Geology Reports⌕ Search

Geology topics

M.M. Morrissey

Publications and source records attributed to M.M. Morrissey.

8 recordsLinked to original sources
USGS2007

A comparative analysis of simulated and observed landslide locations triggered by Hurricane Camille in Nelson County, Virginia

In 1969, Nelson County, Virginia received up to 71 cm of rain within 12 h starting at 7 p.m. on August 19. The total rainfall from the storm exceeded the 1000-year return period in the region. Several thousands of landslides were induced by rainfall associated with Hurricane Camille causing fatalities and destroying infrastructure. We apply a distributed transient response model for regional slope stability analysis to shallow landslides. Initiation points of over 3000 debris flows and effects of flooding from this storm are applied to the model. Geotechnical data used in the calculations are published data from samples of colluvium. Results from these calculations are compared with field observations such as landslide trigger location and timing of debris flows to assess how well the model predicts the spatial and temporal distribution. of landslide initiation locations. The model predicts many of the initiation locations in areas where debris flows are observed. Copyright ?? 2007 John Wiley & Sons, Ltd.

Hydrological Processes↗
USGS2004

Transient hazard model using radar data for predicting debris flows in Madison County, Virginia

During the rainstorm of June 27, 1995, roughly 330-750 mm of rain fell within a 16-hour period, initiating floods and over 600 debris flows in a small area (130 km2) of Madison County, VA. We developed a distributed version of Iverson's transient response model for regional slope stability analysis for the Madison County debris flows. This version of the model evaluates pore-pressure head response and factor of safety on a regional scale in areas prone to rainfall-induced shallow (<2-3 m) landslides. These calculations used soil properties of shear strength and hydraulic conductivity from laboratory measurements of soil samples collected from field sites where debris flows initiated. Rainfall data collected by radar every 6 minutes provided a basis for calculating the temporal variation of slope stability during the storm. The results demonstrate that the spatial and temporal variation of the factor of safety correlates with the movement of the storm cell. When the rainstorm was treated as two separate rainfall events and a larger hydraulic conductivity and friction angle than the laboratory values were used, the timing and location of landslides predicted by the model were in closer agreement with eyewitness observations of debris flows. Application of spatially variable initial pre-storm water table depth and soil properties may improve both the spatial and temporal prediction of instability.

Environmental & Engineering Geoscience↗
USGS2001

Trends in long-period seismicity related to magmatic fluid compositions

Sound speeds and densities are calculated for three different types of fluids: gas-gas mixture; ash-gas mixture; and bubbly liquid. These fluid properties are used to calculate the impedance contrast (Z) and crack stiffness (C) in the fluid-driven crack model (Chouet: J. Geophys. Res., 91 (1986) 13,967; 101 (1988) 4375; A seismic model for the source of long-period events and harmonic tremor. In: Gasparini, P., Scarpa, R., Aki, K. (Eds.), Volcanic Seismology, IAVCEI Proceedings in Volcanology, Springer, Berlin, 3133). The fluid-driven crack model describes the far-field spectra of long-period (LP) events as modes of resonance of the crack. Results from our calculations demonstrate that ash-laden gas mixtures have fluid to solid density ratios comparable to, and fluid to solid velocity ratios lower than bubbly liquids (gas-volume fractions <10%). This difference results in synthetic far-field spectra with higher impedance contrasts and narrower spectral bandwidths for ash-laden gas mixture than spectra for bubbly liquids. Spectral characteristics are described in terms of the quality factor Q-1. Q-1 is measured by the ratio of the frequency of the dominant spectral peak to the bandwidth of the peak measured at one half of its amplitude. This factor expresses the losses of energy due to elastic radiation Q-1r and other dissipative mechanisms Q-1i at the source, Q-1 = Q-1r + Q-1i. Spectra for LP events recorded at active volcanoes such as Galeras in Colombia and Kilauea in Hawaii, have Q-1 factors in the range of 0.1-0.002. The Q-1r factors due to radiation loss calculated for a sphere filled with a H2O-CO2 or H2O-SO2 gas mixture, vary between 0.0015 and 0.0040 with a change in wt% H2O at 800-1600 K and 10-50 MPa. For gas-rich mixtures, Q-1r has a strong dependence on resonator geometry (spherical versus rectangular). The spectra from a resonating sphere filled with gas-rich mixture yields values of Q-1r an order of magnitude smaller than those from a rectangular crack. For a resonating crack filled with an ash-gas mixture (or pseudogas), Q-1r varies parabolically from ???0.006 for an ash-rich mixture, to 0.0015 or 0.0023 for a H2O-rich or CO2-rich mixture at 800 K and 25 MPa. For low (<20%) gas-volume fraction fluids (foams, bubbly fluids and ash-rich pseudogases), the magnitudes for Q-1r are independent of crack geometry. Spectra associated with a foam (gas-volume fractions 10-90%) or bubbly basalt (gas-volume fractions <10%) may have a dominant spectral peak with values of Q-1r on the order of 0.01 and 0.1, respectively. The spectra from a resonating sphere filled with a foam containing >20% gas-volume fraction yields values of Q-1r similar to those for a rectangular crack. As with gas-gas and ash-gas mixtures, an increase in mass fraction narrows the bandwidth of the dominant mode and shifts the spectra to lower frequencies. Including energy losses due to dissipative processes in a bubbly liquid increases attenuation. Attenuation may also be higher in ash-gas mixtures and foams if the effects of momentum and mass transfer between the phases were considered in the calculations. ?? 2001 Elsevier Science B. V. All rights reserved.

Journal of Volcanology and Geothermal Research↗
USGS1999

Air blasts generated by rockfall impacts: Analysis of the 1996 Happy Isles event in Yosemite National Park

The July 10, 1996, Happy Isles rockfall in Yosemite National Park, California, released 23,000 to 38,000 m 3 of granite in four separate events. The impacts of the first two events which involved a 550-m free fall, generated seismic waves and atmospheric pressure waves (air blasts). We focus on the dynamic behavior of the second air blast that downed over 1000 trees, destroyed a bridge, demolished a snack bar, and caused one fatality and several injuries. Calculated velocities for the air blast from a two-phase, finite difference model are compared to velocities estimated from tree damage. From tornadic studies of tree damage, the air blast is estimated to have traveled <108–120 m/s within 50 m from the impact and decreased to <10–20 m/s within 500 m from the impact. The numerical model simulates the two-dimensional propagation of an air blast through a dusty atmosphere with initial conditions defined by the impact velocity and pressure. The impact velocity (105–107 m/s) is estimated from the Colorado Rockfall Simulation Program that simulates rockfall trajectories. The impact pressure (0.5 MPa) is constrained by the kinetic energy of the impact (10 10 –10 12 J) estimated from the seismic energy generated by the impact. Results from the air blast simulations indicate that the second Happy Isles air blast (weak shock wave) traveled with an initial velocity above the local sound speed. The size and location of the first impact are thought to have injected <50 wt% dust into the atmosphere. This amount of dust lowered the local atmospheric sound speed to ∼220 m/s. The discrepancy between calculated velocity data and field estimated velocity data (∼220 m/s versus ∼110 m/s) is attributed to energy dissipated by the downing of trees and additional entrainment of debris into the atmosphere not included in the calculations.

Journal of Geophysical Research B: Solid Earth↗
USGS1998

Rock-fall hazards in the Yosemite Valley

This report with map shows the areas of highest rock-fall hazard in four selected parts of Yosemite Valley, California, defined by the National Park Service. Two specific levels of hazard are recognized and identified from rock falls ranging in size from individual boulders to moderate-sized events with volumes of 100,000 m3. These different levels of hazard are associated with areas within the base of the talus (red line in Plate 1) and those normally beyond the base of the talus which we refer to as the rock-fall shadow (yellow line in Plate 1). Rock falls of even greater size, exceeding 100,000 m3, referred to as rock avalanches, are a potential hazard, but the extent of this hazard is not easily defined. Rock avalanches are considered to be much less likely to occur because of the relatively few prehistoric rock-fall avalanche deposits that have been recognized in the Yosemite Valley. With the configuration of the steep valley walls and the relatively narrow valley, it should be noted that there are no absolutely safe or zero probability areas from large rock avalanches. This study has shown in map form where rock-fall hazard exists and has given general indication of the expected frequency of these events; however, the study does not quantify the probability at any specific location, nor evaluate the risk to people or facilities to such events.

Open-File Report↗
USGS1997

A numerical investigation of choked flow dynamics and its application to the triggering mechanism of long-period events at Redoubt Volcano, Alaska

We use numerical simulations of transonic flow through a crack to study the dynamics of the formation of shock waves downstream from a nozzle-like constriction inside the crack. The model solves the full set of Navier-Stokes equations in two dimensions via an explicit multifield finite difference representation. The crack walls are assumed to be perfectly rigid, and elastic coupling to the solid is not considered. The simulations demonstrate how the behavior of unsteady shock waves near the walls can produce recurring step-like pressure transients in the flow, which in turn induce resonance of the fluid-filled crack. The motion of the shock waves is governed primarily by smooth, low-amplitude pressure fluctuations at the outlet of the crack. The force induced on the walls scales with the amplitude of the shock, which is a function of the magnitude of the inlet pressure, aperture of the constriction, and thickness of the boundary layer. The applied force also scales in proportion to the spatial extent of the shock excursion, which depends on the fluctuation rate of outlet pressure. Using the source parameters of long-period (LP) events at Redoubt Volcano, Alaska, as a guide for our simulations, we infer that coupling of the shock to the walls occurs for crack inlet to outlet pressure ratios p i /p o >2.31 and that the position of the shock front becomes most sensitive to outlet pressure fluctuations for flow regimes with p i /p o >2.48. For such regimes, fluctuations of outlet pressure of up to ±0.5 MPa at rates up to 3 MPa/s are sufficient to induce pressure transients with magnitudes up to 12.5 MPa over 0.1–2.5 m of the walls within ∼0.5 s. These flow parameters may be adequate for triggering the LP events in the precursory swarm to the December 14, 1989, eruption of Redoubt. According to the flow model the recurrence rate and amplitudes of L.P events are inferred to be a manifestation of the response of a shallow hydrothermal reservoir to the sustained injection of superheated steam from a magma column roofing below this reservoir.

Journal of Geophysical Research B: Solid Earth↗
USGS1997

Long-period seismicity at Redoubt Volcano, Alaska, 1989-1990 related to magma degassing

The mass of exsolved magmatic H2O is estimated and compared to the mass of superheated steam (25-50 Mtons) released through the resonating crack producing the December 13-14, 1989 swarm of long-period seismic events at Redoubt Volcano. Results indicate degassing of a H2O-CO2-SO2-saturated magma upon ascending from at least 12 km to 3-4 km beneath the crater as the source of the superheated steam. The mass of exsolved H2O (3.2-250 Mtons) is estimated from solubility diagrams of H2O-CO2-saturated silicate melts for the ascent history of the Redoubt magmas. Crystal size distribution, seismological, petrological, and geochemical data are used to constrain the ascent history of the two andesitic magmas prior to the eruption. Two stages of crystallization are inferred from crystal size distributions of plagioclase crystals in andesites erupted in December 1989. The first stage occurred 30-150 years before the eruption in both magmas and the second stage occurred at least 8 years and 15 years before the eruption in the dacitic andesite and rhyolitic andesite, respectively. The depths of crystallization are constrained from the spatial and temporal variations of volcano-tectonic earthquakes locations (Lahr et al., 1994) and from the P-wave and S-wave velocity structures (Benz et al., 1996). These data suggest that the rhyolitic andesite magma ascended to a depth of 7-8 km within at least 15 years of the eruption. Within at least 8 years of the eruption, the dacitic andesite magma migrated to a depth just below the other magma body where it resided until hours to days of the eruption. At this time, the dacitic andesite magma mixed with the rhyolitic andesite magma and established the reservoir for the eruption. Near the top of the reservoir, some of the mixed magma was displaced into fractures which extended 4-5 km toward the surface. This displaced magma created the eruption conduit and released the fluids related to the resonating crack. This scenario is consistent with the trends in major-and trace-element chemistry, and the stability of hornblende in the pre-eruption Redoubt magmas. It also provides a source for the SO2 and CO2 emissions measured during the eruption.

Journal of Volcanology and Geothermal Research↗