Scientists use new methods to better forecast atmospheric rivers

Increased funding and scientific interest initiate launch of new forecasting tools

A remote sensing image of an atmospheric river. IVT stands for Integrated Vapor Transport. Kg/m/s stands for kilogram-meter per second - graphic: Julie Kalansky, Ph.D., California-Nevada Applications Program, Center for Western Weather and Water Extremes, Scripps Institution of Oceanography, UCSD

Earlier this year, the Center for Western Weather and Water Extremes (CW3E) launched a new sub-seasonal to seasonal (S2S) forecast product to better predict the influence atmospheric rivers will have on the Western United States. Better and more accurate forecasting tools for atmospheric rivers are critical for a number of community uses, including water management, agriculture, insurance and commodities trading, to name a few.

The demand for better atmospheric forecasting tools has facilitated the development of the new S2S forecasting products launched by CW3E this year.

“There’s a large community of end-users with an increased investment at the federal, state and local level to fund research that will let us understand better the physical processes that impact predictability [of atmospheric rivers] at these longer lead times,” said Mike DeFlorio, the lead researcher of the S2S prediction group at CW3E and the Scripps Institute of Oceanography at the University of California, San Diego. 

The increased interest and funding has enabled CW3E to collaborate with the NASA Jet Propulsion Laboratory. JPL has developed an algorithm model that considers weather, climate and geometric criteria to detect conditions favorable to the presence of atmospheric rivers.

According to DeFlorio, a focus on sub-seasonal to seasonal forecasting for atmospheric rivers has recently gained momentum in the research community. 

“The S2S topic has become a focus in our field really heavily in the last three to four years,” DeFlorio said. “Sub-seasonal to seasonal prediction is focused on [forecasting] lead times from two weeks up to about six months. So the sub-seasonal range is the two week to six week range and the seasonal part is more like three months to six months out.”

DeFlorio said the ability to forecast the presence, or absence, of atmospheric rivers can have meaningful implications for water levels and usage in the Western US.

“Atmospheric rivers are these narrow, elongated regions of horizontal water vapor transport in the atmosphere. They play important roles in the water cycle for many regions around the globe and in the Western US, in particular, the annual variation in precipitation during the winter is dominated by atmospheric rivers. They are a major contributor to the surplus or deficit of water in the West during the wintertime.”

Although atmospheric rivers are often associated with precipitation, DeFlorio said that it’s important to make the distinction that the presence of an atmospheric river does not guarantee rain or snowfall.

“You don’t define an atmospheric river just based on something like precipitation, although they are associated with precipitation. But fundamentally, they’re based on the movement of water vapor horizontally in the atmosphere. So when we make a forecast of atmospheric rivers, we need to define a threshold for what is an atmospheric river and what isn’t.”

Proper water management can be critical to the conservation and distribution of water resources across the Western US. 

“[The West] is a uniquely challenging region for water managers to manage resources. The absence or presence of just a couple of atmospheric river events during a given season can really make or break water supply that is available for the rest of the year. So being able to predict the atmospheric rivers themselves is vital in helping us understand what water supply will look like throughout the rest of the season and the rest of the year.”

In a typical calendar year, DeFlorio said, 10 percent of days in California will see atmospheric river activity. That percentage increases to about 15 percent of days in the Pacific Northwest, with higher values during the winter season. Although atmospheric rivers are perceived to be more influential in Southern California and the Pacific Northwest, they can affect the Great Basin region and Northern Nevada as well. 

“Some [atmospheric rivers] make it to the coastline and then decay, while others are able to penetrate further inland. So it’s not just a coastal phenomenon and it’s not just the phenomenon that we’re worried about in California. Atmospheric rivers are a global phenomenon and they impact many regions, including the Northern Nevada and Sierra regions.”

Flooding in Sierraville, California on February 10, 2017 – Photo – Bob Tregilus, licensed with a CC BY 2.0 license

When atmospheric rivers pass over the elevated terrain of the Sierra Nevadas, they can influence levels of snowfall in the region. 

“So the water supply issue is one angle of how atmospheric rivers play a role,” DeFlorio said. “But then you also have the snowpack issue, which affects runoff and ultimately can affect flooding and other related problems across the region. So they really play a role in many different elements of the water cycle, regionally in the West.”

Other elements of research at CW3E include forecasting what’s known as “ridging events,” in which the circulation of air in the atmosphere can prevent or deviate the path of storm systems and atmospheric rivers from ever reaching certain parts of the Western US. 

“Typically, the presence of a ridge prevents atmospheric rivers or precipitation from occurring in a nearby location or move the airflow so that those events affect other places that are further away. So we have products that focus specifically on ridging as well.”

DeFlorio said that developing accurate forecasting products for atmospheric river activity will play an increasingly larger role in the future, particularly as the scientific community begins to understand how a warming climate might affect their intensity and frequency. 

“There’s a lot of research looking at the projections of atmospheric river activity in the future as we continue entering into a warming climate. The relationship between climate change and atmospheric rivers is an emerging topic that people are interested in.”

The Truckee River overflows on December 10, 2016 – Photo: Bob Tregilus, licensed with a CC BY 2.0 license

DeFlorio said that a number of studies are underway seeking to understand how atmospheric river activity is projected to increase or decrease in the future. Considering atmospheric river’s common correlation with precipitation levels, there is a growing concern that their increased presence or absence could both facilitate more extreme weather events in the Western US. 

“Many studies project an increase in the extreme wet and dry events in the Western US and in California throughout the rest of the 21st century. That would imply that the atmospheric rivers that do impact our region could become more extreme.”

Such extreme events could include the likes of drought, flooding, landslides and even avalanches. 

“Those [extreme events] are associated impacts that we would have to monitor and will monitor closely going through the 21st century as we deal with the impacts of climate change. But those relationships between atmospheric rivers and things like debris flow and avalanches and other impacts have already been observed and we already are diagnosing those relationships,” DeFlorio said.

Consequently, when these new forecasting tools are used on a wide-ranging timescale such as S2S, scientists can better understand their effect from both short-term and long-term perspectives.

“If we think about forecasting atmospheric rivers on weather timescales, so maybe from a day out to 10 days out, atmospheric river forecasts can provide really specific detail on the projected impacts of atmospheric rivers in the water-banking regions on those shorter timescales. For longer timescales, the forecasts that we make for atmospheric rivers provide less specific detail, but they are compared to average conditions so they provide a broader lens to what’s normal. Those longer range forecasts can be used as maybe a heads up and a signal for people that are interested in the region to pay attention.”

With new forecasting tools like the ones at CW3E, DeFlorio and those with a vested interest in how atmospheric rivers may affect water supplies will now have that information available. 

“The S2S products [at CW3E] are a lot newer and we have more that we anticipate will become publicly available after the company research is published in peer-reviewed journals and the products are appropriately vetted by our S2S team. This is possible primarily because there was sustained investment at these various levels in our country to improve the prediction of those models. The demand is there for better forecasts and the investment is being made into the research community and really in tandem with the applications community to make those improvements.”


Scott King writes about science and the environment for the Sierra Nevada Ally. Support his work between now and the end end of the year, and NewsMatch will match your one-time or ongoing donation, dollar-for-dollar.