The TNFD Locate step is where a country-level footprint becomes a defensible list of priority locations. Here is how to run it, sub-step by sub-step, and the disclosure it produces.
The Taskforce on Nature-related Financial Disclosures (TNFD, the voluntary framework whose final recommendations landed in September 2023) asks a question climate reporting usually lets you skip: where, exactly, does your business touch nature? You cannot describe a dependency on water, soil, or pollination, or an impact on a habitat, until you know which of your sites and suppliers sit on top of the nature in question. Nature risk is site-specific. A drought that idles one sourcing region says nothing about a facility two valleys over.
That is why TNFD’s recommended assessment method, LEAP (Locate, Evaluate, Assess, Prepare, run in that order), starts with Locate. Locate is the step that turns “we operate in 40 countries” into a defensible list of the specific places where nature matters to the business, and the business matters to nature.
This is a walkthrough of that first step as a reporter actually runs it. It assumes you have already read the TNFD LEAP approach guide for the full four-phase picture, and it does not repeat the framework overview. Instead it does one thing that guide leaves open: it follows Locate end to end and shows the disclosure it produces, Strategy Disclosure D, the priority-locations disclosure.
- The TNFD Locate step turns a country list into a defensible set of priority locations. It runs as a funnel of four sub-steps, L1 to L4.
- L4 tests every material site against five sensitive-location criteria. Only one of the five needs to be met for a site to count as sensitive.
- Sensitive is not the same as priority. Priority locations are the union of sensitive locations and material locations, and only priority locations go into Strategy Disclosure D.
- A location screen settles four of the five criteria at a screening level. Key Biodiversity Area overlap, species data, and ecosystem-service provision stay specialist work.
Locate runs in four sub-steps, L1 to L4
TNFD frames Locate as four guiding questions. Read them as a funnel: each one narrows the last.
- L1, span of the business. What are your activities by sector, value chain, and geography, and where are your direct operations?
- L2, dependency and impact screening. Which of those sectors, value chains, and operations carry potentially moderate or high dependencies and impacts on nature?
- L3, interface with nature. Where are those moderate-and-high activities located, and which biomes and specific ecosystems do they touch?
- L4, interface with sensitive locations. Of those, which sit in ecologically sensitive locations?
The funnel matters because the alternative, screening every site on Earth against every nature dataset, is neither affordable nor meaningful. L1 and L2 exist to shrink the problem before you spend money on spatial analysis in L3 and L4.
| Sub-step | Guiding question | What it produces |
|---|---|---|
| L1 Span of the business | What are your activities by sector, value chain and geography, and where are your direct operations? | A mapped footprint of direct operations, plus as much of the value chain as you can name |
| L2 Dependency and impact screening | Which sectors, value chains and operations carry potentially moderate or high dependencies and impacts on nature? | The subset of the business worth assessing in detail |
| L3 Interface with nature | Which biomes and specific ecosystems do those activities interface with? | A business geography converted into an ecological geography |
| L4 Interface with sensitive locations | Which of those activities sit in ecologically sensitive locations? | A list of sensitive locations, tested against all five criteria |
L1: map the footprint, and expect the value chain to be the hard part
L1 asks for your activities by sector, value chain, and geography. For direct operations this is usually a solved problem: you know your plants, farms, mines, and offices, and you have their addresses or coordinates.
The value chain is where L1 gets real. TNFD asks for upstream and downstream activities “where possible,” and for most sectors the material nature story lives upstream, at the sourcing tier you do not own. A food manufacturer’s water and land footprint sits in the fields that grow its inputs, not in the factory. So L1 is less a mapping exercise than a sourcing-visibility exercise: how far up your supply chain can you actually name a location? Where you cannot, you carry that as a known limit into the later steps rather than pretending the site does not exist.
L2: screen for where nature is likely to be material, by sector
L2 narrows the footprint to the activities that plausibly carry moderate or high dependencies and impacts on nature. This is a screening judgment, not a measurement. You are not yet quantifying anything; you are deciding which parts of the business are worth assessing in detail.
Two structures help. The first is the set of five drivers of nature change that TNFD borrows from the global science body on biodiversity (IPBES): climate change; land, freshwater, and ocean use change; resource use and replenishment; pollution; and invasive alien species. Every impact your business has on nature runs through one of these five. The second is sector-level dependency screening, the approach behind tools such as ENCORE (a database that maps how economic activities depend on nature’s services). Sector screening tells you, before you look at a single map, that a cement operation’s dependency profile differs from a software firm’s.
One boundary is worth stating here, because it shapes what the later steps can deliver: dependency screening at the sector level is well supported, but dependency quantification at a specific site (how many cubic meters of a service, worth how much) leans on external economic taxonomies rather than anything you can read off a satellite. Location analysis answers the impact and exposure side of L4 directly. The dependency-value side is a modeled add-on. Keep the two apart in your own head so you do not over-promise the screen.
L3: place the material activities in their biomes and ecosystems
L3 takes the activities that survived L2 and asks where they are and what they touch. Which biomes (broad ecosystem types such as temperate forest or freshwater wetland) and which specific ecosystems does each interface with?
This is the step that converts a business geography into an ecological geography. A vineyard and a data center in the same county interface with nature differently, and L3 is where that distinction enters the assessment. TNFD organizes the whole exercise around four realms: land, freshwater, ocean, and atmosphere. Most corporate footprints resolve mainly to land and freshwater, which is also where spatial data is strongest; ocean and atmosphere are thinner for most reporters.
L4: run the sensitive-location screen against five criteria
L4 is the heart of Locate and the longest part of the work. For each material activity from L2, placed by L3, you ask: is it in an ecologically sensitive location?
TNFD defines a sensitive location as one that meets one or more of five criteria. Only one needs to be met.
- Areas important for biodiversity, including species.
- Areas of high ecosystem integrity.
- Areas of rapid decline in ecosystem integrity.
- Areas of high physical water risk.
- Areas of importance for ecosystem-service provision, including benefits to Indigenous Peoples and Local Communities (IPLC).
The practitioner’s task is to test every asset against all five, and TNFD gives you a starting toolkit: a table of reference datasets for each criterion. What follows is that table, criterion by criterion, with a note on what each dataset actually tells you. TNFD is explicit that these are example datasets for consistency, not a mandated or exhaustive list, and that judgment, local data, and stakeholder engagement matter as much as any global layer. There is no formula that turns this into a checkbox.
| Criterion | TNFD example reference datasets (Table 4) | What it tells you | Global data maturity |
|---|---|---|---|
| 1. Areas important for biodiversity, including species | WDPA and OECMs; Key Biodiversity Areas; IUCN Red List of Threatened Species; STAR | Whether a site sits in or near a protected area, a Key Biodiversity Area, or a threatened species’ range | Protected areas mature; species and KBA overlap need specialist datasets |
| 2. Areas of high ecosystem integrity | Ecosystem Integrity Index; Biodiversity Intactness Index; IUCN Red List of Ecosystems | How intact the surrounding ecosystem still is | Only partly observable; global condition indices carry real uncertainty |
| 3. Areas of rapid decline in ecosystem integrity | The same integrity indices, plus area- and health-based proxies | How fast habitat is being lost | Good directional proxy in land-cover change and deforestation |
| 4. Areas of high physical water risk | WRI Aqueduct Water Risk Atlas; WWF Water Risk Filter | Whether a site sits in a water-stressed basin | Available and global, at basin level |
| 5. Areas of importance for ecosystem-service provision, including IPLC benefits | LandMark and registered ICCAs; InVEST; TESSA; ENCORE; critical-natural-asset mapping | Where healthy ecosystems underpin local livelihoods and IPLC rights | Least reducible to a spatial test; needs rights data and field work |
Criterion 1, biodiversity importance: protected areas are easy, species are hard
TNFD points to the World Database on Protected Areas (WDPA) and other effective area-based conservation measures (OECMs) for the legally protected limb, plus Key Biodiversity Areas (KBAs) and their sub-types, the IUCN Red List of Threatened Species, the STAR metric, and marine datasets for the scientific-importance and species limbs.
In practice this criterion splits into two very different jobs. Testing whether an asset sits in or near a designated protected area is well-served: WDPA is global, maintained, and spatially precise. Testing whether an asset overlaps a Key Biodiversity Area or the range of a threatened species is harder, needs specialist datasets, and is where most reporters reach for a dedicated biodiversity tool.
Criteria 2 and 3, ecosystem integrity: the condition question global data answers least well
TNFD groups high integrity and rapid decline together and points to integrity indices such as the Ecosystem Integrity Index (EII), the Biodiversity Intactness Index (BII), and the IUCN Red List of Ecosystems, plus area- and health-based proxies. It defines integrity as how far an ecosystem’s composition, structure, and function still fall within their natural range, assessed at landscape scale, “such as an ecoregion.”
This is the criterion where careful reporters slow down. Ecosystem extent (how much habitat is there) and change (how fast is it being lost) are observable from satellites. Ecosystem condition (how intact is what remains) is only partly observable, and the global condition indices carry real uncertainty. A practical pattern is to use land-cover change and deforestation as a directional signal for criterion 3 (rapid decline), and to treat a true intactness score for criterion 2 as a specialist input rather than something a location screen hands you.
Criterion 4, water physical risk: available and basin-level
TNFD points to the WRI Aqueduct Water Risk Atlas and the WWF Water Risk Filter. These are basin-level global layers that flag water stress, flood, and quality risk. This criterion is one of the more tractable ones: the data exists, it is global, and it maps cleanly onto a physical-risk reading of nature dependency. The caution is resolution. Basin-level risk is a flag that a site sits in a stressed watershed, not a measurement of that site’s own water balance.
Criterion 5, ecosystem-service provision: the criterion global spatial data serves least
TNFD points to datasets on Indigenous and community lands (such as LandMark and registered ICCAs), ecosystem-service models (InVEST, TESSA), the ENCORE natural-capital-depletion layers, and critical-natural-asset mapping.
This is the criterion least reducible to an off-the-shelf spatial test. It asks where healthy ecosystems underpin local livelihoods and the rights of Indigenous Peoples and Local Communities, and answering it well depends on rights data, stakeholder engagement, and local knowledge more than on any global raster. Most reporters treat criterion 5 as the one that cannot be screened remotely and must be worked on the ground.
Sensitive is not the same as priority, and only priority gets disclosed
Here is the distinction most first-time reporters miss, and it changes what you disclose.
L4 produces sensitive locations: places that meet at least one of the five ecological criteria. But sensitive locations are not what Strategy Disclosure D asks you to report. TNFD defines priority locations as the union of two sets: sensitive locations and material locations, the latter being places where the organization has identified material nature-related dependencies, impacts, risks, or opportunities. A location can be ecologically sensitive without being material to your business, and material to your business without being ecologically sensitive. Priority locations are the ones that are either.
Two consequences follow. First, you do not finish your priority-locations list inside Locate. Materiality is settled in the next LEAP phase, Evaluate, so the priority set is provisional until then, and TNFD asks you to revisit which locations qualify after Evaluate and at each reassessment. Second, the disclosure itself is precise about scope. Strategy Disclosure D, which is TNFD-specific and has no equivalent in the older climate framework it otherwise mirrors, asks you to “disclose the locations of assets and/or activities in the organisation’s direct operations and, where possible, upstream and downstream value chain(s) that meet the criteria for priority locations.” Note what Disclosure D is not: it is not the resilience disclosure (that is Strategy Disclosure C) and it is not a duty to publish every sensitive site regardless of materiality.
The output of Locate, then, is not a single number. It is a documented set: a map or list of assessment locations, the subset flagged sensitive against each of the five criteria with the data behind each flag, and the reasoning that will let Evaluate turn that into the priority-locations disclosure.
What a location screen can and cannot settle
Step back from the five criteria and a pattern is visible. Global spatial data serves them unevenly. Protected-area proximity (criterion 1’s first limb) and water risk (criterion 4) are well-covered by mature global datasets. Rapid decline (criterion 3) has a good directional proxy in deforestation and land-cover change. True ecosystem integrity (criterion 2), Key Biodiversity Area and species overlap (criterion 1’s second limb), and ecosystem-service provision to local communities (criterion 5) are the parts that global layers serve least well and that pull the assessment toward specialist datasets and on-the-ground work.
This is not a flaw in any one provider. It is the shape of the available science, and a reporter who names it, screening what is screenable and marking the rest as specialist or field work, produces a more defensible Locate step than one who implies a single tool closes all five criteria. The reporters who get Locate wrong tend to make one of two errors: they stop at direct operations and never attempt the value chain, or they treat a protected-area overlay as if it had answered all five criteria at once.
| Criterion | Screened? | Layer | What it does and does not settle |
|---|---|---|---|
| 1. Biodiversity importance | Partial | WDPA protected-area proximity | Protected-area proximity, yes; Key Biodiversity Area and species overlap, no |
| 2. High ecosystem integrity | Context only | RESOLVE ecoregions | Ecoregion context, not a site-level intactness score |
| 3. Rapid decline | Proxy | Hansen land-cover change | Tree-canopy loss as a directional signal, not full ecosystem condition |
| 4. Physical water risk | Yes | WRI Aqueduct | A basin-level water-stress flag |
| 5. Ecosystem-service provision (IPLC) | No | none | Needs rights data, stakeholder engagement, and field work |
One way to run the screen at portfolio scale
For the parts of L4 that global spatial data does serve, the bottleneck is usually volume: running the same overlay for hundreds or thousands of assets and suppliers, consistently, with an audit trail. This is the narrow job Continuuiti’s location analysis does. Given a set of addresses or coordinates, it returns, per asset, protected-area proximity from WDPA (with a distance buffer and the protected area’s IUCN management category), ecoregion and conservation-priority context from RESOLVE, deforestation and land-use change from Hansen Global Forest Change, and water-stress flags. Those are the same industry-standard layers TNFD’s own reference table points to.
Placed against the five criteria, that covers four of them at a screening level: protected-area proximity as a biodiversity-importance signal (criterion 1’s first limb), ecoregion as an integrity context rather than an intactness score (criterion 2), deforestation as a rapid-decline signal (criterion 3), and water stress (criterion 4). It does not provide Key Biodiversity Area overlap, species data, or the ecosystem-service-provision information behind criterion 5, which a full assessment adds from specialist sources. And the deforestation layer measures tree-canopy loss, a directional read on ecosystem change, not a complete measure of ecosystem condition. It is the foundational location screen, sized to run across a whole footprint, with the boundary of what it does not answer stated up front.
To make that concrete without naming a real reporter: picture a food company with a sourcing region in a water-stressed basin. The screen places every known sourcing site, flags the ones inside the basin’s high-stress zone (criterion 4) and the ones adjacent to a recent deforestation front (criterion 3), and marks a subset near a designated protected area (criterion 1). That is a defensible sensitive-location list for the screenable criteria, ready to hand to Evaluate. The species overlap and the community-dependency question on the same sites are flagged as the specialist work that Locate has scoped but not closed. TNFD’s own piloting organizations worked this way, combining asset data with protected-area and wetland layers and then adding species and field studies on top.
Locate is the foundation the rest of LEAP stands on
Everything downstream in a TNFD assessment inherits the quality of Locate. Evaluate can only assess dependencies and impacts at the places Locate found. Assess can only price risk at those places. Prepare can only disclose what the first three produced. Get Locate right, carefully scoped and defensibly sourced, and Strategy Disclosure D writes itself from the record you have already built. Get it wrong, and every later phase is confidently assessing the wrong map.
The step is not glamorous. It is address matching, supply-chain visibility, five spatial screens, and a careful line between what the data settled and what it only flagged. But it is the step that makes the difference between a nature disclosure anchored to real places and one anchored to a country list.
Frequently asked questions
What is a priority location under TNFD?
TNFD defines priority locations as the union of two sets: sensitive locations, which meet at least one of the five ecological criteria, and material locations, where the organization has identified material nature-related dependencies, impacts, risks, or opportunities. A location can be ecologically sensitive without being material to your business, and material without being sensitive. Priority locations are the ones that are either. They are what Strategy Disclosure D asks you to report.
What is Strategy Disclosure D in the TNFD framework?
Strategy Disclosure D is the TNFD-specific priority-locations disclosure, and it has no equivalent in the older climate framework that TNFD otherwise mirrors. It asks you to disclose the locations of assets and activities in your direct operations and, where possible, your upstream and downstream value chain, that meet the criteria for priority locations. It is distinct from Strategy Disclosure C, which covers strategy resilience, not locations.
What is the difference between a sensitive location and a priority location?
A sensitive location is a place that meets at least one of TNFD’s five ecological criteria. A priority location is broader: it is any location that is either sensitive or material to the business. So a sensitive site is not automatically a priority site, and a priority site is not always ecologically sensitive. Only priority locations go into Strategy Disclosure D, so you do not disclose every sensitive site regardless of materiality.
Do you have to disclose every sensitive location under TNFD?
No. Strategy Disclosure D asks for priority locations, not every sensitive location. A site is only disclosed if it is either sensitive or material. Because materiality is settled in the next LEAP phase, Evaluate, your priority set stays provisional until then, and TNFD asks you to revisit which locations qualify after Evaluate and at each reassessment.
What are the five TNFD sensitive-location criteria?
The five criteria are: areas important for biodiversity, including species; areas of high ecosystem integrity; areas of rapid decline in ecosystem integrity; areas of high physical water risk; and areas of importance for ecosystem-service provision, including benefits to Indigenous Peoples and Local Communities. Only one of the five needs to be met for a location to count as sensitive.
