Skip to main content

You are viewing a project that is currently in draft state for the Standards Working Group in the Green Software Foundation. This project should not be considered finished or officially supported in any way by the Green Software Foundation or its members.

Data Quality and Implementation Tiers

Data Quality and Implementation Tiers

In short: not all water data is equally trustworthy, and the spec doesn't hand implementers a single approved number to use. This page gives a three-tier way to judge whatever source you do use, and a disclosure rule for each tier, so different reports stay comparable even when their underlying data quality differs.

The specification tells implementers what to quantify. It deliberately does not tell them exactly which dataset to use for every input — data availability varies too much by region and by generation type for a single prescribed source to work everywhere. This guidance instead provides a tiered framework for evaluating whatever data source is chosen, modelled directly on how the SCI framework handles grid carbon intensity data quality: same shape of problem, same shape of answer.

The three tiers​

TierDescriptionDisclosure requirement
Tier 1Plant-level measured water consumption/withdrawal data (e.g. EIA-923 forms)Preferred; disclose source and time period
Tier 2Regional grid-level data derived from plant-level aggregationAcceptable; disclose methodology and source
Tier 3Default NREL 2011 intensity factorsPermitted as fallback; must be explicitly disclosed

This tiering applies most directly to EWIF (see EWIF — Data Sources and Implementation Guidance for the tier-by-tier source mapping), but the same logic — prefer measured, plant-level data; accept aggregated regional data with disclosure; permit a disclosed default as a last resort — applies wherever an SWI assessment needs a water intensity or stress figure and cannot measure it directly.

No fixed lookup table​

GSF does not maintain, and has no plan to maintain, a single fixed set of water-intensity numbers for implementers to look up and plug in. This is a deliberate choice, not an oversight: water data availability, quality, and methodology are evolving quickly enough across regions that a static lookup table would go stale and could mislead implementers into treating a snapshot as ground truth. This guidance instead points to methodology and current data sources, and expects implementers to select and disclose their own source for each input.

Region-specific notes​

RegionStatus
United StatesTier 1 data available now via EIA-923 (monthly and annual).
European UnionNo equivalent Tier 1 source yet; expected around 2028.
IndiaData availability improving as relevant legislation advances, ahead of most other regions outside the US.

Where an implementer is working in a region without an available Tier 1 or Tier 2 source, falling back to Tier 3 (disclosed) is expected and permitted — the tiering exists to make that fallback visible and comparable across reports, not to penalise implementers for data gaps that are, at present, outside their control.


Please submit any comments you have here.