SCIENCE & METHODOLOGY
Every risk score and financial metric UrClimate produces starts from peer-reviewed climate science and ends as a decision-grade, audit-defensible number. This page explains, transparently, how that path is built — from data to model, from hazard to financial impact.
Our scientific stance
In climate risk analytics, trust comes from transparency, not from a black box. We build on four principles:
Transparency
Which model, which scenario, which assumption — all explicitly documented. We do not ship black-box scores.
Reproducibility
Same input, same output. Every calculation is versioned; results can be reproduced.
Auditability
A traceable data lineage from source data to the final number — ready for assurance audits.
Standards alignment
Metrics and outputs are shaped to what regulators and frameworks expect.
Data foundation
The analysis rests not on a single source but on complementary layers:
Climate models and scenarios
- CMIP6 global model ensemble; methodology aligned with IPCC AR6
- SSP2-4.5 (middle-of-the-road, baseline) and SSP5-8.5 (high emissions, stress test)
- Annual projection: 2015–2100
- CMIP7 transition managed by us; CMIP6 is the valid reference for reporting
Observations and local data
- MGM meteorological station data (Türkiye)
- Air-quality indicators
- GPC-compliant emission factors for municipal inventories
- All 1,054 municipalities in Türkiye pre-defined
From the global model to your asset
Global climate models operate at continental scale; the decision for a facility, a loan or a municipality is made at local scale. We close that gap by downscaling projections to the asset location and aligning them with observations. The result is not a continental average but a coordinate-level risk view — at a resolution where you can click a single point on the map.
Hazard modeling
For each location and hazard, the climate index is projected by scenario and year and normalized to 0–1 for comparability (aligned with IPCC AR6). Critically, hazards are not treated as independent — co-occurring (compound) risk structures are part of the model.
Acute hazards
- Flood and extreme rainfall
- Storm and extreme wind
- Wildfire
- Extreme heat waves
Chronic hazards
- Sustained warming
- Drought
- Sea-level rise
- Shifting precipitation regimes
Physics to finance — the layer that sets us apart
Most providers stop either on the engineering side or the analytics side. We translate hazard directly into financial language — this is the layer that ties a climate score to the balance sheet:
Insurance & assets
AAL (Average Annual Loss), PML at multiple return periods, damage-state distribution. UrClimate Score.
Banking
PD / LGD / DSCR and EL = PD × LGD × EAD; IFRS 9 ECL, RWA and ICAAP impact. UrClimate Next.
Public & municipal
GPC-compliant GHG inventory, SECAP/CDP-ready risk report, evidence for climate finance. UrClimate Kent.
Validation and quality
A model is only as good as its grounding in reality. Our approach:
- Testing against observations: projections are checked for consistency against MGM station records and observed events.
- Versioning: model and data versions are recorded; the version behind any output is traceable.
- Documentation: scenario assumptions and calculation steps are documented to the level an assurance audit may ask for.
- Human oversight: automated outputs are reviewed by our expert engineering team.
An honest view of uncertainty
Standards and governance
Finance & climate frameworks
- NGFS · UNEP FI · IIGCC PCRAM 2.0
- IFRS S2 / TSRS · TCFD
- EBA Pillar 3 · BDDK climate guideline · BCBS 239
Public & reporting
- GPC (city GHG inventory)
- CDP Cities
- SECAP (Covenant of Mayors)
Frequently asked questions
Which climate models and scenarios does UrClimate use?
Physical climate projections are based on the CMIP6 model ensemble and aligned with IPCC AR6. Two core scenarios are offered: SSP2-4.5 (middle-of-the-road) as a baseline and SSP5-8.5 (high emissions) for stress testing; the projection axis is annual, 2015–2100. We manage the transition to CMIP7 when it arrives; CMIP6 remains the valid reference for reporting.
How are hazard scores produced?
For each location and hazard, the climate index is projected by scenario and year and normalized to 0–1 (aligned with IPCC AR6). Hazards are not treated as independent — co-occurring (compound) risk structures are part of the model. Acute hazards (flood, storm, extreme heat, wildfire) and chronic hazards (drought, sea-level rise, sustained warming) are handled distinctly.
How are results translated into financial metrics?
Physical hazard is translated directly into financial language: on the insurance/asset side, AAL (Average Annual Loss) and PML at multiple return periods; in banking, PD/LGD/DSCR and EL = PD × LGD × EAD with IFRS 9 ECL, RWA and ICAAP impact. This turns hazard into an auditable, currency-denominated number.
Which standards is the methodology aligned with?
The methodology aligns with NGFS, UNEP FI, IIGCC PCRAM 2.0, IFRS S2 / TSRS, EBA Pillar 3, the BDDK climate guideline and BCBS 239; on the municipal side, GPC and CDP Cities. Outputs are shaped to the structure these frameworks expect.
Are the results defensible against audit?
Yes — that is central to the design. Behind every output, the source data, model version, scenario assumptions and calculation steps are kept traceable. In this era of limited assurance, every reported number must be reproducible and documentable.
See the methodology on your own data
Let’s have a technical call with your team over a sample output and the methodology documentation.
The methodology statements on this page are consistent with the approach published on the UrClimate product pages; the climate-science framing (CMIP6, SSP scenarios, IPCC AR6) rests on public standards.