Imagine you're standing at the edge of a swimming pool, watching a coin on the bottom. As the water temperature shifts — warm spots mixing with cool — the coin appears to wobble and drift. You're not measuring the coin; you're measuring the water, using the coin's apparent position as a probe. That is exactly what ADSBi does, except the 'coins' are commercial aircraft broadcasting their positions via ADS-B, and the 'water' is the lower atmosphere's humidity field bending those 1090 MHz radio signals on their way to a ground-based interferometer. The committed claim: a ground-based interferometer measuring ADS-B signals from commercial aircraft can retrieve high-resolution refractivity gradient profiles of the lower atmosphere, capturing spatial and temporal humidity variability that existing observing systems miss. This is not the first use of signals of opportunity for atmospheric sounding — GPS radio occultation has done this from orbit for decades — but it is the first demonstration that the ubiquitous ADS-B transmissions already filling the sky can serve as a dense, free source of refractivity measurements at scales that matter for weather forecasting. The architecture is elegantly parasitic. ADS-B is a mandatory transponder system: every commercial aircraft broadcasts position, altitude, velocity, and identity at 1090 MHz, roughly twice per second. The ADSBi technique places an interferometer on the ground and measures the angle-of-arrival of these transmissions. Because the aircraft's true position is encoded in the broadcast itself, any discrepancy between the expected and observed arrival angle is attributable to atmospheric refraction — primarily driven by humidity gradients in the boundary layer. By comparing signals from aircraft at different distances and altitudes simultaneously, the authors reconstruct vertical refractivity profiles. The key structural advantage: aircraft are everywhere, flying at all altitudes through the boundary layer and free troposphere, providing continuous geometric diversity that dedicated satellite systems cannot match at local scales. The integrity picture is early-stage. This is an 8-page paper with 7 figures — a demonstration of concept, not a mature validation campaign. The comparison baseline is the existing refractivity measurement ecosystem: radiosondes (launched twice daily, drifting with wind, sparse spatial coverage), GPS radio occultation (good vertical resolution but ~300 km horizontal smearing), and ground-based GNSS networks (column-integrated, no vertical profile). The paper demonstrates sensitivity to small-scale variability that these systems cannot resolve, but the quantitative error budget against radiosonde truth data is not described in the abstract. The authors are from the UK Met Office (Kitchen, Bowler) and affiliated institutions, which lends credibility — these are operational meteorologists who know what the forecasting community actually needs. The milestone question is where this gets interesting for weather prediction. Numerical weather prediction models are fundamentally limited by moisture initialization in the boundary layer — it's the single largest source of forecast error for convective events. Current radiosonde networks provide profiles at ~250 km spacing every 12 hours. If ADSBi can deliver continuous refractivity profiles at ~10 km spacing using existing air traffic, that's a step-change in observation density. The concrete next number: demonstrating retrieval accuracy within 1-2 N-units of radiosonde truth across a statistically meaningful sample of atmospheric conditions, which would make this viable for operational data assimilation. The obvious experiment not yet run — or at least not described in this abstract — is a systematic validation campaign comparing ADSBi retrievals against co-located radiosonde launches across diverse weather conditions (clear sky, frontal passages, convective environments). The honest read: this is almost certainly being planned or underway, given the Met Office affiliation. The authors are likely saving it for the follow-up paper that makes the operational case. The other gap: quantifying how ADS-B signal density (which varies enormously between busy airspace near airports and sparse oceanic routes) affects retrieval quality. The technique is only as good as its aircraft coverage, and that coverage is not uniform.
