Radon (222Rn) and thoron (220Rn) are widely used to investigate diffuse degassing, fault-zone permeability, hydrothermal circulation, and subsurface unrest, but their signals are not direct proxies for a single process. This manuscript is a critical synthesis and methodological article that develops a reduced inference framework for interpreting radon–thoron observations in volcanic, tectonic, and hydrothermal settings. The framework separates accessible support of the immediate radium parents 226Ra and 224Ra, recoil-scale release into the mobile phase, multiphase transport, geological carrier-gas throughput, and observational closure. It also distinguishes total activity flux from activity concentration and chamber throughput from natural carrier-gas dilution. Synthetic illustrative experiments test the internal behavior of the reduced operator; a concise re-reading of the public Upper Rhine Graben dataset illustrates the limits of concentration-only inference; and published volcanic and hydrothermal examples are used as literature-grounded vignettes. The purpose is not to validate a universal inversion model but to define what can be inferred from different observation packages. The paper, therefore, emphasizes three operational levels: anomaly reporting, mechanism discrimination, and local inversion.
Sebastiano Ettore Spoto (Sat,) studied this question.