A comparison of cooling‐limited and volume‐limited flow systems: Examples from channels in the Piton de la Fournaise April 2007 lava‐flow field. (1st September 2017)

• Record Type: Journal Article
• Authors: Rhéty, Maéva; Harris, Andrew; Villeneuve, Nicolas; Gurioli, Lucia; Médard, Etienne; Chevrel, Oryaëlle; Bachélery, Patrick
• Is Part Of: Geochemistry, geophysics, geosystems. Volume 18:Number 9(2017)
• Page Start: 3270
• Page End: 3291
• Publication Date: 2017-09-01
• Subjects: lava channel -- dispersed flow -- cooling rate -- rheology -- cooling limit -- volume limit; Geochemistry -- Periodicals; Geophysics -- Periodicals; Earth sciences -- Periodicals; 550.5
• DOI: 10.1002/2017GC006839
• Languages: English
• ISSNs: 1525-2027

Abstract

Abstract: Channel‐fed lava‐flow systems lack detailed thermal and textural studies through the boundary between channelized and dispersed flow, and out to the flow front. Here chemical, textural, and morphological analyses were made to define cooling and crystallization rates down the entire system, especially through the zone of dispersed flow. We compare two channel systems active during the 2007 eruption of Piton de la Fournaise, one of which was cooling limited and one of which as volume limited. In the cooling‐limited case, rapid changes in rheology occurred across the zone of dispersed flow, where viscosity increased from 1000 to 1600 Pa s over the last 100 m of the channel system. This was due to an increase in cooling rate from 7°C km −1 over the first 500 m of the system, to 42°C km −1 over the last 100 m, and an increase in microcryst content from 13% to 25%. In the volume‐limited case, the exponentially increasing segment of the down‐flow cooling and viscosity trend is absent. Instead, lava arriving at the flow front is still relatively hot (1161°C compared with a near‐vent temperature of 1167°C) and is thus of relatively low viscosity (1125 Pa s). In the volume‐limited case, because the channel was still in extension when supply to the system was cut, the zone of dispersed flow was extremely short. However, because lava behind the stalled flow front was still hot and fluid, breakouts from the static front resulted in a complex flow front morphology. Plain Language Summary: When lava flows stop they either do so because they have cooled to such an extent that they, essentially, freeze and thus can no longer move, or because the supply of lava is cut; effectively the tap is turned off. We term the former "cooling‐limited" flows, and the latter "volume limited." We show here the differences in terms of rates of down‐flow cooling and crystallization between the two flow types and point to the differences in appearance of the two. For the cooling‐limited case, the channel feeds a zone of lava flow with well‐formed, frozen flow fronts. For the volume‐limited case, the channel extends all the way to the flow front, which is—itself—a mess because of fluid lava inside the stalled flow that then escapes through the stationary front. We show this through a study of lava flows emplaced during the largest historical eruption of Piton de la Fournaise, on the French island of Réunion; this being the eruption of April 2007. Key Points: We define the morphological differences between volume‐limited and cooling‐limited lava flow We define down‐flow cooling and crystallization in a cooling‐limited and a volume‐limited channel We propose a data‐driven model for emplacement of a cooling‐limited flow field fed by multiple short‐lived effusive events