Skip to content
This is NOT an official site of the Government of Canada. Click here for the official AI registry.

Helpful Places

Government of Canada – DTPR AI Register

Back to your search

AI-Powered Weather Forecasting for Canadians

Planning & Decision-making · Ecology

What it collects

About a measurement
Anonymized data
  • ERA5 global atmospheric reanalysis data — a comprehensive record of historical weather conditions including temperature, wind, humidity, and pressure at multiple atmospheric levels, produced by the European Centre for Medium-Range Weather Forecasts.
Operational data
Anonymized data
  • MSC numerical weather and environmental prediction model outputs — structured grid-based forecasting data produced by existing operational models at the Meteorological Service of Canada, used as training and inference inputs for AI components.
Run by
Environment and Climate Change Canada (ECCC)
Where
No fixed location
Kept
Not stated by the Helpful Places.
Shared with
Accountable organization, Me

What it is for

Planning & Decision-makingEcology

Environment and Climate Change Canada is developing AI models to improve weather prediction, drawing on global reanalysis datasets and existing numerical weather models. The system is still in development and will eventually serve both government forecasters and the general public. No personal information is collected or used. Canadians should know this is an active research program — not yet a deployed service — and that all AI models are being integrated alongside traditional forecasting methods.

What it collects and what happens to it

Data taken in

About a measurement
Anonymized data
  • ERA5 global atmospheric reanalysis data — a comprehensive record of historical weather conditions including temperature, wind, humidity, and pressure at multiple atmospheric levels, produced by the European Centre for Medium-Range Weather Forecasts.
Operational data
Anonymized data
  • MSC numerical weather and environmental prediction model outputs — structured grid-based forecasting data produced by existing operational models at the Meteorological Service of Canada, used as training and inference inputs for AI components.

Processing

Classification & Prediction
  • Energy-efficient AI weather models including GraphCast, ForecastNet, AIFS, and the in-development PARADIS model use deep learning and hybrid physical-AI approaches to forecast atmospheric states, including the spectral nudging technique that combines numerical weather prediction with learned corrections.

What it does

Deciding (Analytical AI)
Human decides
  • AI weather models (e.g. GraphCast, ForecastNet, AIFS, PARADIS) predict and classify atmospheric states, producing forecasts and scores from structured meteorological inputs. Human forecasters review and act on these outputs.
Sensing (Perceptive AI)
Human decides
  • The hybrid 'spectral nudging' system ingests raw numerical weather prediction model outputs and transforms them into structured fields that AI components can process, bridging traditional forecasting with learned representations.

Outputs

About a measurement
Anonymized data
  • AI-generated weather forecasts — predicted atmospheric conditions such as temperature, precipitation, wind speed, and other environmental variables over future time horizons, intended for use by government forecasters and eventually the public.
A recommendation or prediction
Anonymized data
  • AI model outputs serve as advisory forecast guidance to human meteorologists, who review and refine them before issuing official forecasts. The outputs are recommendations rather than final authoritative decisions.

Run by

Environment and Climate Change Canada (ECCC)
  • The Meteorological Service of Canada and Science & Technology Branch of ECCC are jointly developing and deploying AI for weather prediction as part of the 2023 AI Roadmap through to 2030.

Government of Canada AI Register — AI for weather prediction (2526-ECCC-002)

Built by

Government of Canada
  • The Government of Canada is listed as the developer of the AI system, indicating that development is conducted in-house rather than by a third-party vendor.

Government of Canada AI Register — AI for weather prediction (2526-ECCC-002)

Kept for

Not stated by the Helpful Places.

Shared with

Available to the accountable organization
  • Weather forecast outputs are available to ECCC employees and MSC forecasters as part of the operational forecasting workflow.
Available to me
  • The register notes that both employees and the public are primary users, indicating that AI-generated forecasts are intended to be publicly accessible once the system moves beyond the current development phase.

Stored

Not stated by the Helpful Places.

How to read the colours

Can it identify you?

Anonymized data
Data about people with the link to who is broken. Stripped of identifiers, blurred, aggregated, or noised so this system can’t reasonably tie a record back to an individual.
Pseudonymous data
Each person’s data is tied to a token (hash, ID, template) that lets this system recognise the same person across events, but the token itself doesn’t reveal a name. Reidentification is possible with extra information.
Identifiable data
The data either contains a direct identifier (name, address, account name, recognisable face or voice, plate number) or carries a token this system uses to look up legal identity during processing.

Who completes the loop?

Human decides
This mode suggests; a person decides what to do next. The AI is always advisory — a human is in the loop on every decision. Example: a triage tool ranks cases for a clinician who chooses which to see first.
Human executes
This mode decides; a person carries out the result. Example: an optimizer plans the day’s trash-collection routes, and drivers run them.
Autonomous
This mode decides and acts on its own. No person reviews each decision or carries out the resulting action.

Definitions from the DTPR standard. Amber is about your data, violet about who decides. The fuller the shape and the deeper the colour, the more identifying the data or the less a person is involved.

What you can do

Ask about this system

Questions go to the Helpful Places, not the vendor.

Your rights

  • Right to Algorithmic TransparencyECCC has published a public AI Roadmap (2023) describing the overall approach to integrating AI into weather forecasting through to 2030. Members of the public may consult the Government of Canada AI Register entry (2526-ECCC-002) for further information about the system's development status and methodology.

Risks and safeguards

  • Environmental harmTraining and running large AI weather models carries a significant compute and energy footprint. ECCC explicitly identifies 'energy-efficient AI weather models' (GraphCast, ForecastNet, AIFS) as a key research priority, indicating awareness of this risk. Mitigation includes selecting architectures optimised for inference efficiency and benchmarking energy use as part of the roadmap evaluation criteria.
  • Societal & cultural harmAI-generated weather forecasts that are inaccurate could erode public trust in meteorological services and affect critical decisions by individuals, businesses, and emergency managers. Mitigation includes the hybrid 'spectral nudging' approach that constrains AI outputs within physically-grounded numerical model frameworks, ensuring AI does not operate independently of established forecasting science.