The University of Arizona

Are More Atmospheric Rivers Headed Our Way?

November 25, 2013
Author Profile: 

Rain drenched the Southwest this past weekend, flooding streets and quenching the thirst of parched soils. The reason: an atmospheric river (AR). If you’ll recall from previous blogs (October 7, 2011 and December 9, 2011), ARs are narrow corridors of water vapor transport in the lower atmosphere that can result in extreme precipitation and flooding when they strike land (see figure 1). An AR in December 2010 produced as much as 6 inches of rain in northwest Arizona, and in January 2010 a powerful winter storm (another AR) delivered 34 percent of Phoenix’s total annual precipitation in only six days. ARs are most common in California, however, where “the state spends about $400 million each year repairing flood damage, and about 90 percent of that is because of ARs” (see previous blog).

 Figure 1: Example of an atmospheric river striking the Pacific Northwest on November 7, 2006. This particular event resulted in over 25 inches of rain in just 3 days. Warm colors represent moist air and cool colors represent dry air. The horizontal band of red and purple colors at the bottom of the image is the Intertropical Convergence Zone (ITCZ) – a normally moist area that some of the strongest ARs can “tap” into, as in this case. Image from Ralph et al., (2011).

ARs also deliver a large fraction of Sierra Nevada snowpack, the main source of water in the area. Authors of a recent study in Water Resources Research assessed the influence of ARs on Sierra Nevada snowpack and suggest another way of predicting these intense storms. They focused on a particular winter (2010/2011) in California’s Sierra Nevada that had “the largest snow accumulation, largest number of ARs, and largest AR-related snow accumulation within the analysis period of [water years] 1998-2011.” During this particular winter, there were 20 AR dates, 14 of which occurred in December, well above the normal number of ARs. All of the December ARs plus one in November occurred during negative phases of both the Arctic Oscillation (AO) and Pacific-North American pattern (PNA)—two modes of variability that affect weather across the Northern Hemisphere. Over the entire analysis period, AR frequency was higher when both circulation patterns were negative. Could this make it easier to predict which winters will have a higher number of ARs, and thus be wetter?

According to John J. Brost, Science and Operations Officer with the National Weather Service (NWS) and SWCCN blog contributor, yes! On long timescales, knowing which climate patterns favor AR advancement into the Southwest (or at the very least into the Sierra Nevada) is a very useful skill. This type of information would be especially useful for the Climate Prediction Center when creating their one- and three-month precipitation outlooks.

However, when predicting weather 7 to 10 days in the future, the job of a lot of operational forecasters, this type of information doesn’t add much to their capabilities. According to Brost, “I would look at this just like ENSO [El-Niño Southern Oscillation]. I never base a 7-day forecast on the phase of ENSO. I never say, ‘it’s going to rain in 3 days and part of that reason for the rain is El Niño.’ I do use ENSO for longer term situational awareness, but the forecast models are very good at predicting the sensible weather out to 7 or even 10 days now.”

Forecast models actually do a pretty good job already of predicting ARs. ESRL has produced some useful AR detection products that forecast Integrated Vapor Transport ( Other models, according to Brost, “identify ARs (with some skill) out to 192 hours [8 days]. The models are much better in the shorter term of course, but we do have some skill in detecting ARs in the 7-day time frame. We (the NWS) produce a 7-day forecast so this is just right for us.”

Looking at the Climate Prediction Center’s site, Brost “noticed we are currently (and have been for most of November) in a positive AO phase, with a negative PNA phase.” So by this metric, based on the new study, we wouldn’t expect many ARs this winter, yet here we have just experienced one. But as Brost said, these climate patterns are just that, and are thus better at predicting AR variability over longer periods of time. Who knows, maybe the AO will switch to a negative phase (which is feasible if you look at past variability), creating the right conditions for a wet winter, which will definitely be welcome here in the dehydrated Southwest.


*Guan, B., N. P. Molotch, D. E. Waliser, E. J. Fetzer, and P. J. Neiman (2013), The 2010/2011 snow season in California’s Sierra Nevada: Role of atmospheric rivers and modes of large-scale variability, Water Resour. Res., 49, doi:10.1002/wrcr.20537.

*Ralph, F. M., P. J. Neiman, G. N. Kiladis, K. Weickman, and D. W. Reynolds, 2011:  A multi-scale observational case study of a Pacific atmospheric river exhibiting tropical-extratropical connections and a mesoscale frontal wave.  Mon. Wea. Rev.139, pp. 1169-1189, doi: 10.1175/2010MWR3596.1.