Christopher R. J. Kilburn

Forecasting giant, catastrophic slope collapse: lessons from Vajont, Northern Italy

Rapid, giant landslides, or sturzstroms, are among the most powerful natural hazards on Earth. They have minimum volumes of similar to10(6)-10(7) m(3) and, normally preceded by prolonged intervals of accelerating creep, are produced by catastrophic and deep-seated slope collapse (loads similar to1-10 MPa). Conventional analyses attribute rapid collapse to unusual mechanisms, such as the vaporization of ground water during sliding. Here, catastrophic collapse is related to self-accelerating rock fracture, common in crustal rocks at loads similar to1-10 MPa and readily catalysed by circulating fluids. Fracturing produces an abrupt drop in resisting stress. Measured stress drops in crustal rock account for minimum sturzstrom volumes and rapid collapse accelerations. Fracturing also provides a physical basis for quantitatively forecasting catastrophic slope failure

Multiscale fracturing as a key to forecasting volcanic eruptions

Abstract Among volcanoes reawakening after long repose intervals, the final approach to eruption (∼1–10 days) is usually characterised by accelerating rates of seismicity. The observed patterns are consistent with the slow extension of faults, which continue to grow until they connect a pre-existing array of subvertical fractures and so open a new pathway for magma to reach the surface. Rates of slow extension are here investigated assuming that gravitational loading and magma overpressure create a fluctuating stress field in the country rock. Fluctuations are due to the intermittent growth of small cracks that cannot be detected by monitoring instruments. The rate of detected events is then determined by the frequency with which the concentration of strain energy around the tips of macroscopic faults becomes large enough to permit fault extension. The model anticipates oscillations in seismic event rate about a mean trend that accelerates with time, and identifies the rate of increase in peak event rate as a key indicator of the approach to eruption. The results are consistent with earlier empirical analyses of seismic precursors and, applied to data before the andesitic eruptions of Mt Pinatubo, Philippines, in 1991, and of Soufriere Hills, Montserrat, in 1995, they suggest that, for such volcanoes, reliable forecasts of magmatic activity might eventually be feasible on the order of days before eruption.

The evolution of lava flow-fields: observations of the 1981 and 1983 eruptions of Mount Etna, Sicily

The eruptions of Mount Etna in 1981 on the north flank and 1983 on the south flank of the volcano were of strikingly different character. The former was a short duration, high effusion rate eruption producing for the most part a simple flow-field; the latter was of relatively long duration and low effusion rate, producing a compound flow-field of overlapping flows.Despite the differences between the eruptive behaviour of these two events and the way in which the flow-field developed, both the flow-fields achieved about the same maximum length. This is considered fortuitous. The evidence suggests that the main 1981 flow stopped because the lava supply ceased and was thus volume controlled. The 1983 flow-field had a more complex history of branching, but in this case it appears that, for the longest individual flow, cooling played an important role in controlling the maximum extent of the flows.

Slow rock fracture as eruption precursor at Soufriere Hills volcano, Montserrat

Breakout of magmatic activity at Soufriere Hills volcano, Montserrat, was preceded by a tenfold increase in rate of earthquake occurrence. A new model of subcritical rock failure shows that this increase is consistent with the growth, possibly episodic, of the magma conduit at a rate controlled by progressive weakening of the host country rock. The preferred weakening mechanism is stress corrosion, by which circulating juvenile and hydrothermal fluids chemically attack the country rock and promote failure at stresses smaller than the rocks theoretical strength. The results illuminate the potential for slow-cracking models to enhance eruption forecasts using the inverse-rate technique combined with traditional monitoring methods.

General patterns of flow field growth: Aa and blocky lavas

Lava flow fields consist of one or more flows. Four ideal emplacement regimes are recognized: (a) that for single flows and (b) that for flow fields dominated by (1) widening, (2) thickening, or (3) lengthening, as a result of generating new flows. Most aa and blocky lavas belong to the flow field widening or single-flow regimes. These two regimes are analyzed assuming advance is controlled by the distal core of a flow, where motion is treated as steady, uniform, and laminar. Because of low deformation rates, the distal core is also approximated to a Newtonian fluid. Widening and, possibly, lengthening are ultimately limited by crustal resistance. After a critical cooling interval, new flows are generated from the upper reaches of the flow field. A simple relation is derived linking flow field dimensions and underlying slope to eruption duration, independent of terms involving gravity or lava chemistry and rheology. The relation well describes field data from several volcanoes (involving lava compositions from K phonolitic tephrite to dacite). This supports the premise that the overall growth of aa and blocky flow fields is systematic and also suggests that such growth may be predictable at time scales greater than, or similar to, the emplacement times of major flows.

Pahoehoe and aa lavas: a discussion and continuation of the model of Peterson and Tilling

Abstract The graphical method of Peterson and Tilling for representing the behaviour of a lava flow is considered in terms of simple rheological theory, assuming that a lava may be approximately described as a Bingham fluid. This approach helps resolve an apparent anomaly in the original treatment of Peterson and Tilling with respect to the representation of the pahoehoe to aa transition. It also suggests further methods for quantitatively defining the limits of the pahoehoe and aa fields, and the transition threshold zone.

Runout lengths of sturzstroms: The control of initial conditions and of fragment dynamics

Sturzstroms are giant landslides that travel kilometers within minutes. Data from terrestrial examples of 0.001–10 km3 suggest that their runout lengths increase in proportion to the square root of their volume. This trend is investigated assuming that runout is controlled by fragmental flow. The results indicate that runout lengths depend on the potential energy available for motion after initial collapse (itself a function of the dimensions of the collapse zone), on the degree of rock fragmentation, and on the rate of momentum loss by fragment collisions in a basal boundary layer, assumed to thicken with time by diffusion. A dependence on initial conditions during collapse would explain claims that a minimum volume is required for sturzstroms to form: Beneath a critical volume, insufficient energy is available to initiate fragmental flow, and so the unstable mass slumps downslope. The drop height (H) is often similar to the vertical extent of the collapse zone and so is linked by geometry to sturzstrom volume. The ratio of H to L (the horizontal distance of effective transport), normally interpreted as a measure of frictional resistance, is thus reinterpreted as an inverse measure of the energy available for runout after collapse. By providing a physical basis for observed trends, the analysis justifies use of empirical limits for forecasting the runout lengths of major landslides.

Fracturing as a quantitative indicator of lava flow dynamics

The traditional classification of lava flows into pahoehoe, aa and blocky varieties reflects differences in how a flow can fracture its surface during advance. Pahoehoe and aa lavas have a low strength upon eruption and require surface cooling to produce a crust that can fracture. Among pahoehoe lavas, applied stresses are small enough to allow the growth of a continuous crust, which is broken intermittently as the flow advances by propagating a collection of lava tongues. Among aa lavas, in contrast, applied stresses are large enough to maintain persistent crustal failure. Blocky lavas are distinguished by having a significant strength upon eruption, so that motion can induce surface fracturing even without cooling. The transition from pahoehoe to aa thus coincides with a change from intermittent to persistent failure of chilling crust, whereas the transition from aa to blocky lava occurs when lava beneath the crust can also autobrecciate. These fracturing characteristics have been used to quantify the transitions between flow regimes and suggest that shear fracture may dominate tensile failure. They also (1) explain the restricted variations with slope of pahoehoe and aa flow thickness, as well as the maximum advance rate for which pahoehoe can form, equivalent to the minimum rate required for aa lava, and (2) provide a revised method for forecasting the maximum potential lengths of aa flows. Applications include improved hazard assessments during effusive eruptions and new evaluations of the emplacement conditions for very large-volume pahoehoe lava flows.

Sampling and major element chemistry of the recent (A.D. 1631–1944) Vesuvius activity

Abstract Detailed sampling of the Vesuvius lavas erupted in the period A.D. 1631–1944 provides a suite of samples for comprehensive chemical analyses and related studies. Major elements (Si, Ti, Al, Fetotal, Mn, Mg, Ca, Na, K and P), volatile species (Cl, F, S, H2O+, H2O− and CO2), and ferrous iron (Fe2+) were determined for one hundred and forty-nine lavas and five tephra from the A.D. 1631–1944 Vesuvius activity. The lavas represent a relatively homogeneous suite with respect to SiO2, TiO2, FeOtotal, MnO and P2O5, but show systematic variations among MgO, K2O, Na2O, Al2O3 and CaO. The average SiO2 content is 48.0 wt.% and the rocks are classified as tephriphonolites according to their content of alkalis. All of the lavas are silica-undersaturated and are nepheline, leucite, and olivine normative. There is no systematic variation in major-element composition with time, over the period A.D. 1631–1944. The inter-eruption and intra-eruption compositional differences are the same magnitude. The lavas are highly porphyritic with clinopyroxene and leucite as the major phases. Fractionation effects are not reflected in the silica content of the lavas. The variability of MgO, K2O, Na2O, and CaO can be modelled as a relative depletion or accumulation of clinopyroxene.

Surfaces of Aa Flow-Fields on Mount Etna, Sicily: Morphology, Rheology, Crystallization and Scaling Phenomena

Most of Etna’s historical lavas have produced aa flow-fields. Despite the classification, they support a variety of surfaces in both the pahoehoe and aa categories. From observations of surface features at metre to submillimetre scales, two morphological series have been recognized: a sequence from pahoehoe to aa along a single flow, and an evolutionary trend among pahoehoe surfaces near boccas down a flow-field. The first series is more prominent and is superimposed on the second. Both series are associated with increasing crystallinity. The pahoehoe-aa sequence, however, is characterized by relatively higher crystallization rates and shear rates, indicating greater undercoolings and imposed stresses during emplacement. The identification of crystallization rate as a controlling factor demonstrates the influence of crystallization kinetics on lava development. The pahoehoe-aa transition also shows evidence of self-similar patterns at scales from at least millimetres to tens of centimetres, and may be accompanied by a self-feeding mechanism which couples large- and small-scale changes in lava rheology and shear rate.