Virtual reality isn’t all fun and games. It’s also helping with flood mitigation.
“We’ve built this tool that can characterize the risks around flooding and the influence that good management practices within the agricultural landscape can have on flood resiliency downstream,” said Steven Frey, a senior scientist at Aquanty Inc.
The company, a research spinoff from the University of Waterloo specializing in computer simulations, has been conducting a computer modelling project on Manitoba’s Assiniboine River Basin.
“It becomes a kind of virtual reality tool that we can play with and run these ‘what if’ scenarios.”
Like Alberta, Manitoba — and in particular, the Assiniboine River Basin — has experienced catastrophic flooding in the past decade. In 2011, the area experienced a one-in-300-year flood that stretched from late April to the end of June, costing more than $1 billion in property damage. Just three years later, heavy rainfall in June sparked a second flood that was just as devastating.
“This area has suffered from billion-dollar floods in 2011 and 2014, so flood risks are paramount,” said Frey.
The Manitoba Forage and Grassland Association launched Phase 1 of the Aquanty project in 2016 to quantify how forages and grasslands could impact water flow within the Assiniboine River Basin.
Using a software simulation tool, Frey and his project team recreated the landscape of western Manitoba and eastern Saskatchewan to create a digital world — or a sort of virtual reality — where they can look at how different management practices on the landscape might influence water flow during flooding.
“It digitally reflects what’s happening in the real world with the landscape, the soil types, and the subsurface conditions, and then it simulates the movement of water,” he said. “It gives you an indication of how the real world would respond if those types of scenarios were to happen.”
In one scenario, for example, Frey looked at the cross-section of a river and watched what happened in a floodplain that didn’t have any protection.
“The flood comes through, the river spills its banks, and the floodplain gets inundated.”
Then Frey looked at what would happen if there was some flood protection on the riverbanks. In that scenario, there was no flooding.
But river flow isn’t the only factor; the rise of the groundwater table plays a role, too. Through this virtual reality tool, Frey is able to explore the role that soil health plays on flood risks on the landscape.
And that’s where farmers come in.
“What’s happening on the agricultural landscape here is influencing how severe the floods can be downstream,” said Frey. “Different agricultural management practices do different things to the soil. Different crops and different plants have different influences on the subsurface structure of the soil. The amount of carbon that’s stored in the soil does as well.
“So can we start thinking about agricultural land management in terms of flood mitigation?”
The Prairies were once covered in perennial grasses, which created a large buildup of carbon in the soil. With that carbon came soil attributes that promoted high water infiltration and strong water retention capacity within the soil.
But annual crops have smaller root systems, resulting in less biomass below the soil surface and less carbon capture in the soil.
“That change below the surface influences how the landscape responds to floods and droughts,” said Frey.
So in the simulator, Frey has “started playing with how water sits on the landscape,” focusing on the role of different soil structures and levels of soil health on water movement both above and below ground. And in a simulation of the 2011 flood where the land is better able to hold water, the magnitude of the flood peak was reduced by as much as 12 per cent.
“We can see that with a reasonable change in the ability of the land to hold water, we could influence the magnitude of the flood peak by 10 or 12 per cent,” said Frey.
It makes sense, he said. Using agricultural land management practices — such as grasslands and perennial pastures — to boost carbon levels in the soil reduces run-off.
“Healthier soils can hold more water, and that means less water leaves the agricultural landscape, leading to less water in the rivers during flood conditions,” he said.
“But if we allow our carbon levels to be continually diminished, we’re going to reduce the soil health, reduce the capacity of the landscape to hold water, and potentially increase our flood peaks.”
Frey is, however, quick to caution that this is still “only a model” and not an exact representation of how flooding might play out in the real world. A second phase of the project, if approved, will further explore how soil health, soil carbon, and landscape management can be optimized to respond to flooding events.
“Even though the math behind the simulation is really complicated, the real world is even more complicated,” said Frey. “But it does show that the system is sensitive to what we’re doing on the land surface and that the magnitude of the flood peaks can be influenced by landscape changes.
“There’s a lot of opportunity here to utilize healthy agricultural landscape management practices to benefit society as a whole that often isn’t quantified when we’re thinking about on-farm practices.”