Promising New Tool for Ice Safety – Sensors Used to Study Soil

Emily Whitaker (far left) and students ready to monitor ice on northern Wisconsin lakes.

Every winter, as lakes and rivers freeze over in the colder reaches of the Northern Hemisphere and again when they thaw in the spring, anglers and skaters and snowmobilers all start asking the same question – is the ice safe to go out on?

While ice can never be deemed safe with 100% certainty, there are some accepted rules of thumb about when ice fishing or pond hockey or cross-country skiing are relatively reasonable pursuits. The recommended minimum thickness is four inches for a person wandering out on the ice, while anglers with a fishing shanty hitched to their truck should wait until there’s a foot or more before driving it to their favorite spot.  

There’s a catch to these safety guidelines – more often than not, the only way to measure ice thickness is to go out and drill a hole. There’s no knowing if it’s safe until you’re already out on it.

Courtesy Minnesota DNR

But there might be a more practical way to answer the safety question. And it involves an unlikely tool.   

That tool, says Center for Limnology graduate student, Emily Whitaker, is a particular kind of probe that is usually buried in the ground and used to measure the amount of water in the soil. The thing is, she says, initial models of these water content sensors developed notoriety among researchers for being “really bad at their job.” While the company that made them addressed the flawed instruments by coming out with newer models that worked much better, Whitaker says, “we wanted to find a way to repurpose the first generation.”

The first step came when Whitaker was still an undergraduate at Dickinson College in Pennsylvania. In the summer of 2015, she worked for UW-Madison professor of atmospheric and oceanic sciences, Ankur Desai. Desai was collaborating with David E. Reed a visiting professor at Dickinson and Reed had “this crazy idea that maybe [the soil water content sensors] could measure changes in ice thickness,” Whitaker says. The idea behind the instruments, she explains, is that each sensor has two probes and measures how efficiently an electrical signal can move between them. Water is a good conductor of electricity so when water levels in soil are high, signals should easily move between probes. When the soil dries up, however, that conductivity falls.

To see if the same was true with ice, Whitaker spent the summer of 2015 building a lake in a trashcan. The trashcan was filled with water, one of the soil water sensors was lowered in and then the whole ensemble was wrapped in insulation and placed into an environmental chamber mimicking Antarctica conditions. Whitaker, Reed and Desai conducted 24-hour runs where they lowered temperatures to well below freezing in the environmental chamber and monitored ice formation. They then brought temps back up and recorded the thaw.

Sure enough, Whitaker says, “we found we could record the ice we were making and its thickness in our fake lake.” The only problem, she says, is “the real world doesn’t freeze like that.”

An array of soil moisture probes (far left) sit frozen in the surface of Lake Mendota, while a nearby weather station collects data. Photo: E. Whitaker.

That’s why, in the early months of 2016, another team led by David Reed headed out onto a frozen Lake Mendota, in Madison, Wisconsin. Using long, wooden dowels and an innumerable amount of plastic zip-ties, they created a half dozen arrays, mounting several soil moisture and temperature sensors to each dowel. They then drilled holes in ice, set up their sensors and allowed them to freeze into the lake. Over a couple of weeks, the instruments recorded data on water temperature and ice thickness from the surface of the ice to two feet down into the lake.

 In a paper on the experiment published this year, Whitaker and her collaborators write that the experiment demonstrated a new method of monitoring that “allows low-cost and high-frequency measurements of ice thickness.”

This is a big deal, Whitaker says, because the only ways to know how thick ice is on a lake without going out on that ice is very expensive, involving radar or underwater buoys or, perhaps, some very tricked-out drones. And, even then, those methods only give a snapshot of conditions once or twice over the winter.

Ice conditions can change day-to-day, however, based on variables like air temperature or big rain events. Ice conditions also vary across the surface of a lake. Ice that is ten inches thick in one location may be less than half that in a nearby location due to things like currents in the lake or water inflows coming in from land. What’s more, reliable winter ice cover is becoming more intermittent as lakes warm, meaning that winter’s outdoor enthusiasts may not be able to count on regular ice seasons anymore.   

Photo: John Magnuson

Whitaker envisions an array of sensors as a way of getting both better data on true winter ice conditions and improving ice safety. If researchers can come up with a way to install the instruments in a lake before it freezes and have those instruments send their data to shore, then we could monitor ice formation and thickness in real time at many points across a lake.

“In a dream world,” Whitaker says, “you could make it an app.” Already, she says, some people are thinking about asking ice fishermen to report ice thickness on a smartphone app whenever they drill a hole.

Whatever they come up with, it will be better than the current approach. Right now our approach to measuring ice thickness for safety is, well, less than scientific.

“Usually,” she says, “people just guess.”