Drilling down: Using nested models for deeper insights
Nested within Canada1Water’s continental-scale simulation are higher-resolution models that researchers like hydrogeologist Melissa Bunn can use to zoom in on specific watersheds and answer practical questions — aided by a built-in data and modelling framework that can run high-resolution models anywhere across the country.
Q: How did you come to be involved with the Canada1Water project?
MELISSA BUNN: Part of my job is doing impact assessments related to water resources — answering questions like, “If a mining operation digs an open pit here, what will happen to nearby streams and rivers?” Canada1Water will make it possible to answer those questions more accurately by integrating groundwater, surface water, permafrost and climate in a single model. We wanted to test the model for a localized watershed, so we picked Carcajou Canyon in the Northwest Territories.
Q: Why was Carcajou Canyon a good test case?
“You can feed in site-specific data but use Canada1Water to augment it and fill in some conceptual blanks, which allows you to make fewer assumptions.”
MELISSA BUNN: There were a few factors. It’s relatively small, about 7,800 square kilometres. It’s in the North, and we really want to better understand the impacts of climate change in permafrost regions. There’s more and more mining being done in those regions, so this is important. We also wanted to work with an area where there’s a mix of continuous and discontinuous permafrost, and where the surface water is well gauged. Carcajou ticked all those boxes.
Q: What have you found so far?
MELISSA BUNN: We’ve run the model and there is a good match to surface water flow and seasonality in HydroGeoSphere.[2] What’s great is you can input site-specific data but use Canada1Water to augment it and fill in some conceptual blanks, which allows you to make fewer assumptions.
Q: Are there any unexpected things you’ve been able to do with the Canada1Water data?
Melissa Bunn is a Research Scientist with the Geological Survey of Canada with a background in groundwater research and numerical modelling.
MELISSA BUNN: We were actually able to use it to generate a 3D-printed model of the water flux and permafrost at Carcajou. Groundwater modelling can be abstract, so having something you can hold in your hand makes it real. You can see the topography and the system in the 3D-printed model. I use it in conversations and presentations because it helps people instantly relate to what we’re talking about.
Q: Has your use of the model benefited the Canada1Water project?
MELISSA BUNN: The benefits flow both ways. With groundwater modelling, you have a question and have to pick a scale to try to answer it. That scale can range from an area the size of a backyard sandbox to a whole watershed like we have with Carcajou. Canada1Water gives us a conceptual understanding of regional groundwater flow where otherwise all we’d have are assumptions and a few drill holes. That lets us create smaller models with more locally detailed information on river networks and geology so we can answer more specific questions about the effects of climate change.
Q: Are there still questions you’re hoping to answer with Canada1Water?
MELISSA BUNN: For sure. We want to know how sensitive the model output will be to our assumptions, and to test how accurately permafrost is represented in the model. For water resource planning on scales like Carcajou Canyon or smaller, the climate piece is usually missing, especially where groundwater is concerned. To have that built into Canada1Water is great, as are the mid- and end-of-century timesteps. As we get into long or shorter-range planning in the North, a tool like Canad1Water will be priceless.
[2] HydroGeoSphere is Aquanty’s integrated hydrogeological modelling platform and the engine that will drive the Canada1Water integrated hydrologic simulations.
A WATERSHED TO HOLD IN YOUR HAND Bunn’s team generated this 3D printout of Carcajou Canyon using the C1W model. The multicoloured section shows different permafrost conditions; the blue shows the water flux as of the start of the model run (1981).