Radial spreading of viscous-gravity currents with solidifying crust

Abstract

We have investigated the effect of a solidifying crust on the dynamics and surface morphology of radial viscous-gravity currents. Liquid polyethylene glycol was admitted into the base of a tank filled with cold sucrose solution maintained at a temperature below the wax freezing point. As the radial current advanced away from the inlet, its surface solidified and deformed through a combination of folding and fracturing. For the warmest experiments, during which solidification did not occur, the radius of the current increased in proportion to the square root of time, as demonstrated both experimentally and theoretically for isothermal viscous fluids by Huppert (1982). When cooling was sufficiently rapid, solid crust formed and caused the spreading rate to decrease. A cooling model combining conduction in the wax with convection in the sucrose solution predicts the distance from the source at which the solid crust first appearedProgressively colder experiments revealed a sequence of surface morphologies which resembled features observed on cooling lava flows and lava lakes. Flows in which crust formed very slowly developed marginal levees which contained and channelled the main portion of the current. Colder flows with more rapid crust growth formed regularly spaced surface folds, multi-armed rift structures complete with shear offsets, and bulbous lobate forms similar to pillow lavas seen under the ocean. The same transitions between modes of surface deformation were also generated by keeping the ambient water temperature constant and decreasing the extrusion rate. This demonstration that surfaces can exhibit a well-defined sequence of morphologies which depend on the underlying flow conditions offers the prospect of more successful interpretation of natural lava flows.

Long submarine lava flows: Observations and results from numerical modeling

“…Therefore, on the basis of the rate of crust growth, we predict that higher strain rates (typically associated with greater flow velocities and effusion rates) are required to generate submarine jumbled flows than to disrupt the surface crust on a similar subaerial flow, and this may explain the relative scarcity of submarine jumbled flow when compared to the abundance of subaerial a'a flows. To constrain the rate of crust growth, we used the depth of the 1005 K isotherm within the flow, which is the glass transition temperature [Ryan and Sammis, 1981;Fink and Griffiths, 1990] for basalts. During the solidification of lava lakes in Hawaii, however, surface drills reportedly "gave way" when they encountered lava with temperatures near 1340 K [Peck and Kinoshita, 1976].…”

“…It is important to note that the drill bit was turning when they "gave way," and therefore this mechanical solidus is necessarily strain rate dependent. On the basis of the thick, glassy rinds observed on submarine basalts, we believe that the glass transition temperature (determined through a range of thermophysical experiments) is the more appropriate choice for the solidus temperature of submarine basalts [e.g., Fink and Griffiths, 1990;Griffiths and Fink, 1992;Gregg and Fink, 1995]. To facilitate comparisons between the submarine and subaerial basalt flows, we used the glass transition temperature for both cases.…”

“…[Fink and Griffiths, 1990;Griffiths and Fink, 1992]. If the cooling rate is high in comparison to the effusion rate, lava will travel only a short distance from the vent before solidifying, resulting in a pile of relatively short flows.…”

Morphology, eruption rates, and rheology of lava domes: Insights from laboratory models

“…As an alternative, we and others have turned to laboratory analogs to better parameterize the growth of lava domes. Previous laboratory-based models of lava flow emplacement have shown that both their morphology and dynamics are controlled by the ratio of eruption rate to cooling rate [e.g., Fink and Griffiths, 1990;Griffiths and Fink, 1993]. Here we will review the hypothesis that the same variables play a dominant role in the development of lava domes, which occupy the high-strength, high-viscosity, more silicic, and more crystal-rich end of the lava flow spectrum.…”

“…However, if approximate experimental crust thicknesses (relative to dome dimensions) in each morphological regime carry over to lava domes of the same type, the carapace strengths estimated above imply that volatile pressures will range from less than 106 Pa for axisym- Here we suggest two alternative hypotheses. The first is an expansion of our earlier argument that the magma lacked significant yield strength as it emerged from the vents, but that each pancake is better considered to be near the opposite or pillow end of the spectrum of morphologies seen in the viscous experiments, i.e., in the "levee" regime, where flows tend to remain symmetric for a time [e.g., Fink and Griffiths, 1990]. Our second and preferred hypothesis is that the erupted magma spread in a manner controlled by a strong carapace but with an eruption style influenced by an interior yield strength.…”