Earth LinC  /  Evidence-led feature Learning review  ·  Updated 22 September 2026

Nature-based Solutions  —  ten weeks with the public record

From Planting
to Proof

What makes a Nature-based Solution credible, investable and repeatable? Not how green it looks. Whether its mechanism, money, delivery and evidence can survive being checked by someone who did not build it.

A watershed cross-section with an evidence line running above it Upland forest on the left drains through a river channel into a restored wetland, which sits beside a small community on the right. Above the landscape, a dashed line connects five nodes labelled Design, Fund, Deliver, Measure and Verify. Dotted arrows show water infiltrating toward a groundwater line below. DESIGN FUND DELIVER MEASURE VERIFY UPLAND WETLAND COMMUNITY GROUNDWATER
Cover diagram, original. Read it left to right for water, top to bottom for the claim: every part of the landscape below the dashed line has to be designed, funded, delivered, measured and verified above it.
The system chainEight links

A project is not a list of activities. It is a chain of claims.

Ten weeks of case study kept returning to the same shape. A credible Nature-based Solution — an NbS, meaning an intervention that works through an ecosystem process rather than merely alongside one — is a managed evidence chain. Break any single link and the public claim quietly stops being true, even while the trees keep growing.

  1. 01RiskWhich societal risk is reduced, for whom, and at what frequency of event?
  2. 02NatureWhich ecological mechanism does the physical work — infiltration, roughness, storage, shade, sediment capture?
  3. 03DesignAn intervention sized to the mechanism, not to the available budget line.
  4. 04FinanceWho pays, on which instrument, against which risk, over how many years?
  5. 05DeliveryProcurement, contracts and a clear RACI — who is Responsible, Accountable, Consulted and Informed.
  6. 06DataIndicators, sampling design, and QA/QC — the quality assurance and quality control that make a field record defensible.
  7. 07VerificationAn identifiable, independent process tests the claim before anyone is paid against it.
  8. 08LearningWhat actually changes in the next design because of what was measured here.
Assumption 01

Planting trees or making green space is automatically an NbS. Without a named risk and a named mechanism, it is landscaping.

Assumption 02

Hectares restored and people trained prove impact. They prove effort. Impact is a change in what the system does.

Assumption 03

Positive economic value means the project is bankable. Value spread across a floodplain is not cash arriving in an account.

The caseRecharge Pakistan
Used as a documented learning case only

Five facts we deliberately did not fill in

Recharge Pakistan — an ecosystem-based approach to flood and water security financed in part through the Green Climate Fund (GCF), the UN climate finance mechanism — was the anchor case for this study. We read its public documents; we did not implement it, visit it or independently evaluate it.

Which makes the next part the honest part. Every number below exists in the public record, and every number below can change between an approved proposal, a signed agreement and the latest official report. Publishing a figure you have not re-opened on the day you publish is the exact failure this article is about. So the register stays open, with the primary source one click away.

We left the numbers blank on purpose.
Approximate total project value — to verify Gap GCF FP207 page
GCF financing amount and instrument — to verify Gap Funding proposal
Expected direct beneficiaries — to verify Target Funding proposal
Degraded landscape area targeted — to verify Target GCF FP207 page
Preparation and approval timeline — to verify Official reported GCF FP207 page

Source hierarchy when documents disagree: signed legal agreement → approved proposal → latest official report → official project page → secondary analysis. Record the document date and your access date beside every figure you copy out.

Status of a numberSix labels

Every number should say how it knows

A statistic without a status is a rumour with a decimal point. Six labels are enough to keep an entire portfolio honest, and they cost nothing to apply. If a figure is derived, show the formula, the inputs, the units and the assumption that carried it.

  • Official reportedPublished by the implementing or funding institution about its own project.
  • Independently verifiedChecked by an identifiable third party through a named process.
  • Project targetAn intention in an approved document. Not an achievement.
  • ModelledAn estimate from a model with stated inputs, not a measurement.
  • DerivedCalculated here from other figures. Formula shown, or it does not count.
  • GapUnavailable, conflicting or out of date. Mark it; never interpolate silently.
The vocabularySeven words that decide a claim

Activity, output, outcome, impact

Most disputes about whether a project "worked" are really disputes about which of these words was used. Each term on first use, plainly:

Activity
What the project does. Excavate a channel, plant riparian vegetation, run a training week.
Output
What exists once the activity stops. A restored wetland. Twelve trained operators.
Outcome
A change in how the system behaves, measured against a defined condition.
Impact
The human consequence of that change: loss avoided, water secured, health protected.
Assumption
What must stay true — upstream land use, rainfall regime, maintenance budget — for the chain to hold.
Attribution
How you know the change came from the intervention and not from a dry year or a new embankment next door.
Verification
A named, independent process that tests the claim. Absent that, never write the word "verified".
One sentence, three different claims
  • Output"Wetland restored."
  • Outcome"Peak flow reduced under a defined design event."
  • Impact"Avoided loss attributable to the intervention."
Six decisionsWhere projects are made or broken

Six decisions, each with a public reference behind it

None of these are late-stage questions. Every one of them is cheaper to answer in the first month than in the fifth year.

Integrity

Is this genuinely an NbS?

Name the societal challenge and the ecosystem process doing the work. If you cannot name both, you have a landscaping budget.

"What breaks if the ecosystem is removed from this design?"

IUCN Global Standard →
Causal design

Why should it work?

Draw the mechanism — storage, roughness, infiltration, shading — before selecting a single activity. The drawing is the hypothesis.

"Which physical process converts this activity into that outcome?"

World Bank NbS handbook →
Baseline

Compared with what?

A counterfactual is the world without the project. Without one, improvement and good weather look identical in the data.

"What would have happened here anyway?"

GCF results framework →
Feasibility & safeguards

Can it be delivered fairly?

Land tenure, consent, displacement of water or risk downstream, and who carries the inconvenience during construction.

"Who loses something for this to work, and were they asked?"

Benefits and costs of NbS →
Funding & contracts

Who pays and who maintains?

Instruments, risk allocation, payment triggers, and a maintenance obligation that survives the closing ceremony.

"Whose budget line pays for year seven?"

Procurement framework →
Evidence

What can honestly be claimed?

Match the strength of the language to the strength of the method. Modelled is not measured; reported is not verified.

"Would this sentence survive an auditor reading the raw data?"

State of Finance for Nature →
Explainer ABlended finance

Why the cheapest money sits at the bottom

Blended finance stacks capital by appetite for risk. The layer at the base is not the biggest — it is the bravest. It absorbs the first losses and pays for the evidence that lets the layers above it exist at all.

Blended finance capital stack Three stacked isometric blocks. The widest block at the base is grant and technical assistance capital, the middle block is concessional capital, and the smallest block on top is commercial or results-based capital. COMMERCIAL CONCESSIONAL GRANT & TA
Read upward: each layer will only enter once the layer below has made the risk legible.
  1. Top — commercial or results-based

    Wants contracted cash flow or a verified result to pay against. It does not buy potential.

  2. Middle — concessional

    Below-market, patient, willing to wait for an ecosystem to behave like an asset.

  3. Base — grant and technical assistance

    Pays for feasibility, safeguards, baselines and monitoring: the work that turns a landscape into an investable proposition. Cut it, and nothing above it can price the risk.

Explainer BEvidence to payment

The distance between a field note and a payment

In payment-for-ecosystem-services schemes (PES) and results-based finance, money moves only when a claim clears the whole run. Each step below is a place a claim can legitimately fail — which is precisely what makes the surviving claims worth something.

  1. Step 01Field observationRecorded by a named person, at a known time and place.
  2. Step 02QA/QCChecks, duplicates, calibration, corrections logged not overwritten.
  3. Step 03AnalysisMethod stated in advance, applied the same way each cycle.
  4. Step 04Independent verificationA third party with no stake in the answer.
  5. Step 05Acceptance or issuanceThe result is formally accepted, or a credit is issued.
  6. Step 06Payment & benefit sharingMoney reaches the people whose land and labour produced the outcome.

What to conclude: verification is not paperwork bolted on at the end. It is the load-bearing step that lets a cash payment be attached to an ecological change at all.

FindingsWhat ten weeks of case study revealed

Six things that were not obvious in week one

Integrity before branding

The label "nature-based" is applied far faster than the standard behind it. Screening against a published standard early is cheaper than defending the term later.

Ask: which societal risk does this reduce, for whom, and at what return period?

IUCN Global Standard →

Outputs are not outcomes

Hectares, seedlings and training days measure effort. They are necessary for delivery management and insufficient for an impact claim.

Ask: what would we measure to show the system behaved differently?

GCF results framework →

Boundaries have different jobs

The catchment explains the mechanism. The administrative boundary controls consent and budget. The beneficiary footprint decides who can claim the benefit. Three maps, three purposes, routinely confused.

Ask: which boundary was this map drawn for, and who is outside it?

World Bank NbS handbook →

Hybrid beats either/or

The grey-versus-green argument is mostly a false choice. A structure can guarantee a threshold; a wetland can absorb variability and keep improving with age. Together they often outperform either alone.

Ask: what does the engineered component do that the ecosystem cannot, and vice versa?

Benefits and costs of NbS →

Economic value is not cash flow

A benefit-cost ratio above one says society gains. It does not say a lender gets repaid. Bankability needs an identified payer, a trigger and a contract.

Ask: who pays whom, on what trigger, in which year?

State of Finance for Nature 2023 →

Maintenance is the investment

Operations and maintenance (O&M), monitoring and post-event recovery are not overheads on the project. For a living asset they are the mechanism by which it keeps performing.

Ask: who holds the maintenance budget in year seven, and under which contract?

Procurement framework →

Field practiceStrategy meets site

The minimum chain a field record has to survive

Strategy fails at the last hundred metres, on a riverbank, in bad light, with a phone at 12 per cent. The bridge between an investable design and a defensible claim is an ordinary discipline repeated every visit. A checklist or a local tool can help a person follow it — but no tool validates or verifies a project. People and processes do that.

  1. 01Safety firstNo record is worth an incident.
  2. 02GPS and timeWhere and when, before anything else.
  3. 03Repeatable photoSame point, same bearing, every visit.
  4. 04Measurement methodNamed, versioned, unchanged mid-season.
  5. 05Sample IDUnique, traceable, readable when wet.
  6. 06QA/QCDuplicates, calibration, range checks.
  7. 07Corrective actionLogged with a reason, never erased.
  8. 08Approved recordSigned off by someone accountable.
ActionNext project

Five things a young practitioner can do on the next project

  1. Ask what societal risk is being reduced.In one sentence, with a population and a frequency in it.
  2. Draw the ecological mechanism before choosing activities.One diagram, on paper. If it cannot be drawn, it cannot be defended.
  3. Separate outputs from outcomes in the log frame.Two columns, permanently. Never let one migrate into the other.
  4. Record assumptions and evidence status beside every number.Official, target, measured, modelled, derived or gap. Six words, applied every time.
  5. Budget maintenance and monitoring before claiming success.A project that cannot fund year seven has not finished designing year one.
TransparencyWhere this article is uncertain

What this article is, and is not

Uncertainty should be as visible as success, so here it is at full size rather than in a footnote.

  • This is a learning synthesis built from public documents. The author has not implemented, visited or independently evaluated Recharge Pakistan.
  • Project targets are not achieved results, and nothing here converts one into the other.
  • Several NbS benefits remain modelled until they are monitored and verified in the field.
  • Financing figures, instruments and institutional procedures change between proposal, agreement and reporting.
  • Source versions and access dates must be re-checked before any formal reuse of the figures referenced here.
  • No endorsement by the Green Climate Fund, IUCN, the World Bank, UNEP or any other institution is claimed or implied.
  • Any capability or progress percentage from the underlying course material is an editorial learning illustration — never project performance data.
摘要 · Abstract in Simplified Chinese

什么样的自然解决方案(NbS)值得社会出资、信任并复制?答案不在于项目看上去有多绿,而在于它的生态机理、社会价值、融资结构、证据链与长期运维能否经得起外部审视。本文以公开文件为基础,把一个项目理解为一条可管理的证据链:风险 → 生态机理 → 设计 → 融资 → 交付 → 数据 → 核查 → 学习。文中每一个数字都标注证据等级,未经复核的关键数据被刻意留白,并直接链接到官方来源。本文为案例学习综述,不代表任何机构立场。

ReferencesPrimary and official sources

Open these before you quote this

Source list assembled 22 September 2026. Open each link and record your own access date before reuse.

  1. IUCN Global Standard for Nature-based SolutionsSupports the integrity test: what qualifies as an NbS and how a design is screened against criteria.
  2. Green Climate Fund — FP207: Recharge PakistanOfficial project page. Supports project value, financing, area and timeline figures left open in the register above.
  3. Recharge Pakistan funding proposalApproved proposal. Supports targets, beneficiary estimates and the ecosystem-based design logic.
  4. GCF Integrated Results Management FrameworkSupports the output / outcome / impact distinction and indicator and verification expectations.
  5. World Bank — Nature-Based Solutions Technical Handbook (PDF)Supports causal design, hybrid grey-green systems and spatial boundary selection.
  6. World Bank — Assessing the Benefits and Costs of NbS for Climate Resilience (PDF)Supports the distinction between economic value and project cash flow.
  7. World Bank Procurement FrameworkSupports contracting, risk allocation and maintenance obligation points.
  8. UNEP — State of Finance for Nature 2023Supports the financing-gap context and the blended finance discussion.
One transferable idea

Nature is not the absence of engineering. In a credible NbS project, nature, engineering, finance, governance and evidence are designed as one system.