Tang et al. (2026) Numerical Simulation of a Heavy Rainfall Event in Sichuan Using CMONOC Data Assimilation

Identification

• Journal: Remote Sensing
• Year: 2026
• Date: 2026-04-10
• Authors: Xu Tang, Cheng Zhang, Aiming Wu, Rui Sun, J. Liu
• DOI: 10.3390/rs18081126

Research Groups

• School of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology, Nanjing, China
• College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing, China

Short Summary

This study demonstrates that assimilating CMONOC GNSS tropospheric products (Zenith Total Delay/Precipitable Water Vapor) into the WRF model significantly improves the simulation of heavy rainfall events over the complex terrain of the Sichuan Basin by enhancing initial moisture and low-level convergence.

Objective

• Evaluate the impact of assimilating CMONOC GNSS tropospheric products (ZTD/PWV) on the numerical simulation of heavy rainfall events over the Sichuan Basin.
• Determine the extent to which high-frequency CMONOC GNSS ZTD/PWV assimilation corrects initial moisture and dynamic biases over the complex terrain of the Sichuan Basin.
• Investigate how these initial condition adjustments physically propagate to improve the simulation of heavy rainband location, intensity, and vertical structure.

Study Configuration

• Spatial Scale: Two-way nested domains: outer domain at 27 km resolution covering Southwest China, inner domain at 9 km resolution focusing on the Sichuan Basin. Model center at 104.0°E, 30.0°N. Model top at 50 hPa.
• Temporal Scale: Primary event simulation from 00:00 UTC on 10 August to 00:00 UTC on 13 August 2020 (72 hours). An additional robustness test for 21–23 August 2021. Data assimilation performed using a 6-hour cycling strategy.

Methodology and Data

• Models used:
  • Weather Research and Forecasting (WRF) model (Version 4.2)
  • WRF Data Assimilation (WRFDA) three-dimensional variational (3DVar) system
  • Physical parameterization schemes: WSM6 microphysics, RRTM longwave and Dudhia shortwave radiation, YSU planetary boundary layer, Noah land surface model, Kain–Fritsch cumulus parameterization.
• Data sources:
  • Assimilated: Crustal Movement Observation Network of China (CMONOC) GNSS tropospheric products (Zenith Total Delay (ZTD) and Precipitable Water Vapor (PWV)) from 23 stations, hourly.
  • Initial and Boundary Conditions: NCEP Final (FNL) Analysis reanalysis (0.25° × 0.25° spatial resolution, 6-hour temporal resolution).
  • Verification:
    • Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (GPM) (IMERG) Final Run (V06B) (0.1° × 0.1° spatial resolution, 30-minute temporal resolution).
    • Upper-air sounding temperature observations at Chongqing (Shapingba).

Main Results

• The Data Assimilation (DA) experiment improved the placement of the primary rainband and the depiction of peak rainfall compared to the Control (CTRL) experiment.
• For the 10 August 2020 event, the observed rainfall core (approximately 40 mm) was underestimated in CTRL (approximately 15 mm) but strengthened in DA (approximately 25 mm).
• Hourly Threat Score (TS) at a 2 mm h⁻¹ precipitation threshold showed DA achieved a higher maximum TS (0.292) than CTRL (0.250), with the largest instantaneous gain reaching 0.061.
• For 72-hour accumulated precipitation, DA yielded higher TS values across all examined thresholds (≥10, ≥25, ≥50, and ≥100 mm), with the most pronounced improvement for heavier rainfall categories.
• Diagnostic analysis revealed that GNSS assimilation introduced positive specific humidity increments (moistening) and negative divergence increments (strengthened convergence) at 850 hPa in the key rainfall region.
• DA produced a more organized and better-positioned vertical ascent core, with improved vertical coherence, compared to CTRL.
• DA reduced the Root Mean Square Error (RMSE) of temperature profiles against radiosonde observations by an average of 15.2% (from 3.68 °C in CTRL to 3.12 °C in DA).
• An additional heavy-rainfall event (21–23 August 2021) showed consistent improvements in precipitation distribution and higher TS values after GNSS assimilation, confirming the robustness of the findings.

Contributions

• Provides targeted, case-based evidence for the beneficial impact of assimilating CMONOC GNSS ZTD/PWV on heavy rainfall simulation over the complex terrain of the Sichuan Basin, addressing a documented research gap.
• Quantifies the improvements in rainband location, intensity, and temporal evolution due to GNSS data assimilation.
• Diagnoses the physical mechanisms (enhanced low-level moisture, strengthened convergence, and better-aligned vertical ascent) responsible for the improved precipitation simulations.
• Suggests that dense ground-based GNSS networks can serve as an effective additional moisture constraint for regional numerical weather prediction, particularly in complex terrain.

Funding

• National Key R&D Program of China (grant 2023YFE0208400)
• Open Fund of Jiangsu Research Center for Underground and Tunnel Engineering Technology (grant 2023SDJJ02)

Citation

@article{Tang2026Numerical,
  author = {Tang, Xu and Zhang, Cheng and Wu, Aiming and Sun, Rui and Liu, J.},
  title = {Numerical Simulation of a Heavy Rainfall Event in Sichuan Using CMONOC Data Assimilation},
  journal = {Remote Sensing},
  year = {2026},
  doi = {10.3390/rs18081126},
  url = {https://doi.org/10.3390/rs18081126}
}