Thesis (MS)--Montana State University--Bozeman, 2010. / Typescript. Chairperson, Graduate Committee: Joseph A. Shaw. Includes bibliographical references (leaves 74-77).

In order to estimate the oxygen fugacity of Hawaiian lavas I have measured the ferric/ferrous ratios of samples from the 1984 eruption of Mauna Loa volcano and from the ongoing (1983-2017) Puu Oo eruption of Kilauea volcano. Fifteen samples were studied from the 21 day Mauna Loa eruption and 86 samples, erupted between 1983 and 2004, of the 34 year long Kilauea eruption. Both studies show that, in order to obtain reliable estimates of oxygen fugacity when, where, and how basaltic lava is sampled is of critical importance. Water-quenched lavas and spatter sampled at, or near vents, are less oxidized than water-quenched samples taken from open flow channels several kilometers away from the vent, or from slowly-cooled solidified flows. Additionally, samples of water-quenched lava traveling in lava tubes are less prone to oxidation than lava flowing in open channels, with oxygen fugacities similar to those of near vent quenched samples. The oxidation state of the rapidly quenched near vent or lava tube samples is at or below magnetite-wüstite (MW). This contrasts with the oxidation state of previously reported values for Hawaiian lavas, which are closer to fayalite-magnetite-quartz (FMQ) or nickel-nickel oxide (NNO). From this I conclude that the initial oxygen fugacity of Hawaiian parental magmas is close to MW and not FMQ, and that previous estimates of the oxidation state of Hawaiian magmas, based on data from solidified lava flows, were too high. This implies that the plume source of both Mauna Loa and Kilauea magmas is also close to MW, but not as reduced as the mantle source of mid-ocean ridge basalts.

In 1960, Kīlauea volcano in Hawaii erupted, destroying most of the village of Kapoho and forcing evacuation of its approximately 300 residents. A large and unprecedented social science survey was undertaken during the eruption to develop an understanding of human behavior, beliefs, and coping strategies among the adult evacuees (n = 160). Identical studies were also performed in three control towns located at varying distances from the eruption site (n = 478). During these studies data were collected that characterized ethnic grouping and attitudes toward Hawaiian cultural issues such as belief in Pele and two lava flow mitigation measures-use of barriers and bombs to influence the flow of lava, but the data were never published. Using these forgotten data, we examined the relationship between Hawaiian cultural issues and attitudes toward the use of barriers and bombs as mitigation strategies to protect Kapoho. On average, 72% of respondents favored the construction of earthen barriers to hold back or divert lava and protect Kapoho, but far fewer agreed with the military's use of bombs (14%) to protect Kapoho. In contrast, about one-third of respondents conditionally agreed with the use of bombs. It is suggested that local participation in the bombing strategy may explain the increased conditional acceptance of bombs as a mitigation tool, although this can not be conclusively demonstrated. Belief in Pele and being of Hawaiian ethnicity did not reduce support for the use of barriers, but did reduce support for bombs in both bombing scenarios. The disparity in levels of acceptance of barriers versus bombing and of one bombing strategy versus another suggests that historically public attitudes toward lava flow hazard mitigation strategies were complex. A modern comparative study is needed before the next damaging eruption to inform debates and decisions about whether or not to interfere with the flow of lava. Recent changes in the current eruption of Kīlauea make this a timely topic.

The 2014-15 lava flow crisis at Kīlauea volcano, Hawai‘i and post-September 2015 elevated unrest at adjacent Mauna Loa volcano provided opportunities to assess households’ psychological and behavioral responses to different levels of volcanic activity. Weused the Protective Action Decision Model to examine stakeholder perceptions and confidence in warnings, in addition toattitudes toward lava flow mitigation strategies, such as diversion by berms and bombing, and people’s acceptance of additional risk to personal property in exchange for protecting important elements of their community, such as schools, major roads, electrical substation, and shopping centers. Respondents’ confidence in events important in decision-making during emergencies and evacuations were significantly correlated with their past experience with lava forecasts. Consistent with previous studies, overall support for the two different mitigation measures was higher for earthen berms/trenches than it was for bombing/blasting. Finally, diversion acceptance was strongly correlated with residents’ perceptions of lava flow diversion strategies.