. Experts in Canada warn that without sufficient oversight, the expansion of AI data centres could put significant strain on freshwater supplies and local ecosystems.
Water contamination
AI data centres require vast amounts of freshwater for their cooling systems. The contamination issues arise at several points in this process:
- Contaminated wastewater: A significant portion of this water is used for evaporative cooling towers. At the end of the cooling cycle, the water is contaminated with chemical coolants, minerals, and heavy metals. This wastewater is then discarded, potentially contaminating local waterways.
- Strained local supplies: As climate change puts a greater strain on water resources, the high volume of water consumed by data centres can create or worsen water stress in local communities. This is an especially significant concern in drought-prone regions, including some areas in Canada.
- Indirect contamination from power generation: The electricity used to power data centres and their cooling systems often comes from thermal power plants (like coal or nuclear). These plants also require large amounts of water and are sources of thermal pollution and hazardous waste.
Hazardous electronic waste (e-waste)
The rapid evolution of AI technology means that specialized hardware, like GPUs and servers, is frequently replaced. The improper disposal of this equipment contributes to a growing e-waste crisis.
- Toxin leaching: When e-waste ends up in landfills, toxic substances like lead, mercury, and flame retardants can leach into the soil and groundwater, eventually contaminating rivers and local drinking water.
- Ineffective recycling: Current e-waste recycling methods often involve shredding and smelting. While this can recover some metals, it can also lead to the incineration of most plastics and chemicals, creating air pollution.
Air contamination from energy consumption
The massive electricity demands of AI data centres also contribute to air contamination, especially when powered by fossil fuels.
- Diesel generators: Many data centres rely on backup diesel generators for emergency power. The testing and operation of these generators emit significant air pollutants.
- Increased fossil fuel reliance: Growing AI demand is, in some cases, being used to justify extending the life of fossil fuel infrastructure. This results in local air pollution, such as fine particulate matter, which harms human health.
Concerns specific to Canada
- Lack of oversight: Experts warn that Canada lacks sufficient regulations to address the environmental impact of the growing AI industry. This lack of transparency allows data centres to operate with little public accountability for their water and energy use.
- Water-stressed regions: Concerns have been raised about proposed data centre projects in regions already experiencing water shortages. For instance, a planned data centre campus in northern Alberta, a drought-stricken area, is expected to require millions of litres of water daily.
- Indigenous communities: Indigenous groups, such as the Sturgeon Lake Cree Nation, have raised concerns that they were not properly consulted on the impact of these large-scale AI projects on their territories and water resources.
- Policy gap: Environmental organizations are calling for mandatory reporting on water and energy use, along with binding efficiency mandates for Canadian data centres.
