Shehaj et al. (2025) A feasibility study to reconstruct atmospheric rivers using space- and ground-based GNSS observations

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

Identification

• Journal: Repository for Publications and Research Data (ETH Zurich)
• Year: 2025
• Date: 2025-11-18
• Authors: Shehaj, Endrit, Leroy, Stephen, Cahoy, Kerri, Chew, Juliana, Soja, Benedikt
• DOI: 10.3929/ethz-c-000788286

Research Groups

Not explicitly mentioned in the provided text.

Short Summary

This study investigates the minimum number of Low Earth Orbit (LEO) Global Navigation Satellite Systems (GNSS) radio occultation (RO) satellites required to accurately characterize the morphology of Atmospheric Rivers (ARs) using machine learning. It concludes that at least 36 satellites are needed for refractivity mapping and 48 for continuous column-integrated water vapor (IWV) mapping, with ground-based GNSS data significantly improving land-based AR mapping.

Objective

• To determine the number of LEO RO satellites needed to quantify the structure, location, and timing of Atmospheric Rivers (ARs).

Study Configuration

• Spatial Scale: Global (for constellation design), Pacific basin (for AR morphology characterization), land, and ocean.
• Temporal Scale: Hourly RO counts, up to several days (AR duration), 12-hour forecasts.

Methodology and Data

• Models used: Neural Networks (NNs), Machine Learning (ML) models, European Centre for Medium-range Weather Forecasts (ECMWF) system (for forecasts and reference data).
• Data sources: Existing GNSS radio occultation (RO) observations (e.g., COSMIC-2, METOP), simulated RO observations from Walker constellations, 12-hour forecasts from the ECMWF system, and ground-based (GB) GNSS for column-integrated water vapor (IWV) over land.

Main Results

• Optimal LEO RO constellations for AR monitoring consist of 3 or 6 orbital planes with inclinations between 85 and 90 degrees.
• At least 36 RO satellites are required to characterize the morphology of ARs in the Pacific basin with useful precision and accuracy when mapping refractivity at 2 km geopotential height using ML.
• A constellation of 48 RO satellites is needed to continuously and accurately map column-integrated water vapor (IWV) fields, thereby reconstructing AR morphology with useful precision and accuracy, using a two-consecutive-NN framework.
• Mapping refractivity into IWV using RO is less accurate over land than over oceans.
• Incorporating IWV from ground-based GNSS significantly improves IWV fields and AR path/shape mapping over land due to its higher spatial and temporal resolutions.

Contributions

• Quantifies the minimum number of LEO RO satellites necessary for effective and continuous monitoring and characterization of Atmospheric River morphology.
• Identifies optimal orbital configurations (number of planes and inclinations) for future RO satellite constellations aimed at AR monitoring.
• Develops and validates a machine learning framework to map sparse RO observations to high-resolution AR structures (refractivity and IWV fields).
• Highlights the complementary role of ground-based GNSS data in enhancing the accuracy of AR mapping, particularly over land areas.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Shehaj2025feasibility,
  author = {Shehaj, Endrit and Leroy, Stephen and Cahoy, Kerri and Chew, Juliana and Soja, Benedikt},
  title = {A feasibility study to reconstruct atmospheric rivers using space- and ground-based GNSS observations},
  journal = {Repository for Publications and Research Data (ETH Zurich)},
  year = {2025},
  doi = {10.3929/ethz-c-000788286},
  url = {https://doi.org/10.3929/ethz-c-000788286}
}