How A 30-Year Rain Event Became a 100-Year Flood

Lake Monona, Madison, WI. Photo: A. Hinterthuer

In case you’ve somehow missed the news, let’s just say that it’s been wet in Wisconsin over the last week. Really, really wet

Here in Madison we’re still holding our breath, filling up sandbags and hoping no new major rain event pushes water-levels even higher. Already, our lakes have hit near-record water levels, causing floods in low lying areas (like our basement!) and prompting beach closures and slow-no-wake zones for the entire chain of lakes.

But, says, Daniel Wright, an assistant professor in the department of Civil and Environmental Engineering here at UW-Madison, it could have been so much worse. 

Wright recently wrote a report that says the August 20th storm “clearly demonstrated Madison’s vulnerability to extreme rainfall.”

The report is excellent and you should read the whole thing here, but we’ll highlight the main points below. 

In Madison, At Least, It Could Have Been Much Worse

First, some basics. All creeks, rivers and lakes have watersheds. The Yahara watershed is 281 square miles of farmland, urban development and a little bit of wetlands that survived the land-use change to farms and cities. Any rain that falls on those 281-square miles and starts flowing downhill and downstream eventually ends up in Lakes Mendota, Monona, Waubesa and Kegonsa. 

So, obviously, a lot of rain raises the levels of our lakes. But, says Wright, while the amount of rain that fell in places like Cross Plains and Black Earth on August 20th was historically massive, the amount that fell in the Yahara watershed wasn’t. 

Left map: National Weather Service radar and citizen rain gauge observations for Aug. 20, 2018 storm. Right map: “Worst Case Scenario,” if the storm had occurred 15 miles to the northeast, directly over the upper Yahara watershed. Radar data courtesy of Dr. Bong-Chul Seo at the University of Iowa.

The image above highlights the part of the Yahara watershed that is upstream (i.e. flows into) Lake Monona and the “rain map” of the August 20th event. You can see that, while the western edge of the Yahara watershed got a ton of rain, most of the watershed received between 3 and 6 inches. While that’s still a lot of water, it’s not historic. As Wright explains in his paper:

“The rainfall that fell inside the 281 square mile watershed works out to about 4.5 inches, equivalent to 33,272 Olympic swimming pools of water. This is a lot of water to be sure, but if the storm had occurred 15 miles to the northeast—as is shown in the “shifted” rainfall map on the right —it would have dropped 8 inches of rainfall across the watershed, almost double what actually occurred. Given the flooding we’ve seen over the past week, it is reasonable to assume that this “worst case scenario” would have been absolutely catastrophic for lakefronts and other low-lying parts of Madison.”

Non-Historic Rain, But Historic Flood?

Lake Monona is on the verge of an unprecedented flood, despite missing the worst of the rain. Photo: A. Hinterthuer 

Thankfully we didn’t get the worst of the rain. By Wright’s estimation (using software designed by the Hydroclimate Extremes Research Group), the amount of rain the Yahara watershed got last Monday fell somewhere in between a 20 to 50 year recurrence interval.  He settled on 30 years as a best guess, which means, Wright writes, “over the course of a 30-year mortgage, a homeowner will, on average, experience one storm similar in magnitude to the August 20th storm.” 

So why, then, did that rain produce a flooding event we’d only expect to see once over the course of three or four 30-year mortgages? 

Well, for starters, Lake Mendota. 

We don’t like pointing fingers at the lake that started it all for limnology here in North America, but, well, its water levels are 5 feet higher than their natural levels thanks to the Tenney Lock and Dam. 

The problem is, people like living on lakes. And, once they’re there, they also like having access to the lakes, which means a pier and a boat and maybe a little beach or swimming area and, most important, consistent water levels. 

But consistently high water levels in Lake Mendota come with a cost. When we get a big rain, the lake doesn’t have the capacity to take that extra load in. It’s already so high that the only thing city officials can do is open the lock and dam and let the excess run downstream. 

Roaring river
Photo: Amber Arnold, Wisconsin State Journal 

Back during the 2008 flood, says Eric Booth, a research scientist in the Hydroecology Lab at UW-Madison, calls were renewed to lower Lake Mendota’s levels so that the water from big rain events in the upper Yahara watershed could be held back and more slowly released downstream.

“People were interested in using Mendota more as a flood control reservoir,” Booth says, “but push back came from shoreline homeowners and people in places like Westport who wanted their docks accessible all summer.” Over the last couple of decades, Booth says, “it’s been a back and forth between those that want to draw Mendota [water levels] down and those that want it consistent. We might see renewed calls depending on how much worse things get this year, but it’s an old argument that quiets down between huge flood flood events. My old advisor called it the ‘flood memory half life.'” [EDITOR’S NOTE: Booth is not implicating all shoreline homeowners – depending on where a property is situated, high water levels are most definitely NOT desired, and many advocate for lower levels]

In addition to Lake Mendota’s artificially high water levels, is the fact that a lot of the rain we got on August 20th fell in the city and surrounding suburbs, where infrastructure like roads and parking lots and roofs keep water from infiltrating into the soil and simply shove it quickly downstream. 

Stormwater runoff in Arlington, Virginia.
Stormwater runoff in Arlington, VA. Photo: Tyrone Turner. 

“It is likely that the impacts would have been less severe if the rain had been concentrated in the northern portion of the watershed in Columbia county,” Wright’s paper says. “This is because the largely agricultural landscape there can soak up more rainfall than the urban and suburban areas in and around Madison.”

In fact, the Wisconsin State Journal highlighted these very problems in an excellent article on August 5th, just two-weeks before the big storm. It’s important that we continue this conversation because storms like the one we saw last week – or the folks in western Wisconsin saw last night – are becoming more common.

And, while storms like that cause floods, it’s our decisions here on land that make them historic.

You can download a copy of Wright’s report here.