Canopy scores African countries and projects for people making decisions about where to put conservation money. The country view answers one question: how ready is this country for a conservation deal. C-RISK scores that readiness from 1 to 5 across five weighted slices, and publishes it as a country score on a 0-100 scale. They are one model on two scales, not two models.

The five slices are: fiscal capacity (can the sovereign carry or restructure the instrument), deal environment (can capital move, and is there official-sector cover), conservation asset (are the parks actually managed, measured by PACE), delivery capacity (can anyone execute), and traction (is anything actually closing). A slice with no data abstains and is left out of the mean rather than scored zero.

Two gates can cap the result regardless of the number: active armed conflict and sovereign distress. Governance is still scored and shown, but as context at zero weight. A separate model, C-CARBON, rates individual carbon projects and reads C-RISK for the project's host country. The diagram below shows how everything connects.

DATA FEEDS Sovereign data Ratings, spreads, IMF Policy and donors Flow, market policy, donors PACE scores Protected-area performance Operator estate Managed-estate scale Deals and gap Nature-finance precedent THE FIVE SLICES Fiscal capacity Can the sovereign carry it Sovereign Fiscal Lens 54 countries 0.30 Deal environment Can capital move Capital flow, policy, donor environment 0.20 Conservation asset Are the parks managed PACE performance 199 protected areas 0.20 Delivery capacity Can anyone execute Managed-estate scale, operator depth 0.15 Traction Is anything closing Swap precedent, pipeline, financing gap met 0.15 READINESS C-RISK Readiness score, 1 to 5 Weighted mean of the populated slices × 20 COUNTRY SCORE Country score C-RISK rescaled 20-100 All 54 African countries Two gates cap the tier Active armed conflict: at best T3 Sovereign distress: at best T2 Never the score, never the rank order CLASSIFICATION Country tier: T1 T2 T3 T4 Governance is scored and shown alongside the score, at zero weight in it. DOWNSTREAM C-CARBON Africa-specific carbon project ratings country risk
Click any box to open its methodology page.

The formula

readiness = weighted mean of ( fiscal capacity 0.30 + deal environment 0.20 + conservation asset 0.20 + delivery capacity 0.15 + traction 0.15 ) 1–5
country score = readiness × 20 20–100
Each slice is scored 1 to 5. The mean is renormalized over the slices that carry data, so a slice with no data abstains rather than dragging the score to zero. Because the slices floor at 1, the score cannot fall below 20: the attainable range is 20 to 100, not 0 to 100. Tiers: T1 ≥ 80, T2 60-79, T3 40-59, T4 < 40, four equal bands of 20 covering that range. These are the four readiness calls (4.0 / 3.0 / 2.0) carried onto the 0-100 scale, so a country's tier and its call cannot disagree; the edges moved to meet them in v2.1 and v2.2. On the current cohort T1 is empty. Two gates can then cap the tier: active armed conflict at T3, sovereign distress at T2. A gate only ever lowers a tier. It never changes the score or the rank order.

The five slices

Fiscal capacity

0.30
Can the sovereign carry or restructure the instrument

Read from the Sovereign Fiscal Lens: credit rating, sovereign bond spread, multilateral backstop status, and fundamentals. This slice leads on purpose. For a debt-for-nature swap the binding constraint is almost always whether the sovereign can carry the paper. All 54 African countries.

Deal environment

0.20
Can capital move, and is there official-sector cover

Capital-flow freedom, carbon and market policy, and the donor environment. This is the plumbing question: whether money can enter, be converted, be deployed against a conservation instrument, and leave.

Conservation asset

0.20
Are the parks actually managed

Keystone protected-area performance from PACE, aggregated to the country. PACE scores 192 major African protected areas across nine dimensions covering biodiversity, ecosystem services, security, ecotourism infrastructure and revenues, community development, leadership, budget, and fundraising. The slice abstains where a country has no cohort protected area.

Delivery capacity

0.15
Can anyone execute

The execution capacity of the conservation operators working in the country's protected areas, led by the managed-estate scale of the strongest operator present, whether that is a state agency, an NGO, or a trust, and widened by how many operators are present.

Traction

0.15
Is anything actually closing

Nature-finance deal precedent, ongoing pipeline, and how much of the conservation financing gap is being met. Forward-looking slices say a country should be able to transact; traction is the record of whether it has.

Two gates sit outside the weighted mean and can only ever lower the result. Active armed conflict caps the country at Tier 3, and sovereign distress, meaning a fiscal score at or below 2.0 or a debt-distress classification, caps it at Tier 2. A legal-standing gate for active degazettement is defined but abstains, because there is no country-level protected-area-downgrade data across the cohort. Governance is still scored and published, but as context at zero weight rather than as an input.

Downstream

C-CARBON - Carbon project ratings

C-CARBON rates individual African carbon projects across five dimensions: methodology, additionality, financial viability, sovereign risk, and community architecture. Designed as a complement to the global project raters, not a replacement. As of v1.1 (May 2026), C-CARBON's sovereign risk dimension reads directly from C-RISK for the project's host country.

C-CARBON methodology →

Evaluation

Canopy Impact - did protection make a difference?

Every model above measures a state or a change. Impact asks the separate question: is the boundary working? For each protected area it compares the change in human pressure (cropland and built-up ground) inside the boundary against the change over the same years in a 10km ring just outside it, with neighbouring protected areas cut out of that ring so the comparison runs against unprotected land. A park where pressure grew more slowly inside than next door is holding; one where it grew faster inside is not.

This is an evaluation result, not a score. It never enters PACE, C-RISK, C-CARBON or the country score, and it abstains rather than zeroing: parks with an approximated boundary, a ring swallowed by neighbouring protected areas, or a land-cover reading that collapses implausibly are held out and listed with the reason, not published as a result.

Canopy Impact →

Validation

C-VAL - the validation matrix

Every model on this page is tested the same way, across six families: reliability, internal structure, robustness, stress, external benchmarking and bias, plus an uncertainty propagation test for the country score. The full method and results for each test live in one place rather than scattered through the methodology pages. Each cell links to its own detail page with the numbers and a chart.

C-VAL validation matrix →

What this is and isn't

Disclosures

Weights are qualitative. The relative weighting across the five slices reflects practitioner judgment on what matters most for long-horizon conservation capital. A reader who disagrees can recompute any score from the published slice scores. Whether the weights hold up under perturbation, and how the models perform on reliability, internal structure, robustness, stress and external benchmarks, is reported in the C-VAL validation matrix.

This is a restatement of method, not a change on the ground. Until version 2.0, the country score was a weighted blend of four components: C-RISK, PACE, an operator-ecosystem count, and a deployed-project count. Those last two are no longer part of the composite. Operator presence now enters through delivery capacity, and the project count is kept as a display sub-signal on the country pages rather than as an input. The v1 series is preserved in the monthly snapshots. The C-RISK country score methodology sets out the change in full.

Editorial review architecture is currently unstaffed. C-RISK and PACE both have full editorial review pipelines built. Neither is staffed: drafts are promoted to live data after automated parse validation, with no human sign-off in between.

Structural data gaps. Some PACE dimensions (ecotourism revenue, budget) are missing for 50%+ of 192 PAs; these are disclosed in the PACE methodology page rather than papered over with imputation.