Kerns et al. (2025) Global Atmospheric River Lifecycle Detection Using Integrated Water Vapor and Vapor Transport

Identification

• Journal: Scientific Data
• Year: 2025
• Date: 2025-12-08
• Authors: Brandon W. Kerns, Shuyi S. Chen
• DOI: 10.1038/s41597-025-06198-9

Research Groups

University of Washington, Department of Atmospheric and Climate Science, Seattle, USA

Short Summary

This study introduces the global Atmospheric River Lifecycle Detection (ARLiD) method and a 44-year database that uniquely incorporates both Integrated Water Vapor Transport (IVT) and Total Precipitable Water (TPW) to detect and track atmospheric rivers throughout their entire global life cycles. This dual-variable approach enhances consistency in AR identification, particularly in phases where TPW is a stronger signature than IVT.

Objective

• To develop a global, lifecycle-tracking method for Atmospheric Rivers (ARs) that integrates both Integrated Water Vapor Transport (IVT) and Total Precipitable Water (TPW) for improved detection consistency and to generate a comprehensive, long-term database.

Study Configuration

• Spatial Scale: Global, on a 0.25° latitude/longitude grid.
• Temporal Scale: 44-year period (June 1980 – June 2024), with hourly data resolution. Only AR systems tracked for at least 48 hours are included.

Methodology and Data

• Models used: Atmospheric River Lifecycle Detection (ARLiD) method, which employs a blob detection technique combining IVT and TPW. This includes a Laplacian of Gaussian (LoG) filter, a dilation/erosion method for tropical moisture extent, and a modified Large-Scale Precipitation Tracking (LPT) algorithm for temporal tracking, allowing for system splitting and merging.
• Data sources: ECMWF Reanalysis 5 (ERA-5) hourly data, specifically total column water vapor (TPW) and eastward/northward components of integrated water vapor transport (viwve, viwvn), vertically integrated from the surface to 1 hPa.

Main Results

• A novel global AR detection and tracking method (ARLiD) and a 44-year database (1980-2024) were developed, uniquely integrating both IVT and TPW.
• The dual-variable approach significantly improves consistency in identifying the full AR lifecycle, extending detection equatorward and poleward beyond methods relying solely on IVT.
• AR initiation is primarily observed off the east coasts of major continents (Asia, North America, Australia, South America), while dissipation is favored in the eastern North and South Pacific Oceans.
• Most AR systems and families exhibit durations of less than 1 week, though a notable 7% of AR families and 0.6% of AR systems persist for 3 weeks or longer.
• AR frequency is maximized over oceans at subtropical to polar latitudes (30-60°N/S), with ARLiD showing a lower overall frequency compared to some other ARTMIP products but similar to TempestLR.

Contributions

• Presents the first global, lifecycle-tracking method and dataset for Atmospheric Rivers (ARs) that explicitly incorporates both Integrated Water Vapor Transport (IVT) and Total Precipitable Water (TPW).
• Provides a more consistent and comprehensive identification of ARs throughout their entire lifecycle, including phases where TPW is a stronger signature, and extends detection beyond IVT-only methods.
• Offers a publicly available, long-term (44-year) global database of AR systems and families, including their centroids and spatio-temporal masks, for diverse research applications.

Funding

• NOAA CPO research grant NA21OAR4310263
• DOE’s Regional and Global Modeling and Analysis (RGMA) program (Integrated Coastal Modeling (ICoM) project, contract PNNL/UW PO516639)

Citation

@article{Kerns2025Global,
  author = {Kerns, Brandon W. and Chen, Shuyi S.},
  title = {Global Atmospheric River Lifecycle Detection Using Integrated Water Vapor and Vapor Transport},
  journal = {Scientific Data},
  year = {2025},
  doi = {10.1038/s41597-025-06198-9},
  url = {https://doi.org/10.1038/s41597-025-06198-9}
}