Styxis
Company
Knowledge

Look beyond nameplate capacity to compare hydrogen water use

Compare water intensity with actual runtime, output, water source and abstraction/cooling boundaries intact.

Capacity hides operating conditions

Two hydrogen facilities with the same nameplate capacity can use water under different runtimes, sources and cooling boundaries. Capacity alone cannot represent actual consumption. Truthound can check a prepared definition table, not judge a plant's water risk or location.

The data need in the JRC study

The JRC's October 8, 2026 account examines hydrogen deployment and basin-level freshwater pressure. It states a freshwater assumption and calls for consistent data on actual runtime, capacity, abstraction and cooling. Conditional modelling is not an observation of every facility.

Keep facility and basin together

Retain the facility-to-basin relationship. A national average should not replace the water context of the actual site. Record coordinate and basin-boundary versions so later changes identify which relationship needs review.

Abstraction is not consumption

Separate abstraction from consumption and process water from cooling water. Preserve units and aggregation periods. For a water-per-kilogram calculation, establish whether the denominator is actual output during the period or nominal potential production.

The unit relationship in Truthound

Truthound Core Python SDK 3.3.4, operating on a prepared Polars table, provides ExpressionValidator for caller-defined Polars row expressions and ColumnPairInSetValidator for supplied allowed pairs. An owner may define a source/measure association or intensity formula, including the policy for a denominator of 0 and for nulls. The software does not supply those physical meanings.

Keep the existing calculation when useful

An existing Pandera 0.35.0 pandas expression can remain. Compare mismatched units, nominal capacity substituted for output and incompatible periods using the same examples. Documented expression support does not establish identical water-intensity validation.

Daily efficiency and seasonal stress differ

Daily operating efficiency and seasonal basin stress have different scopes. A low daily ratio should not become a statement about long-term resource availability. Keep observation windows and scenario assumptions beside any seasonal relationship.

Label observations and assumptions

Label measurements separately from planning assumptions. Filling an absent runtime with maximum available hours can change the intensity meaning. Where estimation is needed, preserve the original unobserved state and owner-approved basis.

Ask the next field question

The useful artifact shows what was actually observed and what remains unknown, rather than ranking facilities automatically. It should identify the next question about basin, source or runtime. Data retaining field meaning can then support the specialist decision.

Conditions

Verify facility/basin relationships, authorised operating data and measurement boundaries.

Alternatives

Retain an existing Pandera 0.35.0 pandas expression with the same water-intensity definitions.

Limitations

  • Input checks do not decide plant siting, water safety or future resource risk.

Evidence

Event checked
Subject checked
Modified