ESRS E3 — Water, and Why the Metrics Survived the Cull
ESRS E3 is the European Union’s water disclosure standard. It sits alongside E1 on climate, E2 on pollution, E4 on biodiversity and E5 on circular economy, gated by the same double materiality test.
The simplification adopted on 3 July 2026 shrank it twice over. The title lost and marine resources. The fifth disclosure requirement was deleted outright. And yet the substantive metric requirement came through untouched, still asking for six separate figures. That asymmetry is the interesting thing about E3, and §7 explains it: the parts that survived are the parts another regulation depends on.
1. The four disclosure requirements
| DR | What it asks for |
|---|---|
| E3‑1 | Policies related to water. Plus a specific confession: if the undertaking has sites in areas with water stress not covered by those policies, it must say so. |
| E3‑2 | Actions and resources related to water — with key actions and resources for areas with water stress called out specifically. |
| E3‑3 | Targets related to water, expressed where relevant against specific geographies such as areas with water stress. |
| E3‑4 | Water metrics — the six figures. See §4. |
Under ESRS (2023) there was a fifth, E3-5, asking for anticipated financial effects. ESRS (2026) removed it from E3 — along with its counterparts in E2, E4 and E5 — and consolidated those disclosures into ESRS 2. Note the pattern in the four that remain: water stress is named in three of them. The standard is not neutral about where your water comes from.
| ESRS 2023 (2023/2772) | Simplified ESRS (2026/1563) |
|---|---|
| E3-1 Policies related to water and marine resources | E3-1 Policies related to water |
| E3-2 Actions and resources related to water and marine resources | E3-2 Actions and resources related to water |
| E3-3 Targets related to water and marine resources | E3-3 Targets related to water |
| E3-4 Water consumption | E3-4 Water metrics |
| E3-5 Anticipated financial effects from water and marine resources-related impacts, risks and opportunities | Removed |
The simplified standards (Delegated Regulation (EU) 2026/1563 (simplified ESRS)) have been in force since 10 November 2026 and apply to financial years starting on or after 1 January 2027; for FY2026 a reporter may use either set (Art 2). Titles as printed in the Official Journal. Read by GreenCalculus from the Official Journal texts (EUR-Lex / Publications Office Cellar) on 27 October 2026.
Provider directory
Need help reporting this to a regulator?
See who does this work. Every listing names the standards it works to, and paid placements are labelled. Including Verdis Group and Bureau Veritas UK.
Browse 4 regulatory reporting providers →Do this work? A listing is US$390 a year. Get listed →
2. Which version applies
ESRS (2026) was adopted on 3 July 2026 and is mandatory for financial years beginning on or after 1 January 2027. For a financial year beginning in 2026 an undertaking has three options, and it must state which one it took.
- Stay on ESRS (2023), as amended by the Quick Fix Delegated Act — the standard still titled “Water and marine resources”, still carrying E3-5.
- Early-adopt ESRS (2026) in full. Everything on this page applies a year ahead of the mandate.
- Stay on ESRS (2023) and take the named reliefs from ESRS (2026) — materiality method, acquisitions and disposals, non-material activities, value-chain scope, joint operations, Taxonomy presentation, executive summary. All procedural. None of them changes what E3 asks for.
An undertaking on option 3 is still reporting against ESRS (2023) E3 — marine resources included, E3-5 included. The narrowing described on this page arrives only with ESRS (2026) itself. Whichever route is taken has to be disclosed explicitly.
3. Marine resources moved — they did not disappear
The most visible change to E3 is its name. ESRS (2023) called it Water and marine resources; ESRS (2026) calls it Water. It is easy, and wrong, to read that as marine resources dropping out of European sustainability reporting.
What E3 kept is water itself, and the standard is deliberately broad about what that means. Water encompasses freshwater and other types — brackish water, for example — from surface water, groundwater, seawater, produced water and third-party water. Seawater taken in for cooling is water. Produced water from extraction is water.
4. The six metrics
E3-4 is the quantitative core, and it is more explicit than its predecessor. The undertaking discloses the following for its own operations, all in cubic metres or multiples thereof.
| Metric | Note |
|---|---|
| Total water consumption | The headline figure. Not the same as withdrawal — see §5. |
| Total water consumption in areas with water stress | The same number, filtered by geography. Requires site-level location work. |
| Total water withdrawal | What you took. Usually the metered figure. |
| Total water discharge | What you returned. |
| Total water recycled and reused | Anchored to an SFDR indicator — see §7. |
| Total water stored | The one most often forgotten. Reservoirs, ponds, tanks. |
ESRS (2023) asked for a water intensity ratio — total water consumption per unit of net revenue. It is not in the ESRS (2026) list. If your reporting pack computes an intensity figure, it is now voluntary context rather than a required disclosure, and it should not be presented as though the standard demands it.
5. Consumption is not withdrawal
This is the single most common error in water reporting, and ESRS (2026) writes the arithmetic into the standard so it cannot be missed. Consumption may be measured directly, modelled, estimated, or calculated as:
6. Water stress — what it is, and what it is not
Water stress is named in three of E3’s four disclosure requirements, so getting the concept right is not optional. Two clarifications in the standard are worth stating plainly.
For the quantity dimension, the standard names four indicators and their thresholds:
| Indicator | Threshold for “area with water stress” |
|---|---|
| Baseline water stress | Equal to or greater than ‘High’: 40–80% |
| Water depletion | Greater than ‘High’: 25–75% (seasonal) |
| Baseline water depletion | Equal to or greater than ‘High’: 50–75% |
| WEI+ (Water Exploitation Index plus) | Equal to or greater than 40% |
Note what is not in that list: a mandated data provider. The adopted text gives indicator names and thresholds and leaves the dataset to you. It adds that the assessment is typically done at basin level as a minimum, and that tailored methodologies may be used and may draw on local knowledge. So a defensible answer needs the indicator, the dataset, the vintage and the basin resolution recorded — none of which the standard picks for you.
7. Why these metrics survived the cull
ESRS (2026) removed a great deal. It deleted the anticipated financial effects requirement from four environmental standards, made much of E4 conditional, and cut mandatory datapoints across the set. Against that, E3-4 still asks for six separate figures and E3-1 still carries a specific confession requirement. Why?
The adopted text answers it in footnotes. Three of E3’s disclosures are marked as supporting the information needs of financial market participants under the Sustainable Finance Disclosure Regulation, because each is derived from a mandatory principal adverse impact indicator.
| E3 disclosure | SFDR PAI indicator | What the investor is computing |
|---|---|---|
| E3-1 — water policies | #7 | Investments in companies without water management policies |
| E3-1 — stressed sites not covered by policies | #8 | Exposure to areas of high water stress |
| E3-4 — water recycled and reused | #6.2 | Weighted average percentage of water recycled and reused by investee companies |
Two practical consequences follow, and they are worth more than the observation itself.
- These three are the least likely to be simplified away next time. A requirement that another regulation is built on cannot be removed unilaterally. If you are choosing where to invest in data quality, start here.
- The confession requirement in E3-1 is doing real work. Disclosing that you have sites in water-stressed areas not covered by your water policies feeds PAI indicator #8 directly. It reads like an administrative footnote; it is in fact the number an investor screens on.
8. What the work actually looks like
Assuming water is material, this is the honest sequence. Note how much of it is location work rather than metering work — that is the shape of E3.
- Settle materiality. E3 is gated by double materiality. A well-documented negative conclusion is a legitimate outcome, and the reasoning is itself disclosable.
- Geolocate every site. Everything downstream — the stressed-area consumption figure, the policy-coverage confession, the geographic targets — depends on knowing where your sites are, to basin resolution.
- Choose an indicator and a dataset, and record both. The standard names four indicators and no vendor. Your choice, its vintage and the basin resolution are all part of a defensible answer.
- Assemble withdrawal and discharge before consumption. Consumption is usually derived from the other two, so its quality is capped by theirs.
- Do not forget stored water. It is a required metric and it lives in nobody’s utility bill.
- Test policy coverage against the stressed-site list. Any gap is a disclosure, not an embarrassment to be resolved before publication.
- Write policies, actions and targets last. Written before the site map exists, geographic targets are aspiration rather than disclosure.
9. Where E3 sits against the rest of EU water law
Like E2, E3 creates no operational obligations. It creates disclosure obligations about a position other law already governs, and the standard names the frameworks it was written against.
| Instrument | Relationship to E3 |
|---|---|
| Water Framework Directive (2000/60/EC) | The basin-level architecture E3’s geography language assumes. |
| Drinking Water Directive (2020/2184/EU) | Quality standards for the supply side of third-party water. |
| Industrial Emissions Directive (2010/75/EU) | Permits and consented discharge points — a primary source for the discharge figure. |
| EU Water Resilience Strategy | The policy direction the standard was drafted alongside. |
| EMAS Regulation (1221/2009/EC) | Existing environmental management reporting many reporters already run. |
| UN Watercourses Convention (1997) and UNECE Water Convention (1992) | The transboundary water law backdrop for basin-level thinking. |
10. What E3 does not do
- It does not cover what you put into water. Discharges and emissions to water, including to seawater, are E2. E3 covers water as a resource. A single site can easily have material disclosures under both.
- It does not cover marine ecosystems. Those moved to E4.
- It does not cover marine resources. Those moved to E5, as resource inflows.
- It does not cover water-related physical climate risk. Flooding, drought, sea level rise, saline intrusion and the rest sit in E1 as acute and chronic physical risks.
- It does not require a water intensity ratio. ESRS (2023) did; ESRS (2026) does not.
- It does not set abstraction limits. E3 is disclosure. Your licence is your licence.
Frequently asked questions
No. ESRS E3 was retitled from “Water and marine resources” to simply “Water”, but the marine content moved rather than disappearing. Impacts on marine and freshwater ecosystems and biodiversity connected to water use are addressed by ESRS E4, and marine resources are addressed by ESRS E5 as a category of resource inflows. An undertaking that stops reporting marine impacts because E3 changed its name has misread the change.
Four under ESRS (2026): E3-1 policies, E3-2 actions and resources, E3-3 targets, and E3-4 water metrics. ESRS (2023) had a fifth, E3-5 anticipated financial effects. That requirement was removed from E3 — and from E2, E4 and E5 — and consolidated into ESRS 2. ESRS E1 keeps its own anticipated financial effects disclosure.
Withdrawal is what you take; discharge is what you return; consumption is what does not come back. The standard states the relationship directly: consumption equals withdrawal minus discharge over the reporting period, and it may also be measured directly, modelled or estimated. Most undertakings meter withdrawal and derive consumption, which means the quality of the consumption figure is capped by the quality of the discharge figure.
Six, for own operations, in cubic metres or multiples: total water consumption; total water consumption in areas with water stress; total water withdrawal; total water discharge; total water recycled and reused; and total water stored. The water intensity ratio that ESRS (2023) required is not among them.
The standard names four indicators and their thresholds: baseline water stress at or above ‘High’ (40–80%); water depletion above ‘High’ (25–75%, seasonal); baseline water depletion at or above ‘High’ (50–75%); and WEI+ at or above 40%. It does not mandate a data provider. The assessment is typically done at basin level as a minimum, and tailored methodologies may be used, including local knowledge.
No, and the standard is explicit about it. Water stress is one component of physical water-related risk, alongside factors not related to stress such as flooding and droughts, and a full understanding also takes in regulatory and reputational risk. Water stress itself spans quantity, quality and accessibility, while the global indicators capture only the quantity dimension. Screening for water stress is not the same as screening for water risk.
Because other regulation depends on them. The adopted text carries footnotes marking three E3 disclosures as supporting financial market participants under the SFDR, each derived from a mandatory principal adverse impact indicator: water policies feed indicator #7, sites in water-stressed areas not covered by policies feed indicator #8, and water recycled and reused feeds indicator #6.2. A requirement another regulation is built on cannot be removed unilaterally.
Yes. The standard defines water broadly — freshwater and other types such as brackish water, from sources including surface water, groundwater, seawater, produced water and third-party water. Seawater taken in for cooling counts, as does produced water from extraction. What E3 does not cover is what you discharge into the sea, which is ESRS E2.
Related references
