Xu et al. (2026) Satellite Residual Two‐Stage Convolutional Neural Network Heating (R2CH) Algorithm for Retrieving Latent Heat Between the Precipitation Top and the Cloud Top

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Journal of Geophysical Research Atmospheres
• Year: 2026
• Date: 2026-07-06
• Authors: Xuanye Xu, Shuping Yang, Hongyun Zhao, Qiong Wu, Lin Chen, Peng Zhang, Renjun Zhou, Dong Li, Rui Si Li
• DOI: 10.1029/2025jd045988

Research Groups

• National Institute of Information and Communications Technology (NICT)
• University of Tokyo
• Japan Aerospace Exploration Agency (JAXA)

Short Summary

This study develops a novel algorithm, R2CH, to retrieve latent heat release associated with non-precipitating clouds above the precipitation top height. The method effectively captures LH from these clouds and reproduces both horizontal and vertical structures.

Objective

• Investigate the retrieval of latent heat release associated with phase changes in water above the precipitation top height using a deep learning approach.

Study Configuration

• Spatial Scale: Global, focusing on tropical regions.
• Temporal Scale: Monthly averages for a 3-year period (2017-2020).

Methodology and Data

• Models used: Residual Two-stage Convolutional Neural Network Heating (R2CH) algorithm.
• Data sources: GPM and Himawari-8 satellite observations.

Main Results

• R2CH effectively captures latent heat release from non-precipitating clouds above the precipitation top height, with a relative error of 10.23% compared to WRF model simulations.
• The retrieved LH is generally weaker than that from two official GPM LH products but more consistent with them at the layer above the PTH.

Contributions

• This study addresses the underestimation of latent heat release associated with non-precipitating clouds, which accounts for up to 9.74% of the total.
• The R2CH algorithm provides a novel approach for retrieving LH from these clouds and reproduces both horizontal and vertical structures.

Funding

• This research was supported by JST-CREST (JPMJCR18A6) and JSPS KAKENHI (JP20H02025).

Citation

@article{Xu2026Satellite,
  author = {Xu, Xuanye and Yang, Shuping and Zhao, Hongyun and Wu, Qiong and Chen, Lin and Zhang, Peng and Zhou, Renjun and Li, Dong and Li, Rui Si},
  title = {Satellite Residual Two‐Stage Convolutional Neural Network Heating (R2CH) Algorithm for Retrieving Latent Heat Between the Precipitation Top and the Cloud Top},
  journal = {Journal of Geophysical Research Atmospheres},
  year = {2026},
  doi = {10.1029/2025jd045988},
  url = {https://doi.org/10.1029/2025jd045988}
}