Skip to main content
U.S. flag

An official website of the United States government

Publication Date
29 June 2021

Constraining and Characterizing the Size of Atmospheric Rivers: A Perspective Independent From the Detection Algorithm

Development of five independent atmospheric river size estimation methods.
Print / PDF
Powerpoint Slide

Five novel methods for estimation of Atmospheric Rivers (AR) were developed in this work. In particular, these methods estimate the AR size independently of any detection algorithm. We managed to reduce the uncertainty and range in size estimation compared with the current existing methods from the Atmospheric River Tracking Method Intercomparison Project (ARTMIP). We found that the AR area is between 1012 to 1013 m2 (between 1 and 11 areas of California).


This work created five independent methods that provide a narrower range of AR sizes than the currently existing methods. AR boundaries and size definition are still an open research question, but our approach provides an objective way to estimate their size that could help to constraint new detection algorithms. Also, it could provide an objective way to estimate how AR size and geometry would change under different climate change scenarios.


LBL scientists developed five different Atmospheric River (AR) size estimation methods that are independent of the AR detection algorithm. ARs have positive and negative impacts over the regions where they make landfall, particularly for West Coast US water resources. AR size (length, width, and area) is an important characteristic that needs to be studied, and could directly relate to the impacts of AR over land. This work found that North American landfalling ARs originated in the North Pacific have areas between 7x1011m2 and 1013 m2 (between 1 and 11 times the area of California), and their lengths are on average four times their widths. LBL scientists also found that the ARs originating from the Northwest Pacific (WP) (100E-180E) are bigger and more parallel to the equator than those from the Northeast Pacific (EP) (180E-240E). The methods in this work provide a narrower range of size estimation than the current methods and could be used to constrain current and future AR detection methods, and they could be used in future studies to understand how AR size could vary under climate change scenarios.

Point of Contact
William D. Collins
Lawrence Berkeley National Laboratory (LBNL)
Funding Program Area(s)