Cold Wave: how HazardNet classifies and scores it

How far the minimum falls below 16 °C, plus persistence.

How the model arrives at this class

Classified from the same drivers; the districts that score highest are the ones where the minimum genuinely falls, not an a-priori cold-region list.

Uncalibrated softmax. Fog — the mechanism that makes a Bangladesh cold wave dangerous — is not a driver.

Nothing above the WATCH level can be published automatically while no calibration map exists, and nothing at all can be published while the pipeline stamps no model version — the alert engine records that as `publication_blocked` rather than issuing the alert.

The independent physics cross-check

16 °C is the reference minimum at which the temperature term starts scoring; it is a formula constant, not a met-service threshold.

Drivers: minimum temperature in the horizon (T_min, °C) · length of the horizon (days).

clip(x) clamps a term to [0, 1] — max(0, min(1, x)) — which is how every formula in the pipeline bounds its terms. Each term is clipped before it is weighted, so no single driver can run away with a score.

  • Expression: 0.7*clip((16-T_min)/10) + 0.3*clip(duration/5)
  • That expression is executed against scripts/physics_severity.py by scripts/tests/test_content_engine.py — this page cannot silently describe a formula the pipeline no longer runs.

What the current run says about Cold Wave

The run this deployment ships (prediction date 2026-09-16) classifies no district-horizon unit as Cold Wave. An absence of cold wave in one run is not a statement that the hazard cannot occur this season.

Season and geography

Typical season: December–February, concentrated in the north and north-west. This describes when the hazard is climatologically plausible, not when this run flags it.

What this class cannot tell you

  • No fog, no wind chill, no humidity: a dry 9 °C night and a foggy 9 °C night score the same.
  • BMD's own cold-wave criteria (which are station-based and use consecutive-day rules) are not what this proxy implements; the two can disagree, and this page says so rather than implying equivalence.
  • Confidence bins (Certain ≥ 0.85, Probable 0.70–0.85, Uncertain < 0.70) describe the model's own certainty in the class it chose. They are not accuracy, and no accuracy figure is published because no held-out evaluation has been run on real labels.

Inputs behind it

Ground-truth reports (district, hazard, horizon, date, what was observed) are the only route by which this project can publish skill metrics. The contact page reaches the maintainers.

  • Open-Meteo forecast (temperature)
  • ERA5-Land daily aggregates

Questions and answers

Is a high Cold Wave severity a prediction of damage?

No. The severity index is the model's continuous score for how strongly the drivers resemble this class, on a 0.00–1.00 scale. It is not a probability, not a percentage, and it does not model exposure. The physics cross-check is a second, independent estimate; where the two diverge, both numbers are shown rather than averaged.

Why is the confidence not a probability of Cold Wave?

The published confidence is the model's own softmax over its eight classes, labelled uncalibrated_model_softmax in the data. It says how certain the classifier is about the class it picked, not how often that class actually materialises. No calibration map has been fitted yet, which is also why nothing above the WATCH level can be published automatically.

Which districts does this page cover?

The district pages cover all 64 districts. This page summarises the districts the current run covers — a run can be partial, and the coverage stamp on the snapshot says exactly how partial, which the status page reports.

Content reviewed 2026-09-16. HazardNet is decision support, not an official warning service — see the methodology for scope and limitations.

Loading the interactive HazardNet application…