From 𝜃 and 𝜃𝑤 Profiles to Convective Potential: A Layer-by-Layer Stability Diagnosis of Tropical and Subarctic Radiosonde Soundings

Vertical stability controls whether an air column suppresses or releases convection. We implement areproducible, layer-by-layer diagnosis of static stability from radiosonde soundings in Python. The vertical tendencies ofthe potential temperature 𝜃 and of the pseudo-adiabatic (wet-bulb) potential temperature 𝜃𝑤, obtained by integrating thepseudo-adiabatic lapse rate with MetPy, are combined to classify each layer as absolutely stable, dry unstable, absolutelyunstable, or convectively unstable. The convectively unstable layers are further split into selective and latent instability.We apply the method to three contrasting soundings: Boa Vista (tropical Amazonia, 1 January 2023), Brasília (tropicalhighlands, 24 March 2024) and Goose Bay (subarctic Canada, 24 March 2024), and compare it with parcel diagnostics.The tropical columns are convectively unstable over 36% and 30% of their depth, and their unstable layers enclosethe level of free convection. These soundings carry 395 J kg−1 and 134 J kg−1 of CAPE and 54.5mm and 44.2mm ofprecipitable water. The subarctic column is 95% absolutely stable, with zero CAPE. The two layers flagged unstable atGoose Bay are shown to be artefacts of a strict inequality applied to neutral, constant-𝜃 layers. This motivates an explicitneutral class.