Model Catalog / climate

CMIP6 Core Ensemble

Climate Current active

Multi-model climate ensemble backbone for scenario-conditioned physical risk.

Horizon 2025–2100
Geography Global (50–100 km grid)
Resolution Grid-cell physical hazard; sector via asset overlay
Projection years 2030, 2040, 2050, 2060, 2070, 2080, 2100
0.69
hazard
0.57
transition
0.48
resilience
0.77
confidence
Temperature anomaly Precipitation extremes Sea-level rise Tropical cyclone intensity Carbon budget trajectories SSP scenario pathways Implied carbon price
Observed state vs. projected ensemble — the near-term / long-term complementarity

ERA5 is what has already happened — the observational record of the climate system grounded in 80+ years of station data and satellite reanalysis. CMIP6 is what physics-based models project will happen under different emissions pathways. For 1–5 year physical risk, ERA5 is superior: its observational grounding makes it more skilful than any CMIP6 model. For 2040–2100 long-run risk, CMIP6 is the only credible source: ERA5 has no capability to project structural climate change beyond its reanalysis window. The two models are not competitors — they are the near-term and long-term anchors of CE's physical risk framework.

ERA5-Calibrated CMIP6 Core Ensemble
Temporal scope Historical 1940–2025 (observational); 2025–2030 near-term extrapolation 2025–2100 (projected); strongest for 2035–2100 long-run horizon
Physical basis Observation-grounded reanalysis — no structural model uncertainty 40+ physics-based global climate models — explicit deep uncertainty quantification
Scenario coverage Historical baseline only; trend extrapolation is not scenario-conditioned Full SSP1-1.9 through SSP5-8.5 — scenario-conditioned projections for regulatory disclosure
Near-term skill High near-term skill — observationally grounded Lower near-term skill — model drift and internal variability dominate on 1–5yr horizon
Regulatory alignment ECMWF / Copernicus provenance — not directly NGFS/TCFD aligned IPCC AR6 and NGFS Phase IV backbone — direct regulatory legitimacy
Use ERA5-calibrated for near-term operational risk, company disruption calibration, and observational loss event attribution. Use CMIP6 for long-run physical risk trajectory, regulatory scenario alignment (NGFS, TCFD), and uncertainty quantification across SSP pathways. For facility-level precision at any horizon, layer GFDL process credibility on top of CMIP6 ensemble outputs for the specific hazard types where GFDL excels.

Methodology

CMIP6 (Coupled Model Intercomparison Project Phase 6) is the international standard for long-run climate scenario analysis, harmonising outputs from 40+ global climate models under shared socioeconomic pathways (SSP1-1.9 through SSP5-8.5). CE uses the CMIP6 ensemble to construct probability distributions over physical risk parameters — temperature anomaly, precipitation extremes, sea-level rise, and tropical cyclone intensity — for each sector's asset and operational exposure profile. Transition pressure signals are grounded in the implied carbon price trajectory required to achieve each SSP scenario's emissions pathway. Company-level Scope 1+2+3 emissions are mapped to SSP scenarios to determine the gap between current trajectories and model-consistent pathways.

Key Mechanisms

  1. Multi-model ensemble: 40+ global climate models are pooled to produce probability distributions rather than deterministic projections
  2. SSP scenario mapping: each climate pathway (orderly, delayed) corresponds to an SSP scenario that determines the magnitude of physical and transition signals
  3. Implied carbon price trajectory: the carbon price required to achieve each SSP pathway becomes the transition pressure calibration input
  4. Sector asset exposure: company-level physical asset locations are mapped to CMIP6 hazard projections to compute sector-specific hazard scores
  5. Emissions-to-trajectory gap: company Scope 1+3 trajectories are compared to SSP-consistent sector pathways to derive transition pressure adjustment
  6. Carbon cycle feedbacks: land and ocean carbon sinks weaken as warming increases, creating a self-reinforcing emissions-concentration loop — CMIP6 explicitly models this, allowing overshoot risk to be quantified
  7. Regional pattern amplification: Arctic amplification drives mid-latitude weather regime shifts (atmospheric blocking, jet stream displacement) that create disproportionate economic impacts in temperate regions
  8. Tipping point probability: CMIP6 ensemble spread is used to estimate crossing probabilities for climate system tipping points (AMOC weakening, permafrost melt, ice sheet instability) — the fat-tail physical risk input

Score & Confidence Methodology

Hazard scores (0–1) are calibrated against IPCC AR6 WG2 Table 16.SM.1 industry-sector exposure bands. Transition pressure scores use NGFS Phase IV scenario families. Confidence intervals are asymmetric where IPCC likelihood language (likely/very likely) maps to the p17–p83 and p5–p95 ranges. Scores are not actuarially certified — see Known Limitations.

Known Failure Modes

  • Coarse spatial resolution (50-100km grid) requires downscaling for facility-level physical risk assessment
  • Short-term (1-5 year) physical risk signals are less credible than ERA5-calibrated near-term observational anchors
  • SSP scenarios assume smooth policy implementation — the transition pressure signal underestimates delayed-action shock risk
  • Ensemble spread at regional and local scales is very high — for facility-level physical risk, GFDL process credibility or ERA5 observational calibration is more appropriate than ensemble mean
  • Carbon cycle and tipping point mechanisms are expressed as probability ranges, not deterministic outputs — probabilistic interpretation and expert elicitation required for investment-grade use

Best For

long-run scenario diversity and physical risk framing

Strengths

  • Internationally standardised — CMIP6 outputs are the basis for all IPCC AR6 WG2 physical risk assessments and NGFS Phase 4 scenarios
  • Multi-model ensemble captures deep uncertainty: the spread of model outcomes is explicit rather than hidden in a single deterministic projection
  • Long-run horizon (2100) provides the full physical risk trajectory needed for infrastructure and real estate investment decisions
  • Largest ensemble of any climate modelling framework (40+ models) — enables robust uncertainty quantification and explicit representation of deep uncertainty in physical risk projections
  • Directly underpins IPCC AR6 WG2 physical risk assessments and NGFS Phase IV scenario calibration — highest institutional legitimacy for regulatory disclosure and client communication
  • Explicit carbon cycle feedbacks allow endogenous warming overshoot risk — overshoot probability and carbon budget exceedance are computed from model physics, not assumed externally

Maturity & Validation

Model era: Current • Status: active
Core models are internally cross-validated against institutional benchmarks. Advanced modules (DSGE, Monte Carlo, Catastrophe, Commodity) are prototype-grade — not yet independently peer-reviewed. View the full validation record at Validation Registry and current capability status at Capability Registry (JSON).

Scenario Coverage

IPCC SSP1-1.9 — Net Zero ~1.5°C by 2100 (NGFS NZ2050 aligned) IPCC SSP1-2.6 — Below 2°C; ambitious but not net zero by 2050 IPCC SSP2-4.5 — Intermediate; current-policy trajectory with some mitigation IPCC SSP3-7.0 — Fragmented / delayed action; NGFS Delayed Transition aligned IPCC SSP5-8.5 — Fossil-fuel intensive; tail risk / extreme forcing scenario Policy transmission scenarios — use IMF WEO or NiGEM Near-term (1–5yr) physical risk — use ERA5-calibrated for superior near-term accuracy Compound hazard cascade scenarios — use CE Physical Hazard Cascade Model

All SSP scenarios are CMIP6 standard — directly map to IPCC AR6 and NGFS Phase IV scenario specifications.

Calibration Benchmarks

IPCC AR6 WG1 SPM Table (2021) Global temperature ranges by SSP pathway — primary structural calibration of warming trajectory
IPCC AR6 WG2 Chapter assessments (2022) Physical risk sector assessments and hazard exposure mapping by region
NGFS Phase IV Scenario Specifications (2024) Alignment of CMIP6 SSP outputs to NGFS scenario family labels and regulatory disclosure requirements
PCMDI CMIP6 Model Evaluation Metrics Model performance ranking and ensemble quality filtering — basis for model weighting decisions
WCRP CMIP Archive (ESGF) Primary data source for all CMIP6 model outputs; provenance and reproducibility reference
Industry Signal Dashboard — projected signals from this model across all tracked industries
Physical Hazard Pressure by Industry
Physical hazard index (0–1) indicating asset and operational exposure to climate-related physical risks.
Transition Pressure by Industry
Regulatory and market pressure from the low-carbon transition — 0 (low) to 1 (high).
Adaptive Resilience by Industry
Resilience index (0–1) — the industry's estimated capacity to adapt to physical and transition risk.
Industry Context
Energy
CMIP6 provides the physical hazard calibration for energy infrastructure — temperature stress on thermal efficiency, water availability for cooling, and extreme weather disruption to transmission networks. Aramco's Gulf operations face SSP2-4.5 wet bulb temperature thresholds by 2040. The transition pressure signal reflects the implied carbon price trajectory that would strand fossil assets — the key risk for ExxonMobil, Shell, and BP under SSP1-2.6.
Agriculture
CMIP6 is the primary climate model for agricultural yield risk — it directly models precipitation variability, soil moisture deficit, heat stress days, and growing season shifts for major crop-producing regions. The ensemble mean projects a 2–6% yield decline per degree of warming for major cereals, with high spatial variance. JBS, Cargill, and Bunge's South American growing regions are among the highest-exposure areas in the CMIP6 agricultural hazard map.
Manufacturing
CMIP6 physical risk for manufacturing centres on industrial facility exposure: heat-related productivity losses, water stress for process cooling (critical for BASF's Verbund system), and flooding of supply chain infrastructure. The CMIP6 ensemble captures fat-tail distributions of compound extreme events that create the highest insurance and operational cost scenarios for ArcelorMittal's coastal steel facilities and Rio Tinto's Pilbara operations.
Transport
CMIP6 provides the physical disruption risk for transport infrastructure — sea-level rise exposure of coastal ports (Maersk's global terminal network), extreme precipitation damage to road and rail (Union Pacific), heat deformation of infrastructure, and cyclone intensity increases for shipping routes. The IMO uses CMIP6 projections as the basis for its climate vulnerability assessment of shipping routes.
Insurance
CMIP6 is the foundation of catastrophe model calibration for insurance pricing. Allianz, Munich Re, and Swiss Re all use CMIP6 ensemble outputs to stress-test their nat-cat exposure books against future climate states. CMIP6 projects increasing loss volatility — higher variance around the mean loss year as tail events become more frequent — which directly compresses reinsurance capacity and drives premium inflation.
Real Estate
CMIP6 physical risk for real estate is the most direct of all sectors: flood inundation maps, coastal erosion projections, wildfire expansion, and urban heat island intensification all derive from CMIP6 ensemble projections. The model projects that 20–25% of current coastal real estate globally faces material climate risk by 2050 under RCP4.5 — directly relevant to Prologis's coastal logistics hubs and Brookfield's global asset portfolio.
 Back to Model Catalog  Open in Workbench