DENVER -- The widespread impacts of September's extraordinary rainfall in Colorado's Front Range, from landslides to peak river flows, are still being cataloged by scientists across the state.

Colorado's landscape is known for its dramatic topography, and that topography also made the land's response to significant rainfall dramatic.

Massive landslides, which tore up roads, destroyed buildings and killed three people, were just one example. U.S. Geological Survey scientist Jonathan Godt showed a map of the area where flood-related landslides occurred.

The triangle-shaped area, 60 miles long and nearly 40 miles across at its widest, ranged from the top of high mountains well above tree level all the way down to the plains.

"The big part of the flooding story and the impact of the floods is a function of the landslide activity and the debris flow activity that was able to transport the debris into the stream channels," Godt said.

He flashed dramatic images of a landslide scouring its way down from near the top of Mount Meeker, a 13,000-foot peak, to state Highway 7 thousands of feet below. Landslides that went through the town of Jamestown left walls of debris 3 to 6 feet thick across the road in some places, Godt added.

Farther east, in the city of Boulder, boulders from a landslide split a building in two.

A force beyond measuring
Another effect of the flooding was the destruction of the very instruments used to measure its impact, said Robert Kimbrough, of the Geological Survey's Colorado Water Science Center (ClimateWire, Sept. 19).

Kimbrough showed graphs of stream flow measurements whose data abruptly halted just when the flooding reached its peak. That's when the gauges used to measure river readings were ripped out of streams.

Among the broken gauges were four on the St. Vrain Creek near Lyons, which were damaged and stopped reporting after a surge, Kimbrough said.

Now, USGS is trying to determine the stream flow records, using measurements of high water marks. Until then, it will not know whether stream flow in those areas set records.

"It could be that we see a new peak flow of record for Lyons," Kimbrough added. The Geological Survey is also still working on determining the likelihood of such floods occurring in various drainages, he said.

When the floods were occurring, there were many questions as to whether they were 100-year floods, which is an event that has a 1 percent probability occurring in a given year. USGS was not yet prepared to issue this number, called a recurrence interval, for the recent floods, he said.

What the weather models didn't know
Burn scars from past wildfires also may have played a role in speeding runoff from certain drainages.

USGS hydrologist Sheila Murphy looked at the contributions of different watersheds to flow levels in Boulder Creek during the flood and found that the Fourmile Creek drainage, which burned in 2010, was a significant source of runoff to Boulder Creek.

"I'm just glad that this storm didn't happen three years ago right after the fire, because I think it would have been much worse," she said.

Scientists also hope to use the floods as a way to improve their weather and hydrologic models, said David Gochis, a scientist at the National Center for Atmospheric Research.

Weather models did not perform very well during the storm (ClimateWire, Sept. 26). This extreme event provides researchers with an opportunity to identify sources of error in their models and improve them, Gochis said.

By adding new sources of data to improve the initial conditions in weather and hydrologic models, he said, researchers hope to improve future predictions. But they still have a lot of work to do.

"Dissertations galore will come out of this event, it's safe to say," Gochis said. The audience laughed, but he was only partly joking. The post-flooding science effort has only just begun, he and other researchers said.

Reprinted from Climatewire with permission from Environment & Energy Publishing, LLC., 202-628-6500