouds within reanalysis frameworks, offering insights into improving the simulation of BLCs. By quantifying biases due to various meteorological factors and parameterizations, this deep learning-driven approach bridges the observational-modeling divide. Surface humidity and parameterization emerge as key limiting factors to affect the representation of BLCs in the reanalysis data. This deep learning approach holds promise for improving the convection parameterization and advancing model diagnostics in weather forecasting and climate modelling.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Tianning Su and Yunyan Zhang

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2024-25', Anonymous Referee #1, 28 Apr 2024
  • RC2: 'Review of the original manuscript - MAJOR REVISIONS', Anonymous Referee #2, 04 May 2024

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on gmd-2024-25', Anonymous Referee #1, 28 Apr 2024
  • RC2: 'Review of the original manuscript - MAJOR REVISIONS', Anonymous Referee #2, 04 May 2024
Tianning Su and Yunyan Zhang
Tianning Su and Yunyan Zhang

Viewed

Total article views: 684 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
433 221 30 684 28 26
  • HTML: 433
  • PDF: 221
  • XML: 30
  • Total: 684
  • BibTeX: 28
  • EndNote: 26
Views and downloads (calculated since 11 Mar 2024)
Cumulative views and downloads (calculated since 11 Mar 2024)

Viewed (geographical distribution)

Total article views: 688 (including HTML, PDF, and XML) Thereof 688 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 13 Jul 2024
Download
Short summary
Using two decades of field observation over the Southern Great Plains, this study developed a deep learning model to simulate the complex dynamics of boundary layer clouds. The deep learning model can serve as the cloud parameterization within reanalysis frameworks, offering insights into improving the simulation of low clouds. By quantifying biases due to various meteorological factors and parameterizations, this deep learning-driven approach bridges the observational-modeling divide.