🌾 Agriculture
Burnable vegetation × FWI hazard intersected with CLC agricultural areas. Economic Vulnerability Index (EFFIS) reflects restoration-cost potential for cropland and forest.
CLIMATE HAZARD RISK ASSESSMENT
Wildfire
Asset-Dependent Threat: The WUI Intersection
A wildfire is an uncontrolled fire occurring in vegetated areas — forests, shrublands, grasslands, or Wildland–Urban Interface (WUI) zones. It is driven by the interaction of fire weather (temperature, humidity, wind speed), fuel availability (vegetation type, moisture content, load), and ignition sources (lightning or human activity).
The ClimEmpower wildfire hazard is defined as the multiplicative product of the Fire Weather Index (FWI) — a Canadian Forest Service index capturing wind, temperature, humidity, and precipitation as a composite fire danger score — and burnable vegetation availability (EFFIS fuel type classification of CORINE Land Cover). A key ClimEmpower methodological adaptation moves burnable vegetation from the Vulnerability component to Hazard, reflecting that fuel availability is a primary trigger condition rather than a secondary sensitivity factor.
More than 785,000 hectaresBurned across the EU in 2022 — one of the worst seasons on record
METHOD OVERVIEW
Asset-Dependent Threat: The WUI Intersection
The assessment combines a mapped climate hazard with the people, assets or ecosystems exposed to it and their vulnerability. Sector-specific indicators then produce the final risk result.
HHazardIntensity & extent
×
EExposurePeople & assets
×
VVulnerabilitySensitivity & capacity
=
RRiskPotential impact
Agriculture
Population & Health
Infrastructure
Biodiversity
Tourism
Layer structure, input variables and scoring logic for the wildfire risk maps.
👥 Population & Health
WUI proportion is the primary vulnerability driver. Exposure via population density and social facilities in fire-adjacent zones. Critical role of WUI density in scoring structural vulnerability.
⚙ Infrastructure
Energy plants, grids, pipelines, transport in fire zones. E-PRTR dangerous elements treated as vulnerability (secondary-disaster potential if fire reaches industrial site).
🌿 Biodiversity
Protected area fraction drives exposure. Ecological Vulnerability Index (1–10 by CLC type) measures susceptibility: broad-leaved forest = 10, mixed = 8, coniferous = 7.
🏨 Tourism
Hotels, golf courses, tourist infrastructure in WUI. No separate vulnerability layer — risk assessed via interaction of FWI hazard with exposed tourism assets only (H × E).
RISK INDEX
1–1,000
Wildfire Risk Index
Composite result of Hazard × Exposure × Vulnerability, expressed on a common 1–1,000 scale.
Limited risk. Maintain existing good practices.
Low impact. Maintain adaptive capacity.
Noticeable risk. Precautionary planning is recommended.
Significant risk. Targeted intervention is needed.
Severe risk. Immediate action is required.
RISK FORMULA
Wildfire × Agriculture
R_agr = H × E_agr × V_agrHH = FWI × BurnVeg_score → rescaled 1–10
FWIw = 0.50
Fire Weather Index (multi-model mean 2021–2025)
Copernicus CDS · ~12km → 100mBurnVegw = 0.50
Burnable vegetation score (CORINE fuel type reclassification)
EFFIS Fuel Types × CLC 2018 · 100mEE = AgrAreas_CLC
AgrCLCw = 1.00
Agriculturally used areas and forests (CLC2 agricultural classes)
CORINE Land Cover 2018 · 100mVV = EconVuln_EFFIS
EconVulnw = 1.00
Economic Vulnerability Index (EFFIS) — reflects restoration costs for agriculture and forest
EFFIS / JRC · Static 2021 · 0–11°RISK FORMULA
Wildfire × Population & Health
R_pop = H × E_pop × V_popHH = FWI × BurnVeg_score → rescaled 1–10
FWIw = 0.50
Fire Weather Index (multi-model mean 2021–2025)
Copernicus CDS · ~12km → 100mBurnVegw = 0.50
Burnable vegetation score (CORINE fuel type reclassification)
EFFIS Fuel Types × CLC 2018 · 100mEE = (PopDens + SocFacil) / 2
PopDensw = 0.50
Population density
ESTAT Census Grid 2021 · 1kmSocFacilw = 0.50
Social facilities (hospitals, schools, care homes)
EU Basic Services (Eurostat) · NUTS-3VV = WUI_proportion
WUIw = 1.00
Wildland–Urban Interface proportion — population density in WUI (scored 1–10 by inh./km²)
OSM Buildings ∩ ESTAT Census Grid 2021 · LocalRISK FORMULA
Wildfire × Infrastructure
R_infra = H × E_infra × V_infraHH = FWI × BurnVeg_score → rescaled 1–10
FWIw = 0.50
Fire Weather Index (multi-model mean 2021–2025)
Copernicus CDS · ~12km → 100mBurnVegw = 0.50
Burnable vegetation score (CORINE fuel type reclassification)
EFFIS Fuel Types × CLC 2018 · 100mEE = (Buildings + Roads + CritInfra) / 3
Buildingsw = 0.33
Building footprints
OpenStreetMap / Geofabrik · VectorRoadsw = 0.33
Road network (primary, secondary, motorways)
OpenStreetMap · VectorCritInfraw = 0.34
Critical infrastructure (energy, transport, water/waste treatment)
Open Power System Data · OSM · GeofabrikVV = DangerousEl_EPRTR
EPRTRw = 1.00
Dangerous Elements — E-PRTR industrial facilities where fire could trigger secondary disasters
E-PRTR (European Commission) · PointsRISK FORMULA
Wildfire × Biodiversity
R_bio = H × E_bio × V_bioHH = FWI × BurnVeg_score → rescaled 1–10
FWIw = 0.50
Fire Weather Index (multi-model mean 2021–2025)
Copernicus CDS · ~12km → 100mBurnVegw = 0.50
Burnable vegetation score (CORINE fuel type reclassification)
EFFIS Fuel Types × CLC 2018 · 100mEE = ProtAreaFrac
ProtAreaw = 1.00
Protected area fraction (Natura 2000 polygon fraction per grid cell)
Natura 2000 / EFFIS (EEA) · Polygon · 2021VV = EcolVuln_EFFIS
EcolVulnw = 1.00
Ecological Vulnerability Index: broad-leaved forest=10, mixed=8, coniferous=7, grassland=7, scrub=5, arable=1
CORINE Land Cover 2018 (reclassified per EFFIS) · 100mRISK FORMULA
Wildfire × Tourism
R_tour = H × E_tour [no V layer — equal ½ weights]HH = FWI × BurnVeg_score → rescaled 1–10
FWIw = 0.50
Fire Weather Index (multi-model mean 2021–2025)
Copernicus CDS · ~12km → 100mBurnVegw = 0.50
Burnable vegetation score (CORINE fuel type reclassification)
EFFIS Fuel Types × CLC 2018 · 100mEE = TourInfra_OSM
TourInfraw = 1.00
Touristic infrastructure — hotels, apartments, golf courses, ports in WUI zones
OpenStreetMap (tags: hotel, apartment, golf_course, port, airport) · Local⚠ No Vulnerability layer identified for this sector — equal weights applied to H and E only (½ each). Range 1–100, scaled ×10.
Range 1–100 (no V component) · scaled ×10 → 1–1,000
HHazard Variables2 variables+
Fire Weather Index (FWI)
Multi-model mean 2021–2025. Reclassified 1–10 per EFFIS danger classes.
DATA SOURCECopernicus CDS
RESOLUTION~12km→100m
Burnable Vegetation Score
Fuel type classification of CORINE Land Cover. Included in Hazard as a primary fire trigger condition.
DATA SOURCEEFFIS Fuel Types × CLC 2018
RESOLUTION100m
EExposure Variables6 variables+
Population Density
Population/Health pillar.
DATA SOURCEESTAT Census Grid 2021
RESOLUTION1 km
Social Facilities
Hospitals, schools, care homes.
DATA SOURCEEU Basic Services (Eurostat)
RESOLUTIONNUTS-3
Critical Infrastructure
Energy, transport, water/waste treatment.
DATA SOURCEOpen Power System Data · OSM
RESOLUTIONPoints/lines
Tourist Infrastructure (Hotels, Golf)
Tourism pillar.
DATA SOURCEOpenStreetMap
RESOLUTIONLocal
Protected Area Fraction
Ecological/Biodiversity pillar.
DATA SOURCENatura 2000 / EFFIS (EEA)
RESOLUTIONPolygon
Agricultural Areas & Forests
Agriculture pillar.
DATA SOURCECORINE Land Cover 2018
RESOLUTION100 m
VVulnerability Variables3 variables+
WUI Population Density
Scored 1–10 by inhabitants per km² in the Wildland–Urban Interface.
DATA SOURCEOSM Buildings ∩ Census Grid 2021
RESOLUTIONLocal
Ecological Vulnerability Index
Broad-leaved forest=10, mixed=8, coniferous=7.
DATA SOURCECLC 2018 (reclassified)
RESOLUTION100 m
Dangerous Elements
Industrial facilities where fire could trigger secondary disasters.
DATA SOURCEE-PRTR (European Commission)
RESOLUTIONPoints
Andalusia is one of Europe's highest wildfire-risk regions: more than 2.9 million hectares of Natura 2000 areas, cork-oak dehesas, and the Sierra Nevada and Sierra Morena pine forests. Far more accurate data is available for this region than for the other four ClimEmpower partner regions. To make the best use of available data, a more elaborated risk calculation method has been established for Andalusia.
The Andalusian method also places greater emphasis on Vulnerability over Hazard and Exposure, since this is the factor the region can influence most efficiently. This is reflected in the following risk calculation formula.
ANDALUSIAN WEIGHTING MODEL
R = H^0.2 × E^0.3 × V^0.5
H weight 0.2 · E weight 0.3 · V weight 0.5 · sum = 1.0Hw = 0.2Ew = 0.3Vw = 0.5
Vulnerability receives the highest weight (0.5) because WUI population density — the primary vulnerability driver — has more than doubled in Andalusia over the 1970–2026 period, as periurban residential development expanded into fire-prone slopes. This is directly addressable through land-use planning and building codes. Fire weather index (Hazard, 0.2) is climate-driven and non-negotiable; asset location (Exposure, 0.3) is largely constrained by existing settlement patterns.
ANDALUSIAN CLASSIFICATION TABLES — VARIABLE THRESHOLDS
HHazard layers (2)+
FWI (Fire Weather Index)
Copernicus CDS · Multi-model mean 2021–2025 → 100mAndalusia uses EFFIS 5-class danger scale anchored to observed fire ignition probability. Sierra Nevada southern slopes and Costa del Sol reach FWI >45 in peak summer — scored 10.
RANGE / CONDITIONSCORECLASS
FWI < 11.22Low danger (EFFIS 1)11.2–21.34Moderate (EFFIS 2)21.3–38.06High (EFFIS 3)38.0–50.08Very High (EFFIS 4)> 50.010Extreme (EFFIS 5)Burnable Vegetation Score
EFFIS Fuel Types × CLC 2018 · 100mAndalusia's cork-oak dehesa (score 7), Mediterranean maquis/garrigue (score 9), and pine plantations (score 8) are dominant fuel types.
RANGE / CONDITIONSCORECLASS
Water / urban / bare0Non-burnable (masked)Arable / crops2Low fuelEsparto / grassland6Medium fuelDehesa / cork oak7Medium-highMaquis / conifer9High fuel (EFFIS primary)EExposure layers (2)+
WUI Population (OSM Buildings × Census)
OSM Buildings ∩ ESTAT Census Grid 2021 · LocalAndalusia has >300 municipalities with WUI population >500 inh/km². Marbella, Mijas, Nerja are among the highest-scoring.
RANGE / CONDITIONSCORECLASS
< 50 inh/km² in WUI2Very sparse50–2004Low200–5006Moderate WUI500–1,5008High WUI density> 1,500 inh/km² in WUI10Critical WUIProtected Area Fraction
Red Natura 2000 de Andalucía · PolygonAndalusia has the largest Natura 2000 network in Spain. Regional polygons provide higher-resolution data than the pan-European EEA dataset.
RANGE / CONDITIONSCORECLASS
0–10% Natura 2000 in cell2Low protection10–30%4Partial30–60%6Significant60–85%8Mostly protected> 85% Natura 200010Core protected zoneVVulnerability layers (2)+
Ecological Vulnerability Index
CLC 2018 reclassified per EFFIS · 100mAndalusian reclassification: holm-oak = 10, cork-oak = 9, Pinus pinaster plantation = 8, macchia = 7. Non-burnable cells masked to score 0.
RANGE / CONDITIONSCORECLASS
Arable / artificial (non-burnable)1MaskedSclerophyllous shrub5Medium ecol. riskConiferous plantations (pine)8High (fast fire spread)Cork oak / holm oak9Very highBroad-leaved Mediterranean forest10Maximum ecol. lossWUI Population Density (V layer)
OSM Buildings ∩ Census Grid 2021 · LocalIn the Andalusian model, WUI density also appears as Vulnerability — measuring how difficult evacuation and containment are in dense WUI zones.
RANGE / CONDITIONSCORECLASS
< 100 inh/km² in WUI2Evacuable100–5004Manageable500–1,0006Challenging1,000–3,0008Complex> 3,000 inh/km²10Critical evacuationRegional wildfire emergency plan; historical fire perimeter database 2000–2024; fuel management priorities
Higher-resolution Natura 2000 polygons than EEA pan-EU dataset (1:25,000 vs 1:100,000)
Historical fire occurrence and burn severity (dNBR) database for Andalusia
Industrial pollutant emission register — Spanish supplemented version with additional facility classifications
Firebreaks & Mechanical Fuel Management
Infrastructure➜ Reduces fire intensity, slows spread; critical in Sierra Nevada / Sierra Morena
Regions: Troodos, Sicily, AndalusiaBuilding Codes & WUI Land-Use Planning
Infrastructure➜ Reduces structural damage and casualties in Wildland–Urban Interface zones
Regions: SicilyWater Resource & Firefighting Infrastructure
Infrastructure➜ Enhances firefighting effectiveness; limits spread to critical assets
Regions: SicilyFire-Resistant Building Materials (WUI zones)
Infrastructure➜ Reduces structure ignition in areas with high WUI population density score
Regions: OBZReforestation & Forest Conservation
Environment & Nature➜ Reduces fuel load over time; increases long-term landscape resilience
Regions: OBZ, AndalusiaPrescribed Burns & Controlled Fires
Environment & Nature➜ Controlled reduction of accumulated fuel load; lowers EVI for target areas
Regions: Andalusia (Plan INFOCA)Adaptive Ecosystem Management
Governance & Planning➜ Improved preparedness through landscape-level fire risk integration
Regions: OBZ, SicilyEarly Warning Systems & Fire Forecasting
Governance & Planning➜ Enables timely evacuation and resource pre-positioning; reduces fatalities
Regions: SicilyEmergency Preparedness & Response Plans
Governance & Planning➜ Reduces casualties; improves inter-agency coordination during wildfire events
Regions: SicilyInclusive Conservation & Community Engagement
Society & Education➜ Local ecological knowledge supplements satellite remote sensing for hotspot identification
Regions: OBZ, SicilyPublic Awareness & Fire Risk Education
Society & Education➜ Reduces human-caused ignitions; improves evacuation compliance
Regions: Sicily, Troodos, AndalusiaHow measures reduce risk
Each measure targets one or more risk components. Governance and planning measures typically reduce Exposure (E) by keeping assets out of hazard zones. Nature-based and infrastructure measures primarily reduce Vulnerability (V). Early warning and awareness measures improve adaptive capacity — the fastest path to a lower risk score.