V43B-1423 1340h
Geotechnical Analysis of the Chaos Jumbles Rockfall Avalanches at Lassen Volcanic Park, California
The Chaos Jumbles rockfall avalanches are a series of three slope failures that occurred from the northwest side of the Chaos Crags approximately 300 years ago. Run out distances for the three rockfall avalanches range from 4.3km to 2.5km. Previous investigations have shown that the failure occurred from one of the six domes of the Chaos Crags. The cause of the failure has been investigated, but has not been determined (Crandell et al, 1974 and Eppler et al, 1987). This research uses geotechnical data obtained from field investigations, laboratory testing and computer modeling to determine the failure processes. Samples and discontinuity data were collected from the accessible areas along the base of the failure scarp. Samples included in situ rock discontinuities for shear strength analysis, rock cores for uniaxial and triaxial strength analyses and field measurement of discontinuity orientations. The data is used to model different failure scenarios along the existing scarp using limit equilibrium, finite difference and distinct element programs. The preliminary results of our analysis show that a set of discontinuities parallel to the failure surface along with there conjugate set perpendicular to the failure surface played a major role in the in the failure at the Chaos Jumbles rockfall avalanches. Research continues on modeling the failure and the run out distances of the avalanches.
V43B-1424 1340h
HSDP II Drill Core: Preliminary Rock Strength Results and Implications to Flank Stability, Mauna Kea Volcano
Selected portions of the 3-km HSDP II core were tested to provide unconfined rock strength data from hyaloclastite alteration zones and pillow lavas. Though the drilling project was not originally intended for strength purpose, it is believed the core can provide unique rock strength insights into the flank stability of the Hawaiian Islands. The testing showed that very weak rock exists in the hyaloclastite abundant zones in the lower 2-km of the core with strength dependent on the degree of consolidation and type of alteration. Walton and Schiffman identified three zones of alteration, an upper incipient alteration zone (1080-1335m), a smectitic zone (1405-1573m) and a lower palagonitic zone from about 1573 m to the base of the core. These three zones were sampled and tested together with pillow lava horizons for comparison. Traditional cylindrical core was not available as a consequence of the entire core having been split lengthwise for archival purposes. Hence, point load strength testing was utilized which provides the unconfined compressive strength on irregular shaped samples. The lowest unconfined strengths were recorded from incipient alteration zones with a mean value of 9.5 MPa. Smectitic alteration zones yielded mean values of 16.4 MPa, with the highest measured alteration strengths from the palagonite zones with a mean value of 32.1 MPa. As anticipated, the highest strengths were from essentially unaltered lavas with a mean value of 173 MPa. Strength variations of between one to two orders of magnitude were identified in comparing the submarine hyaloclastite with the intercalated submarine lavas. The weakest zones within the hyaloclastites may provide horizons for assisting flank collapse by serving as potential thrust zones and landslide surfaces.
V43B-1425 1340h
Slope Failure on Subglacial Volcanoes: Effects of Lithification, Fluid Pressure, and Seismic Loading on Edifice Stability
Large-scale paleolandslides are commonly observed on subglacial volcanoes. Rapid drainage of the surrounding englacial lake, removal of buttressing ice, and hydrothermal alteration are often cited as possible causes for slope failure on subglacial volcanoes. Limit-equilibrium slope stability analyses show that the failure of unlithified slopes during rapid drawdown of the englacial lake is the most likely mechanism for landsliding on a subglacial volcano in the Wells Gray-Clearwater Volcanic Field, British Columbia. Our modeling suggests that large-scale slope instability on subglacial volcanoes most likely requires slope materials that are not completely lithified along with rapid drainage of surrounding water. Pyramid Mountain is a 240 m high subglacial volcano in Wells Gray Provincial Park in central British Columbia, Canada. Geomechanical characterization of hyaloclastite outcrops and landslide mapping were performed in the field. The failure conditions of a prominent landslide on the east flank of the mountain were modeled in SLIDE. In the first scenario, the strength of the current, lithified rock mass was used to test slope stability. The second scenario evaluated the stability of cohesionless, unlithified deposits for a range of possible frictional strengths. For both scenarios, hydrologic conditions were varied iteratively from a fully saturated slope, simulating rapid and complete drawdown of surrounding water, to a dry slope. Results show that rapid drawdown of the englacial lake with simultaneous 0.4 g horizontal ground acceleration would be necessary to cause the rock slope in the first scenario to fail, an unlikely event. In contrast, for the second scenario, rapid drawdown of the lake alone causes the east flank to fail for friction angles ranging from 30-58 degrees. Rapid drawdown is necessary for failure of unlithified deposits below seismic loads of 0.2 g. Hence, edifice failure on subglacial volcanoes probably occurs early, as the englacial lake is draining and before the hyaloclastite has dewatered and lithified.
V43B-1426 1340h
Volcano Instability and Dike Swarms Controlled by Local Stress Field
Hotspot volcanoes in the Hawaiian, Canary, and the Reunion islands have two or three directions of dominant rift zones, which are highly developed. Rift zones in those islands are normally underlain by a sequence of basalt flow units, and consist of dike swarms which are elongated to a specific direction (with several kilometers length). In oceanic volcanic islands(Kilauea, Mauna Loa, Piton de la Fournaise) with rift zones, many large landslides have been generated during the past hundreds of years. Most of such landslides are presumably caused by lateral failures. Destabilization of volcanic edifices resulting from internal and external factors (such as local Stress field by dike intrusions, surface loading by eruption erosion, and seismic activities) should contribute to lateral fracturing and collapsing. In this study, the mechanisms of lateral collapse on the have been examined by numerical and analogue analyses of volcanic edifices with highly developed rift zones. We have conducted analogue experiments on the re-distribution of stress field with which associated the growth of a lift zone by using of the _eanalogue cone_f as a simulation of polygenetic volcano, In the analogue experiments, we will reveal the relationship between fracture distribution and failure (open cracks) developments. First, the gelatin cone was placed on a PVC plate with several holes for the injection of fluid and covered all over with the mixture of starch powder and pulverized soybean. The former is a simulate of thicker outer brittle part that consist of eruption fragments, and the latter is simulated for the underlying inner ductile part that consists of the relatively high temperature material near the magmatic dikes (or fissures). We made a series of experiments by changing of above indicated parameters. Second, the obtained results are compared with the varieties of developed rift zones. Finally, we show that distributions of open cracks produced by intruded dike swarms cause the lateral collapse, although their parameters are greatly affected by the thickness of cumulative fragmentations and inclination of edifice. For the volcanoes with non-dominant rift zones, the lateral collapse on the flank seems hardly generated due simply to the non-heterogeneous distribution of the local stress.
http://epp.eps.nagoya-u.ac.jp/~tacky/
V43B-1427 1340h
Limitations of Deterministic Modelling of Slope Stability on Volcanic Edifices
The conditions leading to the 18 May 1980 sector collapse of Mount St Helens have been the subject of a number of detailed investigations. Preservation of the initial failure plane(s) allowed Voight et al. (1983) and Donnadieu et al. (2001) to undertake back analyses and determine a range of possible failure conditions. While the models proposed offer major insights into potential failure mechanisms, we will demonstrate that deterministic analyses are of limited usefulness because many of the model parameters, such as cohesion, internal friction and pore pressure, are very poorly constrained. This creates problems of non-uniqueness in the solution. An alternative approach involves a series of Monte Carlo simulations to identify potential combinations of parameters that will produce the observed failure plane. Initial input ranges are specified for each parameter and the predetermined model is run repeatedly, with the parameter values for each model selected at random from within the input ranges. The interaction between parameters can be examined in detail, providing a better understanding of the potential failure conditions. This approach, which has been tested initially on a theoretical slope with predetermined failure conditions, highlights the fact that it is impossible to generate a unique model that fits the data when the slope has poorly defined strength parameters. This has clear implications for the validity of commonly used deterministic approaches. This probabilistic back analysis approach has been used to reanalyse the conditions that led to the May 18 collapse on Mount St Helens. Donnadieu, F., Merle, O., and Besson, J.C., 2001, Volcanic edifice stability during cryptodome intrusion, Bulletin of Volcanology, vol 63, p61-72. Voight, B., Janda, R.J., Glicken, H., and Douglass, P.M., 1983, Nature and Mechanics of the Mount St-Helens Rockslide-Avalanche of 18 May 1980, Geotechnique, vol 33, p243-273.
V43B-1428 1340h
Validation of TITAN2D flow model code for pyroclastic flows and debris avalanches at Soufri\`{e}re Hills Volcano, Montserrat, BWI
Soufri\`{e}re Hills Volcano (SHV), Montserrat, has experienced numerous episodes of dome collapses since 1996. They range from relatively small rockfalls to major dome collapses, several $>$10x10$^{6}$ m$^{3}$, and one $>$100x10$^{6}$ m$^{3}$ (Calder, Luckett, Sparks and Voight 2002; Voight et al. 2002). The hazard implications for such events are significant at both local and regional scales, and include pyroclastic surges, explosions, and tsunami. Problems arise in forecasting and hazards mitigation, particularly in zoning for populated areas. Determining the likely extent of flow deposits is important for hazard zonation. For this, detailed mapping (topography of source areas and paths, material properties, structure, track roughness and erosion) has an important role, giving clues on locations of future collapse and runout paths. Here we present an application of a numerical computation model of geophysical mass flow using the TITAN2D code (Patra et al. 2004; Pitman et al. 2004), to simulate dome collapses at SHV. The majority of collapse-type pyroclastic flows at SHV are consistent with an initiation by gravitational collapse of oversteepened flanks of the dome. If the gravity controls the energy for such processes, then the flow tracks can be predicted on the basis of topography, and friction influences runout. TITAN2D is written to simulate this type of volcanic flow, and the SHV database is used to validate the code and provide calibrated data on friction properties. The topographic DEM was successively updated by adding flow deposit thicknesses for previous collapses. Simulation results were compared to observed flow parameters, including flow path, deposit volume, duration, velocity, and runout distance of individual flows, providing calibration data on internal and bed friction, and demonstrating the validity and limitations of such modeling for practical volcanic hazard assessment.
V43B-1429 1340h
Granular Friction in Laboratory Shear-Zone Tests on Volcanic Sediments
We report on detailed laboratory experiments designed to elucidate the frictional behavior of volcanic materials, including pyroclastic flow debris from Soufriere hills volcano, Montserrat, and lahar deposits from Mt. St. Helens. Experiments were conducted in a servo-controlled, double-direct shear apparatus by shearing two 5-mm thick layers of loosely packed pyroclastic material between three roughened forcing blocks under conditions of monitored temperature and humidity. The central block is driven between the stationary side blocks at a precisely controlled displacement rate (typically 10 $\mu$m/s). We studied the effects of loading velocity, normal stress, grain size distribution, and water content. Normal stress was maintained constant during shear. A range of grain sizes and grain size distributions were examined, using material up to 1.0 mm in diameter. Median diameter (Md$\phi$) and the phi deviation measure ($\sigma$$\phi$) were varied, as well as several other distribution parameters. Experiments were conducted on pyroclastic material of fine (0.063-0.125 mm) and coarse (0.5-1.0 mm) grain sizes, as well as two broader grain size distributions (0.125-1.0 mm, and 0-1 mm). Using a normal stress range of 0.75 to 8 MPa we created a Coulomb-Mohr envelope and found that the coefficient of internal friction varies from 0.56 to 0.64 over this grain size range. The residual coefficient of sliding friction increases slightly, from 0.63 to 0.64, with increasing grain size and decreases from 0.61 to 0.56 with a widening of the grain size distribution. Smaller grain sizes and wider size distributions also exhibited higher apparent cohesion. For experiments on the natural grain size distribution (0-1 mm), we varied the shear velocity in the range 10 to 900 $\mu$m/s. These data indicate that the value of residual sliding friction increases with slip velocity, and thus exhibits velocity strengthening frictional behavior. A series of variable normal stress experiments were run at 100 $\mu$m/s so as to create a Coulomb-Mohr envelope that indicates the coefficient of internal friction increases from 0.56 at a shear velocity of 10 $\mu$m/s to 0.65 at 100 $\mu$m/s. Also the apparent cohesion is seen to fall from 78.6 kPa in the low velocity case to 23.6 kPa in the high velocity case. Ongoing experiments include slide-hold-slide tests to identify time-dependent frictional healing effects and saturated tests to investigate the effects of water. Our results indicate that the frictional properties of granular, volcanic sedminent are sensitive to shear velocity, normal stress, and variations in grain size distribution.
V43B-1430 1340h
Headless Debris Flows From Mount Spurr Volcano, Alaska
Sometime between June 20 and July 15, 2004-and contemporaneous with an increase of seismicity beneath the volcano, and elevated gas emissions-a sudden release of impounded water from the summit area of Mt. Spurr volcano produced about a dozen separate debris flow lobes emanating from crevasses and bergschrunds in the surface ice several hundred meters down the east-southeast flank from the summit. These debris flows were first observed by AVO staff on a July 15 overflight and appeared to represent a single flooding event; subsequent snow cover and limited accessibility have prevented direct investigation of these deposits. Observed from the air, they are dark, elongate lobate deposits, up to several hundred meters long and tens of meters wide, draping the steep (up to ~45 degree) slopes and cascading over and into crevasses. A water-rich phase from the flows continued down slope of the termini of several lobate deposits, eroding linear rills into the snow and ice down slope. We infer that the dark material composing these flows is likely remobilized coarse lapilli from the June 1992 tephra fall produced by an eruption of Crater Peak, a satellite vent of Mt. Spurr located 3.5 km to the south. Between 1 and 2 meters of basaltic andesite tephra fell directly on the Spurr summit during the 1992 eruption. The exact mechanism for sudden release of water-laden remobilized tephra flows from the summit basin is not clear. However, observations in early August, 2004, of an 80 m x 110-m-wide pit in the summit area snow and ice suggest the possibility of a partial roof collapse of a summit meltwater basin, likely associated with subglacial melting due to recent heat flux. Such a collapse could have led to the hydraulic surge of meltwater, and rapid mixing with tephra to produce slurries. These slurries traveled down slope beneath the ice surface to emerge through existing crevasses and other easy points of exit on the steep inclines. Mount Spurr is an ice- and snow covered, Quaternary andesitic volcanic complex, comprising a centrally located dome (or stratocone) in a breached, 5-km-wide, glacier-filled caldera that dissects ancestral Mt. Spurr volcano. The summit of Mt. Spurr is 130 km west of Anchorage, AK and reaches 3,374 m in elevation. The summit dome complex is topographically asymmetric, with a steeper southwest side and a more gradually sloping northeast flank To our knowledge, this is the first time such debris flows have been observed near the summit of Mt. Spurr. However, the existence of ponded water near the summit may not be unique to 2004. A review of historical photographs and descriptions of the Spurr summit area indicates a dynamic environment that responds to complex variations in snowfall accumulation, solar radiation, and geothermal heat flux. Other authors have noted variations in summit snow pack and the ephemeral appearance of a snow-filled depression and possibly a water-filled pit in 1964 aerial photographs of the summit. The formation of these debris flows near the summit of Mt. Spurr in conjunction with elevated seismicity below the summit and the development of a collapse pit in summit ice cap suggest that increasing geothermal heat flux, possibly in combination with above normal temperatures and long periods of clear, sunny weather in the region is responsible.