FEATURE: Predicting and planning for extreme precipitation from atmospheric rivers

On Feb. 27, 2019, an atmospheric river 350 miles wide and 1,600 miles long barreled through the sky, funneling moisture from the tropics to Sonoma County and dumping over 21 inches of rain over two days, causing the Russian River to crest at 13.4 feet above flood stage, flooding the town of Guerneville.  Damages were estimated at well over US$100 million.

The phenomenon is not new.  Formerly called a ‘Pineapple Express’, by the late 1990s, scientists discovered that much of the world’s moisture was transported from the tropics to higher latitudes by similar systems, which became known as ‘atmospheric rivers.’  These rivers in the sky can sometimes be the larger than any on earth, capable of transporting 15 times the volume of the Mississippi River.

Atmospheric rivers are doubled edged swords.  In dry conditions, atmospheric rivers can replenish water supplies and quench dangerous wildfires; in wet conditions, they can cause damaging floods and debris flows, and cause damages that can cost hundreds of millions of dollars.  Their presence or absence can make the difference between drought, a normal healthy water supply year, or floods. 

To address the challenges atmospheric rivers present, the Center for Western Weather and Water Extremes (CW3E) was established at the Scripps Institute of Oceanography to advance scientific understanding of atmospheric rivers and their role in extreme events and to improve forecasting capabilities.  Earlier this year, a webinar hosted by the Scripps Corporate Alliance highlighted the Center’s accomplishments.

Scripps Institute of Oceanography is a division of UC San Diego with the stated mission ‘to understand and protect the planet and investigate the oceans, earth, and atmosphere to find solutions to our greatest environmental challenges.’  There are 2300 students, researchers, volunteers, and staff who produce approximately 700 research publications per year and receive approximately $159 million in sponsored research annually.  Among their alumni and staff are 27 Nobel Prize, National Academies, Royal Societies, and National Medal Science winners. 

The webinar presenter was Dr. Anna Wilson, a field research manager for the Center for Western Weather and Water Extremes at Scripps Institute of Oceanography where she works on supporting the development of physically-based accurate representations of atmospheric rivers and other extreme events in forecasts and projections through integration of in situ and remote sensing observations.  In this presentation, she discussed what atmospheric rivers and their importance to the Western US, discussed some of the atmospheric river forecasting and decision support tools that have been developed, and their efforts to improve monitoring of atmospheric rivers.


Dr. Wilson began with some background on the Center for Western Weather and Water Extremes (or CW3E).  The Center was launched by Dr. Marty Ralph at the Scripps Institute of Oceanography at UC San Diego in 2014 with the goal to better understand, predict, and apply extreme weather forecasts and capabilities that are specifically tailored to the unique meteorological conditions of the Western United States.  For California, Washington, Oregon, this is largely driven by atmospheric rivers that mostly come during in the fall and winter months which can penetrate significantly inland even as far as Colorado.

The Center’s mission is to provide 21st Century water cycle science, technology and outreach to support effective policies and practices that address the impacts of extreme weather and water events on the environment, people and the economy of the Western US by revolutionizing the physical understanding, observations, weather predictions and climate projections of extreme events, including atmospheric rivers and the North American summer monsoon as well as their impacts on floods, droughts, hydropower, ecosystems and the economy.

The Center includes a diverse interdisciplinary team of more than 40 staff, graduate students, and Post Docs; they also were with a number of external CW3E-funded collaborators to accelerate the pace of atmospheric river discovery and to apply atmospheric river science to inform water resource decision making.

Having this team is really crucial to achieve our objective and getting as many experts as we can to be working on these problems in order to have the best outcomes possible,” said Dr. Wilson.


Last year, the Center developed a strategic plan with input from key partners and sponsors that identified six priority areas to focus on over the next five years:

Monitoring and Projections of Climate Variability and Change: The goal is to develop a comprehensive understanding of extreme precipitation in the west to inform both current and future resource and risk management. 

Emerging Technologies: Emerging technologies underpins many of the priority areas, but was felt important enough to list as a priority area.  They want to be building expertise in all types of innovative methodologies, such as artificial intelligence, machine learning, and other things in order to add value specifically to weather prediction.

Subseasonal to seasonal predictions of extreme weather:  This is important to any resource manager whose planning is dependent upon knowing how much water will be incoming or not incoming.  The goal is to improve understanding of the predictability of atmospheric rivers on this timescale for more than a week to a few months ahead, and also to become a regional leader in producing outlooks of atmospheric river activity on these timescales over the Western United States.

Modelling capabilities for the Western US:  The Center wants to develop and contribute to state of the art weather hydrology and a coupled modeling system at space and time scales relevant to water resource and emergency management decisions.

Forecast Informed Reservoir Operations:  The Center is bringing together diverse stakeholders including water managers, engineers, environmental consultants that are all working towards a goal of enabling more effective reservoir management through improved weather and water forecasts. 

Atmospheric River Research and Applications:  The Center will be working toward effective, efficient, integrated monitoring of the state of the atmosphere, ocean surface, and land surface and subsurface to support decision makers and stakeholders, and doing this by answering key research questions that will increase forecasting accuracy for high impact events at all relevant lead times.  They will do this by enhancing monitoring and by integrating observations, theory, and modeling into decision support.

The key objectives within the area of atmospheric river research and applications are to further understanding of atmospheric river dynamics; produce forecasting and decision support tools that meet the needs of Western US forecasters, resource managers, and emergency managers; and enhance global atmospheric river modeling through a transformative modernization of atmospheric river measurements both over the Pacific and in the Western US.  The goals of this priority area are to bolster economic strength, resilient water supply, preparedness for emergencies, and the health of ecosystems and communities in the region.

Producing forecasted decision support tools that meet the needs of forecasters and managers requires the input and collaboration of those forecasters and managers, so the partnerships that we develop in that sense are really crucial to the work that we are doing,” said Dr. Wilson.


Atmospheric rivers are relatively long, narrow corridors with a lot of moisture and strong winds that bring water from the tropics to the poles and frequently affect the West Coast in the mid-latitudes.  The graphics on the slide at the bottom left show an atmospheric river making contact with the California coast on February 7, 2017, the day that Oroville Dam’s main spillway was damaged.  And in February of 2019, a series of atmospheric rivers caused the worst flooding in almost 20 years in Northern California in Guerneville. 

Atmospheric rivers cause floods but they also provide water supply,” said Dr. Wilson, noting that the images how the AR storm caused flooding in the town of Sebastopol, as well as filled Lake Mendocino in that same month.  “So the information that we are providing on atmospheric rivers is really important for a number of different industries we work with.  Our federal sponsors, the US Army Corps of Engineers, the state Department of Water Resources, and other regional water agencies want to know what they should be expecting from these atmospheric rivers.  How many, how strong, how much precipitation and how can they be best prepared for them?

Atmospheric rivers are an emerging topic and were highlighted in the Fourth National Climate Assessment, along with hurricanes, severe thunderstorms, and winter storms.  They are now getting widespread attention as an essential phenomenon to understand for the Western US for both being prepared for extreme weather and for the water supply benefits they provide as their presence or absence can make the difference between drought, a normal healthy water supply year, or floods. 

The first step in order to be able to take meaningful action is to have an understanding of what is coming, so the Center is working to advance the state of science to better provide that knowledge.  Over the past decade, about $50 million has been invested at both the federal, state, and local level, which has produced major advances. 

For example, emergency managers might care about work that shows the atmospheric rivers are usually behind levee breaks that can cause really large catastrophes,” said Dr. Wilson.  “Water managers might care about the fact that atmospheric rivers are responsible for filling most of California’s reservoirs, and also responsible for busting droughts, and that’s something that might be important for fire response and other sectors.  Environmental managers might be interested in research on how atmospheric rivers affect fish populations.”

We’ve also made advances in monitoring and predicting, but there is a lot more to do, so continued investment is really essential to be able to really understand and get the most out of what the science can provide for policy making and support.”

One of those challenges is that forecasting large amounts of precipitation is difficult.  The plot shows the frequency of occurrence of different amounts of precipitation; the lighter colored bars are showing greater than 3 inches of precipitation in one day and the darker bars are show greater than 5 inches.  On the x axis is the number of observed occurrences and the y axis is the prediction 24 hours out, 2 days out, and then 3 days out. 

What you can see this really degrades, even at 1 day out, and then again at 3 days, so we can get a lot better at this,” she said.  “The other thing to know is that of the days that have greater than 3 inches of precipitation in one day, most of them are associated with atmospheric river events.”

Another challenge is improving the forecasting of atmospheric landfall positions.  On the plot, the x axis is the distance error of atmospheric river landfall position, the y axis is the forecast lead time in days, and the lines are all representing state of the art, local numerical weather prediction models.

The takeaway here is on average, we still have forecast errors on a five day lead time on an order of 100s of miles, and that’s a position error that’s larger than watersheds,” said Dr. Wilson.  “We are able to see when an atmospheric river is coming and we’re getting better at monitoring and predicting them in some sense, but we really have room to grow in terms of exactly when is an atmospheric river going to make landfall on the west coast, where exactly will it make landfall, and how long will it last.”

The Center for Western Weather and Water Extremes aspires to be a hub for researchers working on these problems, so they hold conferences, such as the biennial International Atmospheric Rivers conference lasted hosted in 2018 and coming up in October 2020. 

Not only do we prioritize bringing together scientists and researchers who are looking at atmospheric rivers from all different angles, but also stakeholders and decision makers like engineers, forecasters, and managers, to discuss atmospheric river science and applications together,” said Dr. Wilson.  “How are people using information about atmospheric rivers and what else might they need?  How can we make our science more actionable? These conferences have oral sessions and poster sessions, but it’s also integrated into the design to have really interactive sessions such as breakout sessions and panel discussions so we can better build relationships between scientists and stakeholders, understand each other, understand each other’s perspectives and needs, so we can coproduce applications of this science.”

An economist at the Center researched insured economic flood losses, and found that for the eleven Western states on average, 84% of those losses were caused by atmospheric rivers and on the West Coast, almost all of the losses are associated with atmospheric rivers.  Atmospheric rivers can also be deadly; the image on the right on the slide at the lower left shows a post-fire debris flow that happened as a result of an atmospheric river near Montecito, California in January of 2018 that killed over 20 people.

In order to communicate the potential impacts of a given atmospheric river based on its strength, the director of the Center Dr. Marty Ralph, along with partners in the National Weather Service, the Department of Water Resources, the Corps of Engineers, and broadcast meteorologists came up with a scale for atmospheric rivers.   The scale reflects the intensity, the amount of moisture and the strength of the winds along the corridor of the atmospheric river, which is known as the Integrated Vapor Transport, or IVT.  It is said to be integrated because it encompasses all of the water and winds from the bottom of the atmosphere up to the top of the troposphere. 

The AR category depends not only the intensity, but also on the duration of the atmospheric river; the longer an atmospheric river lasts, it might be moved up a category because the impacts will be stronger if the atmospheric river is longer.

In addition, the impacts of atmospheric rivers were assessed, and ARs 1 and 2 are generally beneficial more than hazardous as they replenish water supplies.  Category 3 ARs are a mix of benefits and hazardous impacts.  Categories 4 and 5 are mostly harmful and can cause significant flood damages.

In terms of economic impacts, the damages increase by an order of magnitude along with the AR categories of weak, moderate, strong, extreme, and exceptional, so Dr. Wilson said it does align well with the scale that was developed.  She acknowledged the variabilities, noting that the outliers, for example, tend to happen when there are antecedent watershed conditions before the storm comes that may include saturated soils and already full reservoirs, so the conditions on the ground when the atmospheric rivers are hitting is also very important.


The Center in conjunction with their partners and decision makers has developed a multitude of tools that are freely available on their website and updated frequently as new forecasts come out.  Those tools include monitoring of current conditions, interactive maps, and forecasting tools that provide weather analysis and forecast data on a range of spatial and geographic scales.  The tools allow users to display and interact with numerous variables from a single point to a watershed scale with the goal of providing insight into the potential impacts of atmospheric rivers. 

Dr. Wilson then went into detail on one of the available tools, the Atmospheric River Landfall Tool which was developed with the Army Corps of Engineers and is being used in a pilot project for the Forecast Informed Reservoir Operations at Lake Mendocino.

On the graphic on the left hand slide of the slide, the y axis is the latitude along the West Coast, which corresponds to each of the black dots shown in the panel on the map on the right.  On the X axis is the prediction time; the far right side is current time and moving towards the left is going out in prediction time to 16 days; and the colors are showing the certainty.  In the example on the slide, the atmospheric river is a baseline category of AR1.

You can change this to look at whatever strength of an atmospheric river that you want to, and you can also look at how the forecast has changed from prior days as well,” explained Dr. Wilson.  “The darker the color is, the more certain it is that we’ll have an atmospheric river of that strength making landfall.  So the darker colors are really certain, and as we go down into the yellows and greens, it is less certain.  And then the bars on the right are also showing a sense of how long will the atmospheric rivers will last at a point; the longer the bar, the longer the conditions of atmospheric rivers at that particular location.”

This tool was developed along with partners in order to provide the most information in a way that was easy to use and uses the best available science,” she added.


Another really key aspect of atmospheric river research and applications is monitoring.  The Center is working to enhance global atmospheric river monitoring through a transformative modernization of atmospheric measurements, both over the Pacific and the Western US.  On the slide, the picture on the left is one of the meteorology stations that was installed in the Yuba River basin, and on the right, the deployment of buoys on the Pacific Ocean that also measure surface pressure.

The Center for Western Weather and Water Extremes is leading collection, processing, and dissemination of observations of atmospheric rivers and their impacts,” said Dr. Wilson.  “The void of data over the Pacific Ocean has been a consistent challenge, but it’s one that we’re filling through targeted observations at the ocean surface and the atmosphere.”

The Center’s location at Scripps Institute of Oceanography is the opportunity to work with other oceanographers who are also trying to improve the state of monitoring over the Pacific Ocean, said Dr, Wilson, noting that the earlier picture of the launch of the buoys was the result of a collaboration between the Center and the Global Drifter Program located at Scripps. 

In addition to observations over the ocean, intensive monitoring after the atmospheric river makes landfall is needed to provide data to evaluate the atmospheric river as it interacts with the land, as well as hydrologic observations of streams and soils before landfall because those conditions modulate the impacts. 

We really need a well-developed network that’s holistic and over the whole scope of the Pacific and the Western US in order to observe the full water cycle and improve the prediction of weather and water extremes,” said Dr. Wilson.  “So we are then trying to improve observations in space and time in a targeted manner with only the most difference to ensure that those data are easily accessible for partners and other users and also to ensure that our observations are suitable for assimilation into weather models.”

The Center wants to continue to facilitate collaborations through the various groups that they have been involved with.  One is the Russian River Hydrological Data Collection and Modeling Group, which is an effort with researchers working in the watershed which is very densely instrumented and trying to put that together to leverage all of the work that has been done.  Also the Atmospheric River Reconnaissance Modeling and Data Assimilation Steering Committee which collaborates on enhancing monitoring efforts.

In terms of enhanced monitoring of atmospheric rivers, there are several efforts underway, one of those is atmospheric river reconnaissance, which includes things like the buoys on the ocean surface and technology obtaining temperature and moisture profiles from GPS on board aircraft, as well as dropsondes, which is an expendable weather reconnaissance device created by the National Center for Atmospheric Research which is designed to be dropped from an aircraft at altitude over water to measure storm conditions as the device falls to the surface..

Atmospheric river reconnaissance supports western storm predictions and water decisions by developing and testing targeted airborne and buoy observations over the northeast Pacific to improve the forecasting lead times of 0 to 5 days for the landfall and impacts of atmospheric rivers over the US West Coast.  They are pursuing these through collaborative cross-disciplinary science-based strategies using partnerships with NOAA, the Air Force, global modeling centers include the Naval Research Laboratory, the European Center for Media and Range Weather Forecasts, the American National Center for Environmental Prediction, and research centers including the National Center for Atmospheric Research.  Atmospheric river reconnaissance activities are sponsored by the California Department of Water Resources.

Dr. Wilson noted that although they have satellite observations over the ocean, satellites have difficulty being able to see through the high cloud cover, so these reconnaissance observations are crucial for filling the void.  The image on the slide is an example of how they might sample an atmospheric river working with the Hurricane Hunters and the both the NOAA G4 and the Air Force C130.

We have three aircraft at once flying into these storms and we take a number of transects of the atmospheric river and also additional areas around those that might interact with the atmospheric river and cause different landfall characteristics,” said Dr. Wilson.


I hope that I have showed you well how our different collaborative efforts that try to integrate data collection, analysis, and modeling to provide science based improved decision support for the western US,” Dr. Wilson said. “We are trying to advance forecasting, advance decision support tools, and monitoring efforts that support directly the needs of western US forecasters, resource managers, and emergency managers.  Our partnerships supporting this goal are really key for water supply resiliency, mitigation of flood risk, health of ecosystems and communities, our economic strength and its preparedness for emergencies in the Western US.”

We are grateful to the partners that we work with now for helping us to do science that makes a difference and is useful and actionable for them and we look forward to hopefully continuing that,” she concluded.

Question: Can you discuss climate change data that will heat ocean temperature and CO2 concentrations in the context of atmospheric rivers?

Answer:That’s another key priority area for us is looking at atmospheric rivers and monitoring of climate variability and change.  While we do have certainty on things such as the strongest atmospheric rivers look like they will get increasingly strong and the weaker atmospheric rivers may become less frequent, so we might have stronger, longer lasting atmospheric rivers and then fewer precipitation days that aren’t involved with strong atmospheric rivers.  The other thing is that we will expect to see increasing variability year to year, so swings between very, very wet years and then sequences of very, very dry years.”

The other thing I want to mention around climate change is just that the rain/snow line is rising, so what we’re going to likely see is more atmospheric rivers that bring rain rather than snow, and that’s something that will require a lot of work on how we can most effectively manage our existing infrastructure that was really built to take advantage of the natural reservoir of the snowpack in many cases.”

Print Friendly, PDF & Email