This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues.Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. and we concurrently derived SO 2 masses for more than 130 Strombolian explosions and 50 gas puffs. From this, we show erupted SO 2 masses have a variability of up to one order of magnitude, and range between 2 and 55 kg (average $ 20 kg), corresponding to a time integrated flux of 0.05 7 0.01 kg s À 1 . Our experimental constraints on individual gas puff mass (0.03-0.42 kg of SO 2 , averaging 0.19 kg) are the first of their kind, equating to an emission rate ranging from 0.02 to 0.27 kg s À 1 . On this basis, we conclude that puffing is two times more efficient than Strombolian explosions in the magmatic degassing process, and that active degassing (explosions þpuffing) accounts for $ 23% (ranging from 10% to 45%) of the volcano's total SO 2 flux, e.g., passive degassing between the explosions contributes the majority ( $ 77%) of the released gas. We furthermore integrate our UV camera gas data for the explosions and puffs, with independent geophysical data (infrared radiometer data and very long period seismicity), to offer key and novel insights into the degassing dynamics within the shallow conduit systems of this open-vent volcano.