By: Johannes Fiegenbaum on 7/29/25, 11:26 AM · Last updated September 5, 2026
A supplier does not have to be in a disaster zone to stop delivering. Flood, drought, wildfire and heat hit the sites your suppliers depend on, and the effect reaches you as a longer lead time or a production stop. This guide covers how to map that exposure across supplier tiers, measure it at site level, keep it under review, and report it under CSRD.
Live data: see how physical climate hazards such as flooding, drought and heat hit specific locations in the Fiegenbaum Atlas climate risk dashboard.
Tier 1 suppliers are the ones you contract with and can name. Tier 2 and sub-tier suppliers are the ones your Tier 1 partners depend on, and that is where most physical exposure sits unseen. Two patterns turn an ordinary hazard into a business risk: single sourcing and site concentration. If four Tier 1 suppliers buy the same input from one Tier 2 site, you have one point of failure.
| Hazard | Where it usually sits | Exposure indicator | Public data source |
|---|---|---|---|
| Flooding | Tier 1 plants, logistics hubs | Site in a 100-year flood zone, closure days | EEA flood maps, flood cadastres |
| Heat | Tier 2 process industry, warehousing | Days above 30 degrees, cooling-dependent steps | Copernicus C3S, DWD |
| Drought | Sub-tier water and hydropower dependent | Low-flow days, water stress | EU Joint Research Centre |
| Wildfire | Sub-tier raw material, remote grid links | Burnt area probability | Copernicus EFFIS |
| Storm and hail | Tier 1 sites, open stock, rooftop solar | Historic hail days, gust return periods | National weather services |
| Transition pressure | Energy-intensive Tier 2 suppliers | Energy intensity, carbon price pass-through | EU ETS data |
Energy dependency is the example most maps miss. A Tier 1 assembly supplier looks unexposed, but its aluminium comes from a smelter whose power rests on one hydropower region. A dry summer lifts spot prices, the smelter throttles output, and the cut reaches your line weeks later. The hazard is drought, the exposed asset is in Tier 2, and the damage is measured in lead time. Geopolitical risk compounds the same weak points, because a single-source site under climate stress is also the one with no alternative.
Exposure is measured per site, not per company. A supplier with twelve plants has twelve risk profiles, and the group average hides all of them. Five steps:
Suppliers hold the data you cannot buy: backup power, water permits, whether the site has flooded before. Long ESG questionnaires fail here, because they land with someone who has no reason to answer. Two operational questions do get answers: which exact site produces our part, and what happened the last time production stopped there.
My position: automated supplier climate scoring is a starting point, not an audit-proof result. Screening tools are useful for narrowing hundreds of suppliers to the few dozen worth a real look. They are not evidence. Any score that ends up in a report needs a named data source, a stated scenario and a documented threshold behind it.
A one-off assessment ages badly in the parts that change fastest and not at all in the parts everyone re-runs out of habit. Sites move and sourcing shares shift, hazard projections barely budge.
| What you review | Trigger | Frequency |
|---|---|---|
| Tier 1 site addresses | New supplier, relocation | Yearly |
| Single-source and concentration list | Sourcing decision | Quarterly |
| Downtime and lead time incidents | Any disruption | After each event |
| Adaptation measures at critical suppliers | Audit, contract renewal | Yearly |
| Hazard scores per site | New climate dataset release | Every three years |
| Scenario set and thresholds | New IPCC assessment cycle | Not in between |
After an event, re-check three things and nothing else: how long the supplier needed to restart, whether the outage matched the modelled downtime, and whether an alternative source was genuinely available. Near misses count too. A supplier that stayed up because a levee held is exposed, not resilient.
Supplier exposure is a named disclosure, not a side note. Under the ESRS delegated regulation, ESRS E1 asks for a climate resilience analysis covering the value chain and for the anticipated financial effects of physical and transition risks. That is what a site-level supplier assessment produces: which hazard, which site, which scenario, what it costs. Scoping guidance comes from EFRAG.
The EU Taxonomy pulls on the same data. The do no significant harm criterion for climate adaptation in the Climate Delegated Act requires a climate risk and vulnerability assessment for the activity, including relevant parts of the value chain. One assessment serves both.
On materiality, the practical trigger is concentration. A hazard at a supplier carrying a large share of one input, with no qualified alternative, crosses the line even when the absolute exposure looks modest. The reporting obligation itself now applies to companies above 1,000 employees and 450 million euros net turnover, for financial years from 2027.
A screening result becomes useful when each prioritised risk has a measure, an owner and a number. Cut the ranked list where downtime or single-source share stops being tolerable, and give every remaining item one accountable person in procurement plus a budget line. Measures that earn their cost: dual sourcing for concentrated inputs, contractual restart and notification duties, buffer stock sized to the modelled downtime, and physical protection at the supplier site where a joint investment beats a switch.
Concentration is what makes this affordable. In a climate risk analysis for a southern German dealer group with 18 sites, around 88 percent of the quantified exposure was physical, driven mainly by hail and heavy rain, and hail had already been documented at seven of the 18 locations. The top five sites carried 35 percent of the physical risk. Supplier portfolios have the same shape, so treating every supplier equally is the most reliable way to spend a budget without reducing risk.
Where the exposure sits with a supplier rather than with you, the plan item is a conversation, not a capital request: ask what is already in place, then put the answer in the contract. For tracking measures and evidence, see climate risk management technology.
Measure per site, not per supplier. Collect coordinates for the sites behind your critical parts, pull hazard indicators for each from Copernicus, the EEA or the national weather service, use an intermediate and a high emissions scenario, and convert each hazard into an operational number: downtime days, lead time extension, single-sourced share affected.
Rank by replaceability, not by spend. Start with single-sourced inputs, then suppliers whose sites cluster in one region, then parts with long qualification times for an alternative. A first round of 20 to 30 suppliers chosen this way covers most of the exposure that can stop production.
Not all of it at one interval. Review supplier sites and adaptation status yearly, concentration and single-source shares quarterly, and log every disruption against the modelled downtime. Hazard scores only need recalculating when the underlying dataset is updated, the scenario set only with a new IPCC cycle.
ESG and sustainability consultant based in Hamburg, specialised in VSME reporting and climate risk analysis. Has supported 300+ projects for companies and financial institutions, from mid-sized manufacturers to major banks and insurers.
More aboutA supply chain biodiversity footprint is the pressure your purchased goods place on ecosystems at the places where they are grown, mined or processed. Almost all of it sits ...
Read more →