Extreme-Heat Exposure

What this driver is

Extreme heat exposure is one continuing physical-hazard mechanism: how often and how intensely populations, outdoor workers, and cities are exposed to heat that exceeds health and infrastructure thresholds. It is a mechanism, not a single heatwave headline, not a tourism brochure, and not a welfare score.

The driver is not climate crisis (the warming bundle), not climate adaptation readiness, not wildfire risk, and not drought risk. Heat exposure is the hazard reaching bodies and grids. Readiness is the capacity to absorb it. Wildfire is fire weather.

It applies globally from WMO State of the Climate heat records through IPCC AR6 WG2 heat-health chapters and WHO heat-health guidance. It is not ukraine-war.

The live Factrail series uses an Our World in Data natural-disasters-by-type count as a disclosed, remote proxy. Stored points include 0.38823677 (2020), 0.30038097 (2021), 0.80465332 (2022), 0.82958644 (2023), 0.8519443 (2024), and 0.15019048 (2025). Raw event counts on that chart (5, 3, 40, 45, 50, 1) are OWID’s. A 2025 raw count of 1 is likely an incomplete year, not a unique Factrail “heat ended” coefficient. This page does not invent a unique heat-death integer or Contribution.size.

Mechanism

Heat moves welfare through named channels.

First, physiology. Wet-bulb and apparent-temperature thresholds convert outdoor work and indoor poorly cooled housing into heat illness. WHO and WMO treat heat as a health hazard with dated guidance, not a mood.

Second, labour. Outdoor and poorly cooled indoor work lose hours. That is an hours channel, not a unique GDP size invented here.

Third, power. Cooling load stresses grids. Blackouts then raise indoor heat. Neighbouring energy security.

Fourth, compounding. Heat plus drought plus wildfire smoke is a compound event. Those companions stay on their slugs.

Fifth, attribution. WMO and IPCC treat the raised GMST baseline as increasing the probability of extreme heat. Climate crisis is that baseline. This node is the exposure.

Timing is hours for a heatwave and a season for a death certificate wave. A later WMO heat-record year would be a new fact on the crisis node and an exposure year here.

Evidence base

OWID natural-disasters-by-type CSV is the live proxy (https://ourworldindata.org/grapher/natural-disasters-by-type.csv (opens in a new tab)). WMO State of the Global Climate reports are the official heat-record instruments (https://wmo.int/publication-series/state-of-global-climate (opens in a new tab)). IPCC AR6 WG2 is the heat-health assessment. WHO heat-health action plan guidance is the operational health instrument.

Evidence is strongest for a WMO record year and a WHO guidance text. It is weaker for any unique first-day death coefficient. It is weakest where a disaster-event count of “1” in 2025 is asked to prove heat ended.

Effect on welfare indicators

• Exposure to extreme temperature: direction higher exposure when the heat hazard intensifies. Horizon: the same season. Confidence: high for the sign. No invented death size.
• Under-five mortality: direction higher deaths in documented heat-health failures. Horizon: days to a season. Confidence: medium locally, low as a global coefficient.
• Inability to keep home adequately warm: this series is winter-coded; heat is a neighbouring indoor-temperature problem, not this indicator. Missing coefficients stay missing.

The welfare sign of rising extreme-heat exposure is negative.

Interactions and contested points

Interacts with climate crisis, climate adaptation readiness, wildfire risk, drought risk, water scarcity, and energy security. Reinforcement is obvious when a record GMST year and a heat-health wave sit together. Offsetting is a heat-health plan that actually reaches outdoor workers.

The strongest competing view is that an OWID disaster-type count is too coarse to carry heat exposure. That objection is serious. A WMO annual heat-record table, or a WHO heat-health action-plan coverage series, would be a better meter. Until then the live points stay as lineage.

It does not. Crisis is the warming stock. This node is the heat hazard reaching people. A third competing view is that 2025’s stored low value means heat receded. A raw count of 1 on an incomplete year is not a completed decline. Quote OWID with the year; do not invent a unique Factrail heat-death total.

Established: live OWID points cited above; WMO/IPCC/WHO as instrument class; distinction from wildfire and from adaptation-readiness. Not established: a unique Factrail heat-death total, or that 2025’s stored low value is a completed year. A later WMO heat table would be a better meter. It would not mint a second heat driver. Night-time heat, wet-bulb thresholds, and outdoor-worker hours remain neighbouring measurements. They are not invented here as Contribution.size. A heat-health action plan with a date is a constructive instrument on the adaptation companion; it does not mint a second heat-exposure driver.