Natural climate variation, such as that caused by volcanoes, is the basis for identifying anthropogenic climate change. However, knowledge of the history of volcanic activity is inadequate, particularly concerning the explosivity of specific events. Some material is deposited in ice cores, but the concentration of glacial sulfate does not distinguish between tropospheric and stratospheric eruptions. Stable sulfur isotope abundances contain additional information, and recent studies show a correlation between volcanic plumes that reach the stratosphere and mass-independent anomalies in sulfur isotopes in glacial sulfate. We describe a mechanism, photoexcitation of SO2, that links the two, yielding a useful metric of the explosivity of historic volcanic events. A plume model of S(IV) to S(VI) conversion was constructed including photochemistry, entrainment of background air, and sulfate deposition. Isotopologue-specific photoexcitation rates were calculated based on the UV absorption cross-sections of (32)SO2, (33)SO2, (34)SO2, and (36)SO2 from 250 to 320 nm. The model shows that UV photoexcitation is enhanced with altitude, whereas mass-dependent oxidation, such as SO2 + OH, is suppressed by in situ plume chemistry, allowing the production and preservation of a mass-independent sulfur isotope anomaly in the sulfate product. The model accounts for the amplitude, phases, and time development of Δ(33)S/δ(34)S and Δ(36)S/Δ(33)S found in glacial samples. We are able to identify the process controlling mass-independent sulfur isotope anomalies in the modern atmosphere. This mechanism is the basis of identifying the magnitude of historic volcanic events.