The boundary layer, the atmosphere between the ground and roughly 10km altitude, is where most short-range forecast error originates and has historically been the hardest layer to observe routinely. Aircraft data is sporadic and often missing humidity, which is key. Satellite data is least accurate exactly where the surface interferes most.

This session presents Meteodrone Strato, a weather drone capable of profiling the full atmospheric column up to 10,000m ASL, and the operational network already running unattended across Switzerland and Norway. Results are validated against wind-mast and radiosonde data and meet OSCAR goal requirements. The talk covers why this matters in practical terms: earlier fog and outage detection, improved convective forecasting, and a shift from hardware ownership to a Data-as-a-Service model where customers pay for atmospheric profiles rather than drones.

It closes with NOAA's 3D Mesonet programme as a real-world example of this approach applied at national scale. Attendees will leave with a clear picture of how routine, high-frequency vertical profiling is closing a long-standing observation gap, and what a service-based delivery model looks like in practice.