Home Page
Seattle Public Utilities Home page

Search Climate Change

September 2017

It Will Never Happen Here

Posted: September 11th, 2017

“It will never happen here.”


During Hurricane Harvey, much of metropolitan Houston received as much if not more rainfall than Seattle receives in an entire year. People ask, could we get 50 inches of rain here?  The short answer is no, we could not. But the answer to the follow up—how much rain could we get?—may surprise you.


First, absent runaway global warming, we know that hurricanes, or tropical cyclones, will never reach the Pacific Northwest. The temperature of the ocean off our coast is simply too cold. In order to form, tropical cyclones require sea surface temperatures in excess of 77ºF, and the closest we get to that is in the vicinity of Hawaii.


Tropical cyclones are responsible for every rainfall record between 1- and 7-days around the globe. They essentially lift warm water off the ocean and redistribute it as they spin along. And beyond water temperatures, warmer tropical air begets more water vapor, which usually begets more rainfall.


Topography also enhances rainfall, and mountains obviously boost rainfall amounts in the Northwest, but the key ingredients for extreme, multi-day rainfall have more to do with latitude than altitude. So, if not hurricanes then what is responsible for the still-copious winter rainfall we do get around here?


Our signature extreme event is the atmospheric river (AR). These long, narrow corridors of water vapor are transported northward from warmer regions, often around Hawaii, by large extratropical cyclones. They are responsible for up to half of total annual rainfall throughout western North America, and they are also often called double-edged swords for their ability to simultaneously end droughts and cause floods.


Selected rainfall extremes, each due to ARs:

  • Washington State 24-hour record: 14.26 inches, Mt. Mitchell/South Cascades, November 24, 1986
  • Seattle 24-hour record: 5.61 inches, RG12/Magnolia, December 03, 2007
  • Sea-Tac daily record: 5.02 inches, October 20, 2003


Note that those record numbers don’t come close to what Houston experienced last month. Preliminary data suggest that Harvey’s rainfall was a one-in-a-thousand-year event. Extreme value statistics are notoriously tricky—no one has close a thousand years of reliable data—but there are methods to estimate worst-case scenarios.


The National Weather Service has been taking observations at a handful of regional sites for around 100 years. The federal government last updated precipitation frequency statistics for the Northwest in 1973 (NOAA Atlas 2) and last calculated “probable maximum precipitation” in 1994 (HMR 57). Probable maximum precipitation, or PMP, is “theoretically, the greatest depth of precipitation for a given duration that is physically possible over a given size storm area at a particular geographical location at a certain time of the year.”


Seattle Public Utilities has been measuring rainfall, primarily to
comply with water quality regulations, for 40 years. Its last look at precipitation frequency statistics was in 2003. Since that time the Northwest has seen a dramatic increase in extreme rainfall events, which prompted a review of its statistics that began last year. A final report documenting changes in precipitation at both SPU and NWS rain gages is due by the end of this year, but preliminary results are helping put Hurricane Harvey into perspective.


According to the research, a 1000-year, 24-hour rainfall event in Seattle equals 8.69 inches. Not 50, not 20—only 9 inches. If the 1000-year figure seems too speculative, Seattle’s PMP comes in at 12 inches. Either way, that’s substantially less than what Houston and other places have recently seen. But what about the mountains where rainfall amounts are enhanced?


Seattle Public Utilities’ Cedar River and South Fork Tolt Watersheds have extreme values of 2- to 2-and-a-half times Seattle’s, which put their theoretical rainfall maxima between 20- and 25-inches. Even above the iconically wet rain forests of the Olympic Mountains, PMP values top out at only around 40 inches. “Only.”


In terms of rainfall, it is safe to say that Harvey will never happen here. In terms of flooding, however, we cannot be so smug. The impacts associated with 1000-year rainfall event would represent a catastrophe for regional river valleys and cities.


We’ve caught glimpses of such devastation in recent years. In January 2005, an AR stalled, dropped 13.46 inches in two days at Port Renfrew on Vancouver Island, and caused mudslides throughout British Columbia. In November 2006, a large AR dumped two feet of rain in three days over parts of coastal Oregon and damaged Mt. Rainier National Park. In December 2007, an AR flooded Western Washington, shutting down I-5 in Chehalis, in addition to inundating some Seattle neighborhoods.


Atmospheric rivers typically last a day or two before moving on. Occasionally they stall and cause impacts like those described above. Then there’s another, scarier scenario that researchers have been working to understand in recent years.


In late 1861 into early 1862, California was hit by its worst series of storms on record, flooding the old inland sea that is the Sacramento Delta. The events were reviewed by a coalition of scientists a few years ago and the result, published in 2011 by the U.S. Geological Survey, was called ARkStorm. Meaning “atmospheric river (AR) 1000 (k),” the mega-storm scenario represents the most likely way in which maximum precipitation could be achieved on the West Coast, including the Northwest.


Imagine if one of the locally devastating storms from 2005-07 had stalled for seven days instead of three. That’s essentially how Seattle could reach 12 inches, and our watersheds could get up to 25 inches of rain.


The topic of stalling is a hot one among climate scientists. It happens to be the main reason the Gulf Coast received as much rainfall as it did during Harvey. As this series continues we’ll try to explain why stalling may be on the increase, and we’ll detail the implications not only for rainfall but other local phenomena.


Another future topic will be land use and topography. Houston’s lack of both elevation and land use controls were factors in how the rainfall behaved after it hit the ground. Seattle Public Utilities is constantly looking at ways to mitigate urban and riparian flooding and we can certainly learn from Harvey.


WA State Records
NOAA Precipitation Frequency Data Server (including NOAA Atlas 2 and HMR57)
Flooding in Western Washington: The Connection to Atmospheric Rivers
Seattle Times, “Preparing a city for the worst”