Mishra et al. (2025) Assessment of EOS-07 MHS satellite observations and retrieval of specific humidity profiles using a random forest-based algorithm

Identification

• Journal: Remote Sensing of Environment
• Year: 2025
• Date: 2025-10-10
• Authors: Manoj Kumar Mishra, Rishi Kumar Gangwar, Munn Vinayak Shukla, Prashant Kumar, P. K. Thapliyal
• DOI: 10.1016/j.rse.2025.115066

Research Groups

Space Applications Centre, ISRO, Ahmedabad, India

Short Summary

This study presents a preliminary performance assessment of the EOS-07 MHS satellite instrument, validating its brightness temperature observations and developing a random forest-based algorithm for retrieving specific humidity profiles, which demonstrated good agreement with reanalysis and radiosonde data and improved atmospheric forecasts when assimilated into a numerical weather prediction model.

Objective

• To conduct a preliminary performance assessment of the EOS-07 MHS satellite, including brightness temperature validation and the development and validation of a random forest-based algorithm for retrieving specific humidity profiles.

Study Configuration

• Spatial Scale: Global (satellite observations); assessment of moisture representation in the lower and middle atmosphere.
• Temporal Scale: Preliminary assessment of data from EOS-07 MHS (launched February 2023); a month-long cyclic assimilation experiment.

Methodology and Data

• Models used: Radiative Transfer for TOVS (RTTOV), Weather Research and Forecasting (WRF) model, Random Forest algorithm.
• Data sources: EOS-07 Microwave Humidity Sounder (MHS) observations, Advanced Technology Microwave Sounder (ATMS) observations, ERA5 reanalysis data, Radiosonde observations.

Main Results

• EOS-07 MHS brightness temperature biases relative to RTTOV simulations were within ±1 K, except for channels 1 and 6.
• Intercomparisons with ATMS observations showed brightness temperature biases within ±1 K and a standard deviation of 2–3 K.
• The random forest-based algorithm successfully retrieved specific humidity profiles. Compared to radiosonde data, the mean bias was approximately 0.78 g/kg and the standard deviation was 2.3 g/kg.
• Mean percentage bias for specific humidity was within ±20 % below 800 hPa and between ±20 % and ±40 % above 800 hPa when compared to radiosonde data.
• Relative to ERA5, the mean bias and root-mean-square deviation (RMSD) of retrieved specific humidity were under 30 % and 50 %, respectively.
• Estimated total precipitable water vapor showed a mean bias of 1.7–3.1 mm and a standard deviation of 5.2–5.7 mm compared to ERA5.
• Assimilation of EOS-07 MHS data into the WRF model resulted in improved atmospheric analyses and forecasts, with consistent enhancements in moisture representation across the lower and middle atmosphere during a month-long cyclic assimilation experiment.

Contributions

• First preliminary performance assessment of the newly launched EOS-07 MHS satellite instrument.
• Development and validation of a random forest-based algorithm for specific humidity profile retrieval from EOS-07 MHS observations.
• Demonstration of the positive impact of assimilating EOS-07 MHS data on atmospheric analyses and forecasts using the WRF model.

Funding

Not specified in the provided text.

Citation

@article{Mishra2025Assessment,
  author = {Mishra, Manoj Kumar and Gangwar, Rishi Kumar and Shukla, Munn Vinayak and Kumar, Prashant and Thapliyal, P. K.},
  title = {Assessment of EOS-07 MHS satellite observations and retrieval of specific humidity profiles using a random forest-based algorithm},
  journal = {Remote Sensing of Environment},
  year = {2025},
  doi = {10.1016/j.rse.2025.115066},
  url = {https://doi.org/10.1016/j.rse.2025.115066}
}