Techniques for constraining short term eruptive processes at Kilauea's Overlook Vent

The activity within Kilauea's summit eruptive vent has provided ongoing opportunities to study lava lake dynamics, especially since the active lava surface became persistently visible in early 2010. Patrick et al. attributed 20-30 m lava rise/fall cycles to accumulation and release of gas beneath a shallow crust (2010), and Nadeau et al. (2010) reported rapid SO2 release during two of these rise/fall cycles. Since that time, the summit eruptive vent has widened incrementally from about 70 to more than 160 meters in diameter, and this expanding geometry has been accompanied overall changes in pond dynamics. Recently, pond activity has been temporally dominated by relatively constant magmatic convection with active and persistent gas release. This predominant mode of activity, especially during the past year, has been punctuated by short, (minutes-to-hours-long) more passive periods of little spattering, with decreased magmatic circulation, that usually end in brief, turbulent upset of the lava lake followed by a return to steadier active convection and gas release. In addition to thermal and conventional webcams and seismic monitoring, an array of upward-looking UV spectrometers (Horton et al., 2012) and an SO2 camera system (Kern et al., session V043) are now tracking changes in Kilauea summit degassing, and inferentially, changes in pond activity, by continuously recording SO2 gas release rates during daylight hours. The continuous SO2 emission rate data combined with continuous camera, seismic, and gravity measurements, all support a hypothesis that the dominant process controlling short term behavior of the lava lake surface is shallow--occurring within a kilometer of less of the active free surface. SO2 exsolution occurs predominately at shallow depths (hundreds of meters or less) and would reasonably produce a low-density, bubble-dominated fluid. Observationally, upset of the lava lake surface is accompanied by rapid and coincident increases in SO2 emissions and high frequency seismic amplitude. Continuous gravity measurements indicate a density consistent with a foamy fluid. Transition to passive pond activity results in lowered observed SO2 emissions and seismic amplitude. The shallow process hypothesis is further bolstered by time-series FTIR analyses of gases boiling out of the melt prior to and during turbulent events. These data show that CO2/SO2 ratios drop markedly during lava lake upset intervals, owing to increases in shallowly exsolving gases including SO2, H2O, HCl, and HF. The short term eruptive 'signal' produced by the active lava lake within the summit eruptive vent is superimposed on a typically more slowly-varying 'signal' dictated by deeper, magma supply-related processes. Careful monitoring of short term eruptive processes by the techniques described here is providing a better understanding of the shallow system. This monitoring, combined with contemporaneous observations of slowly-varying surface deformation, seismic energy release, and deep gas release (CO2), are steadily improving our overall understanding of how Kilauea and similar volcanoes work.