Surface deformation and seismic signatures associated with the eruption cycle of Lone Star Geyser, Yellowstone National Park

Tuesday, 16 December 2014
Francisco G Gomez1, Harold E Johnson III1, Adam L LeWinter2, David C Finnegan3, Eric A Sandvol4, Avinash Nayak5 and Shaul Hurwitz6, (1)Univ Missouri, Columbia, MO, United States, (2)Univ Northern Colorado, Denver, CO, United States, (3)U.S. Army Cold Regions Research and Engineering Laboratory, Hanover, NH, United States, (4)Univ Missouri Columbia, Columbia, MO, United States, (5)Berkeley Seismological Lab, Berkeley, CA, United States, (6)USGS California Water Science Center Menlo Park, Menlo Park, CA, United States
Geysers are important subjects for studying processes involved with multi-phase eruptions. As part of a larger field effort, this study applies imaging geodesy and seismology to study eruptive cycles of the Lone Star Geyser in Yellowstone National Park. Lone Star Geyser is an ideal candidate for such study, as it erupts with a nearly regular period of approximately 3 hours. The geyser includes a 5 m diameter cone that rises 2 meters above the sinter terrace, and the entire system can be viewed from a nearby hillside. Fieldwork was accomplished during April 2014. Ground-based interferometric radar (GBIR) and terrestrial laser scanning (TLS) were used to image possible surface deformations associated with Lone Star Geyer’s eruption cycles. Additional observations were provided by global positioning system (GPS) measurements and six broad-band seismometers deployed in the immediate vicinity of the geyser. The GBIR and TLS were deployed approximately 65 meters from the sinter cone of the geyser. The GBIR involves a ku-band radar (1.7 cm wavelength) that is sensitive to approximately half-millimeter changes in the line-of-sight distance. Radar images were acquired every minute for 3 or more eruptions per day. Temporally redundant, overlapping interferograms were used to improve the sensitivity and interpolate a minute-wise time series of line-of-sight displacement, and efforts were made to account for possible path-delay effects resulting from water vapor around the geyser cone. Repeat (every minute) high-speed TLS scans were acquired for multiple eruption cycles over the course of two-days. Resulting measurement point spacing on the sinter cone was ~3cm. The TLS point-clouds were geo-referenced using static surveyed reflectors and scanner positions. In addition to measuring ground deformation, filtering and classification of the TLS point cloud was used to construct a mask that allows radar interferometry to exclude non-ground areas (vegetation, snow, sensors). Preliminary results suggest deformations are very small, with possible uplift around the sinter cone of up to 1 cm. Ongoing analysis is examining temporal variations in the seismological data that may correlate with apparent temporal and spatial patterns of surface displacement.