How Scoring Works

Every location on Criora gets an overall risk grade from A++ (lowest risk) to F (highest risk). This page explains how that grade is built up from raw environmental data.

Three layers of scoring

Raw data  →  Risk types  →  Dimensions  →  Overall grade
(sources)    (15 site,       (5 groups)     (A++ … F)
              3 country)

Each layer answers a different question:

Layer	Answer
Risk type	How exposed is this location to one specific climate hazard?
Dimension	What is the worst risk within an environmental theme?
Overall grade	What is the combined risk picture for this location?

Score scale

All scores use a 0-100 scale where higher means more risk.

Score	Level	Letter band
0-20	Very Low	A++, A+, A, A−
20-40	Low	B+, B, B−
40-60	Moderate	C+, C, C−
60-80	High	D+, D, D−
80-100	Critical	F

The + / − modifiers position a location within its band: B+ is at the lower (better) end of the 20-40 range, B− is at the upper end. A++ is reserved for locations scoring below 5.

Risk types

Criora evaluates each location against 15 site-level risk types, grouped into two environmental dimensions, plus 3 country-level dimensions sourced from country indices.

Site-level risks (from geospatial data)

Climate dimension: 12 risk types:

Extreme Temperature Events
Chronic Temperature Stress
Snow and Ice Hazards
Wind Pattern Changes
Storm and Lightning Events
Precipitation Extremes
Water Stress
Drought
Flooding Events
Coastal and Marine Risks
Sea Level Rise
Health and Workforce Impacts

Nature dimension: 3 risk types:

Wildfires
Erosion and Degradation
Land Movement (subsidence, landslide)

See the risks overview for definitions and the data sources behind each one.

Country-level dimensions (from country indices)

Social: workforce, health, and demographic vulnerability
Governance: institutional quality and regulatory environment
Adaptation: readiness and capacity to respond to climate change

These come from INFORM Risk Index, ND-GAIN, and World Bank ESG indicators.

Dimensions

The 15 site risks roll up into 2 site-level dimensions, joined by 3 country-level dimensions for a total of 5:

Dimension	Source	Covers
Climate	Site	Temperature, precipitation, wind, water
Nature	Site	Wildfires, erosion, ground movement
Social	Country	Workforce and health vulnerability
Governance	Country	Institutional and regulatory environment
Adaptation	Country	Climate-readiness and resilience

Each site dimension takes the maximum score among its risks. This is intentional: a single critical risk should not be hidden by averaging it with lower ones.

ℹ️ Why max instead of average?

Climate risk is driven by tail events. A location with extreme flood exposure and low wind risk is genuinely high-risk, not “moderate on average”. Insurance and IPCC frameworks use the same approach.

Site + country blend

For dimensions that exist on both sides (e.g. climate exposure has both site geography and country adaptation capacity), the final dimension score is a weighted blend:

dimension_score = 0.6 × site_score + 0.4 × country_score

If only site or only country data is available, the score uses what’s there.

Overall grade

The five dimension scores combine into a single overall grade by averaging them:

overall_score = average of (Climate, Nature, Social, Governance, Adaptation)

That score maps to a letter grade with + / − modifier, and is surfaced alongside the driver dimension: the one pulling the grade down hardest.

ℹ️ Why a plain average?

Each dimension is shown separately, so the overall grade is meant as a navigational summary, not a substitute for the per-dimension view. A plain average keeps the math transparent and matches the approach taken by established composite indices such as the UN Human Development Index and ND-GAIN.

One consequence worth knowing: a location reaches the F band only when every dimension is itself in the critical range. A single critical dimension pulls the overall grade down, but does not push the location to F on its own.

💡 Worked example

A site with dimensions Climate 78, Nature 45, Social 60, Governance 55, Adaptation 67 has an average of 61 → grade D+, driver Climate.

Confidence

Each score is reported with a confidence indicator based on:

Data coverage: how many input layers were available for this location
Data recency: how current the inputs are
Spatial resolution: how fine-grained the source data is at this point

Locations in well-monitored regions (Europe, North America) typically have higher confidence than remote areas.

Working with Layers: explore the underlying data on the map
Risks Overview: definitions for each risk type
Data Sources: provenance of every input
Climate Context: how Köppen climate zones shape relevance