Xu et al. (2025) Correcting Atmospheric Temperature and Vapor Density Profiles of Ground-Based Microwave Radiometer in Diverse Skies by Regression Model and Artificial Neural Network Methods

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

Identification

• Journal: Remote Sensing
• Year: 2025
• Date: 2025-11-25
• Authors: Guirong Xu, Tang Yonglan, Aning Gou, Yajun Wang, Wenze Yang, Jing Yan
• DOI: 10.3390/rs17233819

Research Groups

Not explicitly stated in the provided text.

Short Summary

This study aims to improve ground-based microwave radiometer (MWR) temperature and vapor density retrieval accuracy by correcting deviations against radiosondes using regression and artificial neural network (ANN) models, finding that both models effectively reduce biases but do not significantly enhance retrieval consistency.

Objective

• To correct ground-based microwave radiometer (MWR) temperature and vapor density deviations against radiosondes in diverse sky conditions using regression and artificial neural network (ANN) methods, thereby improving MWR retrieval accuracy.

Study Configuration

• Spatial Scale: Wuhan, China (single location)
• Temporal Scale: 9 years of observations; MWR data at minute-level resolution

Methodology and Data

• Models used: Regression model, Artificial Neural Network (ANN)
• Data sources: Ground-based microwave radiometer (MWR) observations, radiosonde observations

Main Results

• MWR temperature exhibits a cold bias from radiosondes in clear and cloudy skies, but a warm bias in rainy skies.
• MWR vapor density is generally wetter than radiosondes, especially in rainy skies.
• Both regression and ANN models successfully reduce the biases of MWR temperature and vapor density against radiosondes to approximately zero across diverse sky conditions.
• The root mean square error (RMSE) of MWR vapor density in rainy skies shows a marked decrease after correction.
• After correction using the regression model, the RMSE of MWR temperature (vapor density) declined by 14% (7%) in clear skies, 7% (4%) in cloudy skies, and 12% (29%) in rainy skies.
• The ANN model showed slightly better performance, with corresponding RMSE decreases for MWR temperature (vapor density) of 19% (8%) in clear skies, 10% (8%) in cloudy skies, and 12% (30%) in rainy skies.
• The consistency of MWR retrievals with radiosondes was rarely improved after the corrections by either regression or ANN models.

Contributions

• Demonstrates the reasonable ability of regression and artificial neural network (ANN) models to correct ground-based microwave radiometer (MWR) retrieval deviations for temperature and vapor density in diverse sky conditions (clear, cloudy, rainy).
• Quantifies the improvements in bias reduction and RMSE for MWR retrievals using these correction methods.
• Identifies that while biases are reduced, there is remaining room for further improvement in MWR retrieval accuracy, particularly regarding consistency with radiosonde data.

Funding

Not explicitly stated in the provided text.

Citation

@article{Xu2025Correcting,
  author = {Xu, Guirong and Yonglan, Tang and Gou, Aning and Wang, Yajun and Yang, Wenze and Yan, Jing},
  title = {Correcting Atmospheric Temperature and Vapor Density Profiles of Ground-Based Microwave Radiometer in Diverse Skies by Regression Model and Artificial Neural Network Methods},
  journal = {Remote Sensing},
  year = {2025},
  doi = {10.3390/rs17233819},
  url = {https://doi.org/10.3390/rs17233819}
}