Water is the one thing the prairies never quite have enough of in a dry year. Ask any rancher who has watched a dugout shrink in August, or any irrigation district that has rationed its members through a drought. So it should stop people cold to learn how much of Alberta’s data centre rush is being planned in exactly the places that can least spare a drop.
A recent investigation found that roughly 76% of the data centre campuses proposed in this province sit in regions already rated under high or extremely high water stress. A single large facility can use as much water for cooling as thousands of acres of irrigated farmland. That is not a rounding error in a basin where farmers, feedlots, and towns are already competing for the same river.
The farm was here first
None of this means Alberta should slam the door on AI. It means the people who grow our food should not be the ones who pay for it in lost water. Rocky View County understood that when it turned down a major proposal in part because neighbouring landowners feared being starved of supply. That was not a county afraid of the future. That was a county doing the math on a finite resource and deciding its producers came first.
The technology to do this right already exists
Here is the part the loudest promoters tend to skip. Data centres do not have to be thirsty. Closed-loop and air-based cooling systems use a fraction of the water of older designs. In a dry southern basin, that should not be a nice-to-have. It should be the price of admission. If a company wants to build here, it can build the version that does not compete with a hay crop. Make it a condition of the licence and let the market sort out who is serious.
Put the watershed in the room
The fix is not complicated. Irrigation districts, municipalities, and the farmers downstream should be at the table before a single water licence is granted, not informed after the fact. Alberta can absolutely chase the largest projects on the continent and still insist that water for people, livestock, and crops comes ahead of water for servers. Those two goals only conflict if we let them. The same principle applies whether the project is a giant in the north or a campus next to a small town that deserves a real say in what happens to its supply.
If it comes down to a choice in a dry year, who should get the water first, the feedlot down the road or the server hall on the edge of town?
