See previous NWS assessments

More info:

The National Weather Service is most often in the business of talking about what's coming. But this week, it has a team in Boulder exploring instead the question: What happened?

Specifically, the NWS wants to evaluate how its forecasters performed in the days leading up to, and during, the historic record-shattering rains and flooding that ravaged the Front Range in September.

A team of about 10 specialists brought in from around the country is conducting on-site interviews this week with people from a wide range of disciplines and professions -- including this reporter -- to evaluate its staff's forecasting of, and response to, one of the most extreme weather events ever recorded in the Boulder area.

The purpose of the NWS "service assessment," according to a statement from the NWS's David Vallee, is "to evaluate the performance of NOAA's NWS offices affected by the event. It is designed to (a) identify and share best-case operations, procedures, and practices, and (b) address service deficiencies. Partner engagement is a key component to help us identify services that worked and deficiencies to improve."


Vallee is one of two officials heading up the assessment team. He is the hydrologist in charge at the NWS Northeast River Forecast Center in Taunton, Mass.

There was a widespread acknowledgment, during and after the storm, that no one had predicted a storm of the magnitude Boulder witnessed, in the days that preceded it.

For example, Boulder meteorologist Matt Kelsch had told the Daily Camera at the storm's peak, "If you had had asked me (in the previous week), I would have told you it looked like a wetter, cooler period (ahead). But there is no way I would have told you we were going to have 10 inches of rain."

Boulder registered 17.15 inches of rain from Sept. 9 to Sept. 16, including 9.08 inches in one 24-hour period.

Nezette Rydell, NWS Boulder meteorologist in charge, said the review launched this week "is not routine," and can be a lengthy process -- for example, the agency's 66-page assessment from Hurricane/Post-Tropical Cyclone Sandy, which struck the Northeast in late October 2012, was only recently completed.

"It can take from three to six months, from start to finish," said Rydell. "They will write a report. There will be best practices that are shared, and there will be findings, equipment that didn't work, processes that could benefit from changing."

The team's review also has a "social science" component, Rydell said, examining how information conveyed by forecasters affected, or was utilized by its recipients.

"Sometimes we will find out that a county 100 miles away did something, takes some action based on our forecast, and we'll say, 'Oh no, that's not what we meant,' " Rydell said. "We think we know what we are saying, but, we're looking at how other people respond."

Not every member of the team is from the NWS, or its parent agency the National Oceanic and Atmospheric Administration. For example, joining Vallee in leading the analysis of the Colorado flood is Robert Kimbrough, associate director of the United States Geological Survey's Colorado Water Science Center.

Kimbrough is the official who first reported to the Camera that on the morning of Sept. 12, a gauge on Boulder Creek at North 75th Street showed the streamflow to have reached a level with a one-in-100 chance of occurring in any given year -- commonly known as a "100-year flood."

Subsequently, some experts concluded that in certain locations, rain had fallen at a pace that made the rainfall a one-in-1,000-year event.

Rydell cautioned that those calculations were made in the week immediately after the storm, and that stating such things with certainty takes more protracted gathering and studying of available data.

"(State climatologist) Nolan Doesken says, 'Be careful what the first number (reported) is, because that's the number everyone remembers,' " Rydell said. "He is still collecting rainfall information. He is still adding to the rainfall reports, to get just as thorough a documentation of the rain that fell, as possible."

She added, "They're still out there, literally knocking on doors, saying, 'How much rain did you get? At which point did your gauge wash away?' "

Kelly Mahoney, a research scientist at the Cooperative Institute for Research in Environmental Sciences at the University of Colorado, said she has met several of the assessment team members during their visit to Boulder this week. She said she'd "love" to work on such a project.

"It's a really neat compilation of people," Mahoney said. "They're basically doing detective work and trying to piece together what happened, looking at different components and seeing what different people's perceptions are.

"Sometimes, there's a lot of separation between those two things, the perception and reality."

Contact Camera Staff Writer Charlie Brennan at 303-473-1327 or