Short- and long-term deformation of Kilauea and Mauna Loa volcanoes, Hawaii
Abstract/Contents
- Abstract
- Kilauea volcano, Hawaii has sustained a near continuous eruption interrupted by multiple intrusions within its east rift zone (ERZ) and experienced multiple slow-slip events (SSEs) since 1983. Klauea also experiences large earthquakes, some with tsunamigenic potential. This magmatic and tectonic activity marks Kilauea as one of Earth's most active volcanoes. Though not currently active, Mauna Loa is Earth's largest volcano and is capable of producing rapidly-moving lava flows that can travel from the eruptive fissure to the coast within four hours. Assessing the short- and long-term volcanic and seismic hazards at these volcanoes requires an understanding of the active structures on which these events occur. The work presented here uses geodetic data recorded at Kilauea and Mauna Loa by tilt meters, continuously recording global positioning system (GPS) stations, campaign GPS stations, and satellite based synthetic aperture radar. These data are used with linear and non-linear optimization techniques to analyze short- and long-term deformation due to intrusions, SSEs, and secular slip. The first simultaneous intrusion and SSE ever observed at Kilauea occurred 17-19 June 2007. The complex series of events were recorded in great detail by GPS, tilt meters, and synthetic aperture radar. Markov-Chain Monte Carlo optimization of GPS, tilt, and interferometric synthetic aperture radar (InSAR) data was used to determine source parameters for a single dike, represented as a uniform opening rectangular dislocation, and a summit magma source, represented as a point pressure source, in a homogeneous, isotropic, linear half-space. The optimum model is a an ENE striking dislocation with approximately 2 m of opening, a dip of approximately 80 degrees to the south, and extending from the surface to approximately 2 km depth. Determining the precise sequence of events during the 17 June 2007 intrusion and SSE is important since flank extension and rift opening are linked, i.e. intrusions may influence episodes of flank slip and vice versa. Tilt meters installed on the ERZ and the coast can provide data on the timing of the intrusion and SSE. Diurnally and tidally notch filtered tilt meter data show the intrusion likely preceded the slow-slip event. Eleven SSEs have been identified at Kilauea since 2001 when they were first observed as centimeter scale transients in continuous GPS data. These events release energy equivalent to a Mw 5-6 earthquake over 2-3 days as seaward slip along the decollement beneath Kilauea's south flank. Slow-slip and tremor propagation are known to occur during some SSEs in subduction zones. Tremor accompanying slow-slip has not been observed at Kilauea nor has definitive observation of slow-slip propagation. Electronic tilt meter data with microradian precision collected at one minute intervals during the 2007 intrusion and SSE, 2010 SSE, 2011 intrusion, and 2012 SSE were analyzed. The analyses indicate areas of slow-slip likely overlaps areas of secular slip along the decollement and that slow-slip likely propagated eastward during the 2010 and 2012 SSEs. Tentative evidence suggests an SSE occurred during the 2011 intrusion beginning east of coastal tilt meter KAE and propagating eastward. Installation of additional tilt meters along the coast to the east and west of KAE would provide data valuable to understanding SSE at Kilauea. Ground deformation recorded by GPS stations at Kilauea and Mauna Loa volcanoes, Hawaii, were analyzed using principal component analysis (PCA) with mode rotation. Though widely used in the Earth sciences to deconvolve time series data into distinct modes theoretically representing distinct geophysical sources, PCA is prone to a problem termed mode mixing. Mode mixing occurs when the signature of one geophysical source (e.g. an intrusion) appears in multiple modes, a significant problem if the modes are interpreted physically. Mode rotation attempts to ensure each mode represents a distinct geophysical source. The mode rotation technique uses linear combinations of modes to minimize the correlation between any two modes in a time period of interest, during which the geophysical event occurs. Eleven years (1998-2009) of detrended continuous GPS data at Kilauea and 11 years (1998-2009) of campaign data from Mauna Loa were analyzed. Following rotation, modes of deformation at K\={\i}lauea represent episodic east rift zone intrusions, movement of magma beneath the summit and rift zones, transient motion of the south flank associated with SSEs, and southwest rift zone inflation. The spatial density of stations is insufficient for PCA to distinguish individual intrusions (such as the 1999 and 2007 intrusions) spatially, but these events are recorded in the temporal modes. At Mauna Loa two modes representing summit magma sources and flank sliding were found. Modeling of these modes show decollement slip ceased beneath Mauna Loa in 2002. During the 2003-2004 time period Kilauea experienced few episodic events and relatively little summit deformation. Therefore GPS velocities and InSAR line-of-site velocities nominally capture secular deformation at Kilauea and can be used to model the active structures responsible for the observed deformation. A variety of models were tested whose basic elements include a southwest rift zone (SWRZ) that is continuous through the summit to the ERZ, a basal decollement that extends from north of the rift zones to tens of kilometers offshore, and a summit magma chamber represented by an isotropic pressure source. Results indicate that ERZ opening reaches a maximum near Makaopuhi crater at several kilometers depth; opening occurs at depth further downrift. SWRZ opening is negligible. Decollement slip south of the summit area extends beneath the south flank just offshore. High decollemet slip rates also occur within an aseismic area 3-4 km from the base of the rift zone; seismicity occurs south of this zone in areas of little to no decollement slip. Modeling of the average displacements due to the 2005 and 2010 SSEs show slipping areas during slow-slip may be distinct from secularly slipping areas, though we can not rule out overlap. Secularly slipping areas may represent velocity strengthening regions while areas of slow-slip and seismic activity may represent velocity weakening asperities transitioning to a velocity weakening area. This transition may be caused by a change in material properties due to dewatering and incipient metamorphism of sea floor sediments as pressure and temperature increase nearer the bulk of Kilauea and Mauna Loa. Large earthquakes such as the 1975 Kalapana earthquake may occur if rupture grows to sufficient size to link multiple velocity weakening areas.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2014 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Sinnett, Daniel Kenneth |
|---|---|
| Associated with | Stanford University, Department of Geophysics. |
| Primary advisor | Segall, Paul, 1954- |
| Thesis advisor | Segall, Paul, 1954- |
| Thesis advisor | Beroza, Gregory C. (Gregory Christian) |
| Thesis advisor | Zebker, Howard A |
| Advisor | Beroza, Gregory C. (Gregory Christian) |
| Advisor | Zebker, Howard A |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Daniel Kenneth Sinnett. |
|---|---|
| Note | Submitted to the Department of Geophysics. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2014. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2014 by Daniel Kenneth Sinnett
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...