At What Warming Does Your Neighbourhood Break?
SUEWS run at 6 warming increments reveals the non-linear threshold where heat becomes unworkable
exceeds 100 dangerous hours
from 0 to +2.5°C
(0 to +2.5°C)
Kampong Lama at +2.5°C
The response is non-linear. Dhobi Lines goes from 44h (present) → 70h (+0.5°C) → 113h (+1.0°C) → 182h (+1.5°C) → 276h (+2.0°C) → 424h (+2.5°C). Each half-degree of warming adds more dangerous hours than the last — because more timesteps cross the 35°C threshold as the baseline shifts upward.
Dangerous Heat Hours: The Slider
Drag the slider to see how each neighbourhood responds to warming. The non-linearity is immediately visible — hotspot neighbourhoods accelerate while cores stay cool until higher warming levels.
From Physics to People: Unsafe Work Hours
The ILO estimates that by 2030, heat stress will cost the global economy 80 million full-time jobs worth of productivity. We can translate our model output directly into a policy-relevant metric: hours where outdoor work should cease.
This is not an assumption — it's a direct application of the ILO occupational heat guideline (no outdoor labour above 35°C dry-bulb). We multiply dangerous hours by each neighbourhood's outdoor worker fraction (from the challenge dataset's socioeconomic data). The result shows who loses the most working time.
At +1.5°C warming, Kampong Lama's outdoor workers lose 150 hours of safe working time in a two-month period. That's 62% of the workforce losing ~25% of their productive hours. For daily-wage workers with no income safety net, this is not an inconvenience — it's a poverty trap.
Where Things Break: The Tipping Increments
| Neighbourhood | Type | Present (h) | +0.5°C | +1.0°C | +1.5°C | +2.0°C | +2.5°C |
|---|---|---|---|---|---|---|---|
| Jade Gardens | Refuge | 75 | 108 | 163 | 239 | 353 | 513 |
| Kampong Lama | Hotspot | 68 | 103 | 164 | 242 | 345 | 507 |
| Dhobi Lines | Hotspot | 44 | 70 | 113 | 182 | 276 | 424 |
| Fuzhou Lanes | Hotspot | 36 | 60 | 98 | 156 | 248 | 395 |
| Mlima Moto | Hotspot | 5 | 20 | 44 | 75 | 137 | 277 |
| Zheng He Towers | Core | 0 | 0 | 5 | 19 | 59 | 132 |
Key observations:
- Kampong Lama breaks at +0.5°C — crosses 100 dangerous hours, equivalent to losing >10% of working time for outdoor workers
- Mlima Moto's cliff is between +1.0°C and +1.5°C — goes from 45h to 92h, nearly doubling in a single half-degree increment
- Zheng He Towers (CBD) is protected until +1.5°C — roughness-driven ventilation from tall buildings provides thermal resilience
- The gap between rich and poor neighbourhoods widens with warming — inequality is amplified by physics
At 35°C dry-bulb with 81% relative humidity, wet-bulb temperature is approximately 31°C — the empirically determined limit of human thermoregulation (Vecellio et al., PNAS 2023). Our threshold is not arbitrary; it marks where human physiology fails.
What This Means
The slider reveals three things that a single scenario comparison cannot:
- The response is non-linear — each half-degree of warming is worse than the last. Policy based on linear extrapolation will underestimate future risk.
- Neighbourhoods have different tipping points — defined by their current proximity to the threshold. This creates an equity argument: the poorest neighbourhoods cross into danger first.
- Physical interventions buy time, not immunity — cool roofs eliminate ~88% of present-day dangerous hours, but under +2.5°C even cooled neighbourhoods would face hundreds of hours above threshold. The headroom that interventions provide is quantifiable: approximately +1.5°C of breathing room.
Method
- Model: SUEWS v2026.6.5 via SuPy, 10 neighbourhoods, 5-min timestep, 15-min output
- Warming method: Linear interpolation of forcing Tair and incoming longwave radiation (ldown) between present (0°C) and future (+2.5°C) datasets at 0.5°C increments
- Analysis window: April–May (61 days), excluding 30-day spinup
- Threshold: 35°C dry-bulb (≈31°C wet-bulb at 81% RH — Vecellio et al. critical limit)
- Outdoor worker fraction: From challenge dataset socioeconomic.csv (range: 18–65%)
- Unsafe work hours: Dangerous hours × outdoor worker fraction (ILO occupational heat guideline)
- Limitation: Uniform warming applied — real climate change is heterogeneous. Treat as stress test, not prediction.
Citations
Vecellio, D.J. et al. (2023). Greatly enhanced risk to humans as a consequence of empirically determined lower moist heat stress tolerance. PNAS, 120(42). DOI
ILO (2019). Working on a Warmer Planet: The Impact of Heat Stress on Labour Productivity and Decent Work. International Labour Organization, Geneva.
Järvi, L., Grimmond, C.S.B. & Christen, A. (2011). The Surface Urban Energy and Water Balance Scheme (SUEWS). J. Hydrol., 411(3–4), 219–237. DOI