Extreme weather event powers new storm simulation tool
Jan. 23, 2014 |
Record-shattering rainstorms that hit west-central Wisconsin in June 2008 caused catastrophic damage and lasting images, including the televised failure of an earthen dam containing Lake Delton, which swept away several houses and sections of roads.
The Federal Emergency Management Agency paid $34 million in damage claims after the week-long barrage of rain poured down on the Baraboo area, flooding 810 square miles of land. Sewage treatment plants in scores of communities were swamped, leaking 90 million gallons of raw sewage and contaminating a quarter of area drinking water wells. Interstate highways closed for three days due to flooding.
These epic events overwhelmed local stormwater management and infrastructure. But they also presented an opportunity for communities across the state to ask themselves: What if that same downpour had happened in Eau Claire, Appleton, Madison, or any other location?
Answering that question becomes more crucial each year. Storms with heavy rains are on the rise in Wisconsin, according to UW-Madison scientists affiliated with the Wisconsin Initiative on Climate Change Impacts (WICCI). Local officials who plan, develop and manage stormwater infrastructure must account for this increase -- along with land use changes and development -- to protect local property, water quality and the health of residents.
"If communities can understand their vulnerabilities," says Kenneth Potter, a UW-Madison professor of civil and environmental engineering, "then they can try to correct them."
But statistical analyses and probabilistic climate models don't necessarily win over a skeptical public when it comes to spending money for preparedness. People need to be convinced, and having the ability to visualize "what-if" scenarios provides a powerful tool to illustrate risk.
Potter and colleague David Liebl, a statewide stormwater specialist with UW Cooperative Extension, have developed a new approach to help local decision makers see how their stormwater management systems would handle an enormous rain event. They used NEXRAD data from Milwaukee, La Crosse, and Iowa's Quad Cities, as well as data from rain gages, to estimate the spatial distribution of the Baraboo-area rainfall at a 15-minute time interval from June 2 through June 13, 2008. They then developed a computer program that can be used in conjunction with hydrologic models to determine what would happen if the 2008 storm had been centered over any given region. They are currently applying this approach to the Lake Mendota, Starkweather Creek, and Black Earth Creek watersheds.
Potter and Liebl, who co-chair a WICCI working group on stormwater, say this exercise can help stormwater system designers and managers understand the increasing risk of extreme events in a changing climate. It could also help update design and management tools, which currently rely on rainfall scenarios that do not account for rising temperatures, seasonal shifts in rainfall, changing amounts of rainfall, and impacts to ecosystems, commercial and residential property and farmland.
The innovative program allows local municipalities to test their stormwater management systems and infrastructure through simulation, using the real numbers of the Baraboo storm system. The program also allows users to use new extreme storms when they happen.
Between June 1-15, 2008, 12 to 15 inches of rain fell over the Baraboo River watershed. This map shows the areas that received the most rain, with pink showing the highest volume, and green showing the lowest.
"With the traditional models we use, where we run a whole year of data, we often do not account for the short, intense storm events that cause us problems," says Jeremy Balousek, urban conservation engineer for Dane County's Land and Water Resources Conservation Department. He's using Potter's storm transposition software.
"Being able to test our infrastructure against known damaging storms is very beneficial, especially since we can move the most intense point wherever we want to test," Balousek says.
Despite technological advancements, no one can predict exactly how much precipitation is likely to fall in any specific location, limiting the usefulness of climate models to stormwater managers. Additionally, "thunderstorms are still too small for climate models to capture," says Potter. Models cannot tell ‘when' or even ‘if' an extreme storm will hit a specific municipality, a fact that fails to resolve the budgetary tug-of-war between preparing for the worst (a 50-, 100- or 500-plus-year storm event) or allocating limited financial resources to other demanding projects.
No state or national standards exist to guide local municipalities on how best to prepare infrastructure to cope with catastrophic weather events, Potter points out. Despite the increasing trend toward more extreme rainstorms, most communities only plan for 10-year storm events. The 2008 Baraboo storm was a 500-year flood event.
Balousek says the simulation tool is a tremendous help when speaking with decision makers and citizens about costly design changes to stormwater infrastructure, because it allows them to see the consequences of a catastrophic storm.
"We are seeing how well we can withstand changes," says Balousek. "We can do a great job given the right resources and tools."