State: Flood data need update

Planners blame old information for rain-swamped highways

The failure to anticipate major flooding in Arkansas over the past five years is due, in part, to out-of-date information that experts say will take two years and $300,000 to fix.

The equations hydrologists and engineers use for estimating peak flooding in the state’s watersheds are based on a statewide flood-frequency analysis the U.S. Geological Survey performed in 1995, or nearly 20 years ago.

The magnitude, frequency and timing of floods likely have changed in that time, according to the agency. In particular, it points to the severe flooding that happened in 2008, 2009 and 2011, which likely altered the frequency statistics used in the equations.

Anticipating the likelihood of peak flooding - typically calculated at 25-, 50- and 100-year flood intervals - is crucial for engineers when designing a bridge and the roads leading to and from them.

“These calculations are instrumental for engineers in setting the grade for roadways and bridges,” said Randy Ort, a spokesman for the Arkansas Highway and Transportation Department. “It sounds odd there may be only a 6-inch change in the elevation of a 50-year flood under the old calculations to the new, but 6 inches of water over the roadway is a major difference.”

The science behind the calculations is good but the information no longer may be.

“It’s not that we have bad equations, but we need [updated] information,” said Jaysson Funkhouser, assistant director of the Geological Survey’s water science office in Little Rock. He also oversees the agency’s 160 stream gauge stations and 125 water-quality program sites in Arkansas.

That means the set of equations based on the 1995 data likely are “under-predicting the magnitude and frequency of occurrence of flood events in Arkansas,” according to an agency report released in December outlining the problem. “Because the analyses and, consequently, the regression equations are out of date, there is a need to perform a new analysis of flood frequency” in the state.

The equations are important for engineers when designing drainage structures, culverts, levees and bridges near rivers and streams.

The equations they now use “likely don’t take into account new development and the runoff it causes,” Funkhouser said.

An Arkansas Highway and Transportation Department study cited outdated data as a factor in the unexpected flooding that happened on a section of Interstate 30 in Little Rock three times between 2005, when it was rebuilt, and 2011.

The section, about a mile west of the Otter Creek exit, flooded most recently on Nov. 21, 2011, closing all six lanes for about three hours. It also was closed for more than 10 hours on Christmas Eve 2009 because of high water. The study noted the section has flooded at least one other time since the section was rebuilt in 2005.

Before rebuilding that section of the road, the department relied on existing hydraulic models, according to the 2012 study. The water elevation estimates the models produced couldn’t explain why that section of the roadway had flooded so often, the study found.

The data used to design the section didn’t take into account the increase in surface-water runoff, spurred by development within the Crooked Creek drainage basis and the placement of fillin the stream’s floodplain, which constricted the stream and caused an increase in the depth of the flooding in the vicinity of the I-30 section that flooded, the study found.

To fix it, the department has proposed widening the channel of Crooked Creek in the vicinity of the flood-prone section of I-30 and possibly adding a small levee on the south side of the interstate. A study of the area concluded these steps should offset the placement of fill in the creek’s flood plain, a result of development at the creek’s headwaters near Bryant and the residential development between Bryant and Alexander since the mid-1990s.

The estimated cost of the project is more than $800,000. It is scheduled to be awarded a contract in July, Ort said.

The I-30 project is but one example, he said.

“We need this [new] data for all our projects,” Ort said.

Which is why the department has agreed to contribute $150,000 toward a Geological Survey effort to update the equations by incorporating additional rainfall and stream-flow data collected since the 1995 equations were published. The U.S. Army Corps of Engineers also has agreed to participate in the effort and contribute an identical amount.

Updating the equations used to take place every 10 years, a practice first adopted in the 1940s or 1950s, Funkhouser said.

“What happened is we lost funding,” he said.

The study will take two years to complete, Funkhouser said. The money will be used to pay for a hydrologist, a hydraulic engineer, geographic information specialist and an information technology specialist.

Annual maximum discharge data will be retrieved from the agency’s water information system on streamflow gauges in and near Arkansas that have 10 years or more of peak-flow record and that measure flow in watersheds without “significant regulation and without significant urban development,” according to the Geological Survey’s formal proposal.

That information will be used to compute data for each gauge site and then used to develop a map.

What is described as the most challenging and time-consuming portion of the study is developing the basin characteristics of the watersheds, such as elevation, percentage of forest cover and 24-hour precipitation. The Army Corps of Engineers will oversee that portion of the study.

“They have the expertise and the resources,” Funkhouser said. “Accurately defining the drainage characteristics is the crux of the project. It’s what takes a long time.”

All of that information will be used to analyze whether the three geographic regions used in the 1995 study are still appropriate or if new regions are needed. Once the regions are developed, new equations can be developed for each region, according to the proposal.

Whether the new information will have an impact on highway project designs is unclear.

The study “is basically trying to find out what causes these [flooding] peaks so you can mathematically define them,” Funkhouser said. “What we once thought as a 100-year flood might be a 50-year flood or more frequent. My guess is that is going to be the case.”

Front Section, Pages 1 on 02/03/2014

Upcoming Events