Hypothesis
Signaling data from a mobile network could enumerate backcountry travellers in avalanche terrain, supplying the exposure denominator needed for a fatal-accident rate.
What failed
The method produced estimates for the 2019/2020 Norwegian winter season (13,666 phones, 0–118 per day) and weak correlations with daylight, weekends and bulletin views, but validation against GPS reference positions revealed discrepancies between estimated and true positions.
Broken assumption
Mobile-network signaling data locates people accurately enough outside cities to count them in specific terrain.
Why
The authors attribute the mismatch to an algorithm designed for urban mobility and to the long distances between base transceiver stations in mountainous terrain, concluding this limits signaling data's usefulness outside urban regions.
From the published paper: Toft et al. Journal of Trial and Error, 2023. ↗CC BY