The influence of topographic roughness on lava flow emplacement

Abstract

A quantitative understanding of the factors controlling lava flow emplacement is critical for both hazard assessment and mitigation and for the interpretation of past flow emplacement conditions. The influence of topography with a vertical amplitude smaller than flow thickness (i.e., substrate roughness) is currently not accounted for in most flow emplacement models and hazard estimates. Here, we measure the effect of substrate roughness on flow emplacement through experiments using analog fluids and molten basalt, complementing recent work on the interaction of lava flows with obstacles taller than flow thickness. We present results from three sets of analog experiments, in which corn syrup, polyethylene glycol, and molten basalt were each extruded onto a sloping plane covered with a series of beds of varying grain sizes. We find that flow front advance rates are impacted by bed roughness for all materials, with decreases in average velocities by up to 50% with increases of substrate grain sizes by 5–100 times, ranges analogous with topographic variations found in nature. These decreases in flow front advance velocities are equivalent to up to an order of magnitude increase in fluid viscosity. We interpret this velocity decrease to be caused by the movement of material into void spaces between substrate grains and by enhanced cooling through heat conduction to the substrate due to increased surface contact area. The difference in advance velocity with increasing grain size diminishes with time after initial emplacement as a basal boundary layer is established. Additionally, the experimental flow geometry, measured by the complexity of the flow external perimeter, became increasingly complex with increasing substrate grain size. This effect will act to both slow the forward advance of lava flows and to create irregular emplacement paths of flows moving over rough surfaces. We propose that flow emplacement models should be modified, possibly through a calibrated “effective viscosity” term, to account for bed roughness to increase accuracy in flow prediction and hazard estimation models.