Krakatau revisited: The course of events and interpretation of the 1883 eruption

Abstract

Magma chamber over-pressuring by volatile saturation and/or a magma mixing event may have triggered the 1883 eruption of Krakatau. From the beginning of activity on 20 May to the onset of the 22–24 hour-long climactic phase on 26–27 August, Krakatau produced a discontinuous series of vulcanian to sub-plinian eruptions. Based on contemporary descriptions, the intensity of these phases may previously have been underestimated. The most realistic estimate of eruptive volume (magnitude) is about 10 km3 of dacitic magma. The climax of the eruption began at 1:00 pm on 26 August with a plinian phase which led into a 5-hour-long ignimbrite-producing phase. Caldera collapse most probably occurred near the end of the eruption on 27 August, precluding large scale magma-seawater interaction as a major influence on the eruption column and characteristics of the pyroclastic deposits. Very rapid displacement of the sea by pyroclastic flows remains the best explanation for the series of catastrophic sea waves that devastated the shores of the Sunda Straits, with the last and largest tsunami coinciding with the slumping of half of Rakata cone into the actively forming caldera, perhaps during a period of great pyroclastic flow production. The large audible explosions recorded on 27 August may have been the rapid ejection of large pulses of magma that collapsed to form pyroclastic flows in the ignimbrite-forming phase. Co-ignimbrite ash columns rising in the atmosphere immediately after the generation of each major pyroclastic flow may have contributed to the magnitude of the air waves. A reappraisal of the eruption in the light of this, in conjunction with the pressure (air wave) and tide gauge (tsunami) records from Jakarta, suggests that the relationship between the latter two has been oversimplified in previous studies. Tsunami travel times from Krakatau to Jakarta probably varied more than hitherto thought and there need not be a simple correlation between the times of the explosions and the initiation of the tsunamis. However, tsunamis in the Sunda Straits and vicinity probably were not caused or influenced by coupling with the air waves. Various hypotheses about the cause of the tsunamis and explosions are reviewed and it is concluded that the cause of both is most likely related to the sudden emission of large pulses of magma that led to formation of the Krakatau ignimbrite.