Water Scarcity & Biodiversity Loss: Climate Risks & Corporate Strategy

Water scarcity and biodiversity loss reach a company through two channels, not one. They arrive as physical risk, when a site or a supplier loses reliable water, and as reporting risk, when ESRS E3 and E4 turn material and the datapoints have to be filled with something defensible.

The causal chain between the two is short. Less water in a river or an aquifer means less habitat, lower flow, higher water temperatures and less dissolved oxygen. Species that cannot move or tolerate the change disappear, and the services that depended on them, pollination, natural water purification, flood buffering, weaken with them. Up to 85% of wetlands have already been lost, although they store twice as much carbon as all the world's forests combined (Ramsar Convention), and the IPBES Global Assessment puts up to one million species at risk of extinction.

Live data: See how physical climate hazards such as flooding, drought and heat hit specific locations in the Fiegenbaum Atlas climate risk dashboard.

Water Scarcity and Biodiversity Loss as Corporate Risks

Both topics behave like climate risk, not like nature conservation. On the physical side, a dry summer restricts cooling, process water and inland shipping; on the transition side, permits, abstraction rights and river basin obligations tighten around the same locations. In 2022, one third (34%) of EU territory was affected by water scarcity for at least one season, and the risk stayed high even though withdrawals fell by 19% between 2000 and 2022 (EEA).

Germany shows how industrial the exposure is. Energy companies use 56% of river water for cooling, mining and industry account for another 18%, and groundwater supplies 70% of drinking water (German Environment Agency). A plant that shares an aquifer with a public water supplier is competing for a resource whose allocation is a regulatory decision, not a market one. The logic is that of a physical and transition risk assessment, applied to water and land instead of temperature.

The practical difference is the unit of analysis. Carbon is measured at company level and aggregates cleanly. Water and biodiversity are decided at the level of a single basin or a single hectare, which means a group figure says almost nothing and a site list says almost everything.

What ESRS E3 and E4 Actually Require

ESRS E3 covers water and marine resources, ESRS E4 biodiversity and ecosystems. Both are conditional: they only produce disclosures once the double materiality assessment finds them material, but the assessment itself has to be described under ESRS 2 IRO-1 whatever the outcome. The EU Water Framework Directive supplies the basin classification that E3 leans on, and the Nature Restoration Regulation is why E4 data requests will keep growing: member states had to submit draft national restoration plans to the Commission by 1 September 2026.

Read the right-hand column again, because it is the whole difficulty. Almost every E3 and E4 datapoint needs a location, and most reporting systems are built around legal entities and cost centres instead. My position: E3 and E4 are not hard standards, they are a data architecture problem wearing a disclosure costume. Companies that already maintain a geocoded site and sourcing register fill them in weeks; companies that do not spend a full cycle rebuilding their master data, and no amount of ESRS reading substitutes for that.

Screening Water and Biodiversity Risk in a Double Materiality Assessment

The screening sequence below is the one I run, and it is deliberately ordered so the cheap exclusions happen first.

1. Build the location list before anything else. Own sites, leased sites, warehouses, and the sourcing regions for the three highest volume inputs. Addresses are enough to start; coordinates are needed to finish.
2. Run a basin level water stress screen. Classify each location against the river basin it sits in, using the water exploitation index and the relevant river basin management plan under the Water Framework Directive. Basin, not country: a German site can sit in a stressed basin while the national figure looks comfortable.
3. Overlay protected and sensitive areas. Check each location against Natura 2000 sites, national protected areas and key biodiversity areas. Proximity counts, not just overlap, because E4-5 asks about sites in or near such areas.
4. Add the pressure, not just the place. For each flagged location, name the actual pressure: abstraction volume, thermal discharge, nutrient or chemical load, land sealing, or habitat conversion upstream in the supply chain.
5. Write the IRO in one sentence per finding. Impact or risk, location, pressure, time horizon, and financial channel. If a finding cannot be written that way, it is not yet a finding.

Where does this fail in practice? Almost never at strategy level. It fails at step 1 and step 4: nobody owns a complete site list, and the water intensive step turns out to sit two tiers upstream in an irrigated crop or a wet chemical process that the company does not operate and has never geocoded. Teams then discuss materiality thresholds for weeks while the missing input is a spreadsheet of addresses. A structured water risk assessment and a site level climate risk assessment should therefore share one location register, not two.

Concluding that water and biodiversity are not material is legitimate and common, especially for service companies without production sites. It is only defensible if it is documented rather than asserted. Record the location list you screened, the basin classification and protected area overlay you used with their vintage, the pressures you tested and ruled out, and the threshold that the findings failed to reach. That record is the ESRS 2 IRO-1 disclosure, and it is also what an auditor asks for first. Aligning the method with ISO 14091 keeps the water part consistent with the climate risk work instead of running as a parallel exercise.

FAQs