Active Discussion

If Alberta Builds Gigawatt Data Centres, It Should Build Industrial Ecosystems Around Them

L
lisa
Posted Sun, 1 Nov 2026 - 12:35

Waste-heat greenhouses will not erase the emissions from gas-powered computing. But Alberta can require its largest data centres to become efficient, measurable and connectable parts of a regional industrial system rather than isolated consumers of fuel, land and water.

Alberta's largest proposed data centres should not be evaluated as warehouses filled with computers. At gigawatt scale, they are energy-conversion complexes: fuel, electricity, heat, water and carbon move through them continuously, while communities and ecosystems inherit the consequences.

Global data-centre electricity use was already estimated at about 415 terawatt-hours in 2024, and the International Energy Agency projects rapid further growth, although the pace remains uncertain.[6] Alberta is actively positioning itself for that expansion. Its framework favours projects that bring their own generation, while a new hardware levy is lower for self-generating facilities and disappears for off-grid facilities.[4]

That creates a clear risk: the policy can reward independence from the grid without necessarily rewarding lower emissions.

The answer is not to pretend that a greenhouse beside a server building makes a massive gas-powered project environmentally benign. It does not. Heat reuse is not a substitute for efficient generation, a lower-emission power supply, carbon management, cumulative-effects assessment, responsible water allocation or meaningful Indigenous participation.

But if Alberta decides that very large data centres will be built, it should not permit them as isolated facilities that discard usable resources and externalize long-term liabilities. It should require them to become connectable foundations for regional industrial districts—and make later expansion conditional on measured performance.

Wonder Valley shows why a whole-system test is needed

The proposed Wonder Valley project south of Grande Prairie illustrates the scale. Publicly reported project information has changed over time: the original announcement described a 7.5-gigawatt campus, while a June 2026 update described as much as nine gigawatts of on-site generation at full build-out.[2]

A preliminary analysis prepared for Sturgeon Lake Cree Nation estimated that full operation could ultimately release roughly 18 gigawatts of heat and about 30 million tonnes of carbon dioxide annually. These are preliminary estimates based on public information, not verified engineering results.[1]

The province exempted the project from a provincial environmental impact assessment on the ground that it proposes standard power and water systems, while emphasizing that permits and detailed air, land and water assessments are still required.[3] The counterargument is compelling: standard components assembled at unprecedented scale can create non-standard cumulative effects.

This is precisely why the approval question cannot be reduced to whether each turbine, cooling system or water licence satisfies its individual rule. A facility of this scale needs a whole-system accounting of fuel, electricity, heat, emissions, water, land, transmission effects, Indigenous rights, regional labour and eventual decommissioning.

Nearly all of the energy eventually becomes heat

A gas-powered data-centre complex produces heat in two distinct stages.

First, the generating plant converts only part of the fuel's energy into electricity. The remainder leaves through turbine exhaust, cooling systems and other losses. This is the higher-temperature stream. It may support additional power generation, industrial steam, absorption cooling or process heat, depending on the actual plant design.

Second, nearly all electricity used by servers, networking equipment and cooling equipment ultimately becomes lower-temperature heat inside or near the data centre. Liquid-cooled computing can make this stream easier to collect, but it remains much less valuable than turbine exhaust.

These streams should not be merged into one vague promise of “waste-heat reuse.” They have different temperatures, economics, infrastructure requirements and potential customers.

The order matters: use the hottest energy first, then cascade it into lower-temperature applications. High-grade heat should first do work that low-grade heat cannot. Server heat should go to uses that genuinely need warm water rather than being raised to a much higher temperature without a demonstrated customer.

Build a portfolio, not a miracle greenhouse

No realistic greenhouse development will absorb gigawatts of heat by itself. Nor should a data-centre operator be expected to become a grower, fish producer or forestry company.

The more credible model is a diversified industrial district with several possible customers:

  • greenhouses producing high-yield vine and leaf crops;
  • aquaculture with stable warm-water demand;
  • lumber, oriented-strand-board, grain and oilseed drying;
  • absorption cooling and other seasonal balancing uses;
  • snowmelt, soil heating and low-value flexible loads;
  • industrial processes that can accept appropriately graded heat;
  • treated-water users where the resulting water actually meets their quality requirements.

Even a successful district may use only a minority of the available heat. That is not failure if the numbers are reported honestly. The objective is to capture the economically and environmentally useful share—not to manufacture a claim that every rejected megawatt has become a public benefit.

Carbon dioxide requires even greater care. Turbine exhaust cannot simply be piped into a greenhouse and called food-grade. Any useful carbon stream would require capture, purification, contaminant monitoring, compression, storage and quality assurance. Greenhouse demand would also be tiny compared with the emissions of a multi-gigawatt gas fleet. Carbon utilization can support a local industry; it cannot replace permanent emissions reduction and storage.

The practical obligation is a fence-line interface

The province should require the operator to deliver standardized, metered resource connections to the edge of the approved site. Past that point, producers and industrial customers can decide whether connecting makes economic sense.

A serious interface would include:

  • heat exchangers that isolate the computing coolant from external users;
  • separate mains for low-temperature water and higher-temperature water or steam;
  • published temperature, flow, pressure, reliability and outage specifications;
  • a quality-controlled carbon-dioxide connection only if capture is installed;
  • treated-water connections with disclosed chemistry and permitted uses;
  • metering that distinguishes resources offered, purchased and actually used;
  • space and rights-of-way reserved for future connections.

Both sides must retain backup. The data centre must be able to reject all of its own heat safely if no customer is operating. Greenhouses and industrial customers need backup heat if the computing or generating facility goes offline.

This avoids two bad outcomes. It does not force an AI company to run a food business, and it does not let the operator advertise theoretical reuse that no customer can physically access.

Six conditions for approval

A credible Alberta framework should place the largest conditions in the order that they affect emissions and public risk.

1. A measured generation-performance standard

Approval should specify maximum emissions and fuel input per unit of useful electricity, verified under real operating conditions. A technology name is not enough: performance depends on plant configuration, ambient temperature, part-load operation and internal energy use.

2. Staged expansion

Later power and computing blocks should not receive automatic approval. Their construction should depend on earlier blocks meeting published targets for efficiency, emissions, water, consultation, reclamation funding and resource-interface availability.

3. An emissions-intensity incentive

Data-centre taxes and levy relief should reward lower tonnes of carbon dioxide per megawatt-hour, not simply independence from the public grid. Alberta's current structure reduces the hardware levy for self-generation and eliminates it for off-grid projects.[4] That incentive should be redesigned so that the cleanest reliable supply wins rather than assuming one generation model in advance.

The generation debate is legitimate. Alberta and the system operator emphasize around-the-clock reliability and protection of existing ratepayers. Critics argue that excluding renewable, storage and hybrid configurations suppresses competition and locks in gas emissions.[5] A performance standard can test both claims without writing a preferred technology into the rule.

4. A heat, water and carbon availability obligation

The operator should build the fence-line interface and disclose its performance. The obligation should initially concern honest availability rather than guaranteed third-party uptake. Public reporting must prevent unused pipes from being counted as environmental outcomes.

5. Full-life decommissioning security

A reclamation bond should be funded before construction and updated as the site expands. Alberta should not assume that today's AI demand, owners or financing structures will survive for the full physical life of a multi-gigawatt energy complex.

6. Indigenous participation with real authority

Affected First Nations require more than late-stage information sessions. The framework should provide early consultation, access to independent technical expertise and a genuine option for equity or utility ownership where nations choose it. Ownership cannot substitute for rights, consent requirements or environmental scrutiny, but it can convert some of the project's long-term value into durable local authority and revenue.

The strongest objection must remain visible

The strongest case against this proposal is that an industrial-symbiosis district could provide social licence for a project whose total footprint remains unacceptable. A greenhouse, fish farm and heat main could become decorative beside tens of millions of tonnes of annual emissions.

That objection is valid. The framework must therefore preserve the possibility that the answer is no—or that a smaller, cleaner configuration is the only approvable one.

Heat reuse should never be credited as though heat disappears. It eventually returns to the environment. The benefit exists only where the reused heat replaces another input that would otherwise have been consumed. Likewise, captured carbon used in a greenhouse generally returns to the atmosphere; it is not permanent storage.

The test is not whether the project can display a circular-economy diagram. It is whether independent measurement demonstrates lower fuel use, lower emissions, responsible water use, credible cumulative-effects management, enforceable Indigenous participation and secured end-of-life obligations.

Alberta can write a better rule

Alberta's gas resources, cold climate, industrial workforce and available land may attract large computing projects. Those advantages do not require the province to accept the weakest possible bargain.

A data-centre approval should answer five public questions:

  1. How much fuel, water and land will the complete system require?
  2. What emissions and waste streams will it produce under measured—not advertised—performance?
  3. Which resources can neighbouring industries physically access, at what quality and reliability?
  4. What conditions stop expansion when early promises are not met?
  5. Who pays if the technology, operator or market disappears?

The central question is not whether waste heat is “green.” It is whether Alberta can turn an enormous new demand for energy and water into a system that produces more durable public value, fewer emissions per unit of computation, meaningful Indigenous authority and enforceable protection against stranded liabilities.

That is a standard worth debating before the province commits itself to a generation of gigawatt-scale infrastructure.

Sources

[1] https://www.cbc.ca/news/canada/edmonton/wonder-valley-slcn-court-physicist-9.7403677 — Proposed data centre would release ‘23 Hiroshima-sized atomic bombs worth of energy’ a day, physicist says > "At full operation, it would release around 18 gigawatts of heat a day and about 30 million tonnes of carbon dioxide a year — more than four per cent of Canada’s total emissions, said Davies, whose assessment was based on publicly available information." [2] https://www.cbc.ca/news/canada/edmonton/wonder-valley-newsletter-open-house-9.7217459 — Updated info about water use, timelines for proposed Wonder Valley AI project in Alberta > "It would inhabit approximately 64 square kilometres in the Municipal District of Greenview and will include up to nine gigawatts of on-site power generation capacity when all phases are completed, a spokesperson for the open house confirmed in an email to CBC News." > "The newsletter said Wonder Valley would draw water from the nearby Smoky River for the facility, and noted that the M.D. of Greenview holds a Water Act licence to use up to six million cubic metres of water per year. It also has a preliminary certificate to withdraw up to 24 million cubic metres of water per year." [3] https://www.cbc.ca/news/canada/edmonton/wonder-valley-data-centre-environmental-impact-assessment-9.7158526 — Wonder Valley AI Data Centre exempt from provincial environmental impact assessment, still needs permits > "The Ministry of Environment and Protected Areas says the Wonder Valley AI Data Centre Park project was reviewed by an independent approvals officer who deemed it exempt from the assessment due to its proposed use of standard power and water systems." [4] https://www.blg.com/en/insights/2026/07/data-centre-regulation-in-alberta — Data centre regulation in Alberta: key considerations > "The Financial Statutes Amendment Act, 2025 (No. 2) introduced a 2 per cent levy on computer hardware that applies to grid-connected data centres drawing 75 MW or more." [5] https://www.ctvnews.ca/edmonton/article/why-renewables-are-being-left-out-of-albertas-data-centre-power-plans — Why renewables are being left out of Alberta’s data centre power plans > "Clean energy think tank The Pembina Institute also told the operator it should take a “technology agnostic approach” arguing renewables, energy storage and dispatchable thermal backup can compare to natural gas at a lower cost and with lower emissions." [6] https://www.iea.org/reports/energy-and-ai/energy-demand-from-ai — Energy and AI: Energy demand from AI > "Today, electricity consumption from data centres is estimated to amount to around 415 terawatt hours (TWh), or about 1.5% of global electricity consumption in 2024. It has grown at 12% per year over the last five years."

CanuckDUCK Provenance Package

Status: Verified. This article has a lightweight authenticity reference anchored to CanuckDUCK local Hedera; full content remains in Pond.

  • Package: 091d3914-df27-43e3-965b-e76acf63aca6
  • Hedera topic: 0.0.1068 / sequence 10455
  • Graph pathway and Consensus binding: present as package sections; Consensus may show no linked poll until a public input surface is created.

Verify package · Machine-readable JSON

--
Consensus
Calculating...
0
perspectives
—
views
Constitutional Divergence Analysis
Loading CDA scores...
Perspectives 0