PM2.5 shock (SSP3-7.0)

Source: IPCC CMIP6 (GFDL-ESM4, GISS-E2-1-G, MIROC-ES2L) Year: 2050 Category: Scenario Coverage: Global Format: Raster grid (~100 km) Used in risk analysis: No (contextual layer; not used in materiality scoring)

What it shows

This layer measures the projected change in fine particulate matter (PM2.5) air pollution between today and mid-century under a high-emissions "regional rivalry" future. Instead of showing absolute pollution levels, it highlights where air quality is expected to deteriorate the most, so that future change rather than present conditions drives the air-pollution risk signal.

Note on the scenario. This is the pessimistic pathway used across Darwin's stress test. It is SSP3-7.0, which replaced the previous SSP5-8.5 layer — the IPCC AR6 treats SSP5-8.5 as physically implausible, so SSP3-7.0 is now the high-end reference.

How it is built

The shock is derived from CMIP6 climate-model projections under SSP3-7.0, a high-emissions, regional-rivalry pathway. For each location it compares projected 2050 PM2.5 against a 2014 baseline and expresses the difference as a relative change. The signal is dampened in already very clean areas: where baseline concentrations fall below a health-relevant "very low" threshold, the change is attenuated so that ultra-clean regions such as the Arctic and open ocean do not register misleadingly large percentage shocks. The output is a coarse global grid (roughly 100 km cells).

How to read it

Higher positive values mean a larger projected worsening of fine-particulate pollution by 2050; values near zero indicate little expected change; negative values indicate projected improvement. As clean baseline regions are deliberately down-weighted, the strongest signals appear in already-populated or polluted areas. This pathway typically shows larger increases than the sustainable-development variant.

Class thresholds

The underlying quantity is a physical flux expressed as a relative change: the projected percentage change in fine-particulate (PM2.5) air pollution between a 2014 baseline and a 2050 projection, under the SSP3-7.0 pathway (% change). It is computed as (PM2.5_2050 − PM2.5_2014) / (PM2.5_2014 + 1st_decile) × weight, where weight = min(1, PM2.5_2014 / 5.8×10⁻⁹ kg/kg) attenuates the signal in ultra-clean regions (concentrations below the health-relevant "very low" threshold of 5.8×10⁻⁹ kg/kg).

This layer is not split into discrete risk classes. It is symbolised on the PM2.5 purity shock (%) field as a continuous diverging gradient over its native unit, the shock percentage:

The layer is contextual and is not used in materiality scoring, so it carries no five-class risk table. It does, however, define a single trigger: a location is marked as a High PM2.5 shock zone when the projected degradation exceeds 30 % (purity shock below −30 %), which feeds the Enabling/Supporting Services dimension.

How these thresholds were set. This is a Darwin in-house classification: the PM2.5 concentrations come from the CMIP6 multi-model ensemble (GFDL-ESM4, GISS-E2-1-G, MIROC-ES2L) under SSP3-7.0, but the shock formula, the 5.8×10⁻⁹ kg/kg attenuation threshold, the ±90 % gradient anchors and the −30 % High-shock trigger are fixed round-number cutoffs chosen by Darwin's scenario pipeline, not a published CMIP6 or WWF class break.

Reference: CMIP6 (WCRP Coupled Model Intercomparison Project Phase 6) — wcrp-climate.org/wgcm-cmip/wgcm-cmip6.

Source

CMIP6 climate models (WGCM CMIP6 ensemble), models GFDL-ESM4, GISS-E2-1-G and MIROC-ES2L, scenario SSP3-7.0; 2014 baseline versus 2050 projection.

Comparison with the WWF Risk Filter Suite

The WWF Risk Filter Suite publishes present-day risk indicators, not forward-looking scenario shocks. This layer is a temporal change-variant of an air-quality risk WWF assesses only for current conditions, so it has no distinct WWF equivalent and is a Darwin scenario extension.

Legend

Symbolised field: PM2.5 purity shock (%)

Generated from darwin/layers/layer-pm25-shock-ssp5.toml and risk_indicator_pairs.toml (develop).