Crystals, Bubbles and Melt: Critical Conduit Processes Revealed by Numerical Models

Abstract

Understanding how magma moves within a conduit is an important question that is still poorly understood. In particular, estimation of the magma ascent rate is key for interpreting monitoring signals and therefore, predicting volcanic activity. This relies on understanding how strongly different magmatic processes occurring within the conduit control the ascent rate. These processes are controlled by changes in magmatic parameters such as the water content or temperature and understanding/ linking changes of such parameters to monitoring data is an essential step in the use of these data as a predictive tool. The results presented here are from a suite of conduit flow models based on Soufrière Hills Volcano, Montserrat, that assesses the influence of individual model parameters. By systematically changing these parameters, the results indicate that changes in conduit diameter and excess pressure in the magma chamber are amongst the dominant controlling variables. However, the single most important parameter controlling variations in the magma ascent rate is the volatile content. Therefore, understanding the processes controlling the volatile content within the conduit system and the outgassing of these volatiles is crucial to understanding and predicting potential unrest or eruption scenarios.

Counterfactual Analysis of Runaway Volcanic Explosions

“…Instead, we rely on an elementary first order volumetric modeling approach because our main purpose is to elucidate the counterfactual concept with a well recorded case history. Following Thomas et al (2018), we restrict our calculations to consider magma drawdown volumes just within a conduit of nominal length 5 km, i.e., to a point where the conduit may meet an upper reservoir. We place a narrow uncertainty on this value (5th percentile 4.8 km; 95th percentile 5.2 km) because here we are interested in the influence of conduit radius, as the main uncertain variable; testing reservoir depth as an uncertain variable would be straightforward with our numerical counterfactual model, described below.…”

“…We use the Soufrière Hills Volcano conduit geometry outlined by Thomas et al (2018) in their investigation of conduit property controls on seismicity as a function of depth and, in or analysis, adopt their point values for the conduit dimension parameters h n , r n , as central values for our uncertainty distributions. The latter parameters are shown on Figure 2 and comprise: r 1 = radius of open crater h 1 = depth of crater to top of upper conduit h 2 , r 2 = length and radius of upper conduit h 3 , r 3 = length and radius of conduit constriction (see Thomas et al, 2018) h 4 , r 2 = length and radius of lower conduit Additionally, for uncertainty quantification in our calculations, we introduce randomized variations ε hn , ε rn on these parameters (see Figure 2 and Supplementary Information for details).…”

“…FIGURE 2 | Schematic model conduit geometry for Soufriere Hills volcano (after Thomas et al, 2018; see text); conduit radii and lengths are denoted by r n and h n , with related uncertainties ε n . Dimensions z 4 and H refer to nodes on Figure 3 and represent BBN random variables: overall conduit length and depth to effective level of the magma reservoir, respectively.…”

“…At each Monte Carlo iteration, these radii and depth spans are converted jointly to conduit sector (and crater) volumes (row 5), summed in row 6 to enumerate total conduit volume (Vol_total) and the volume of the upper three depth sectors to the bottom of the constriction (Vol_123). The latter node was included to allow us to test the impact, if any, of the Thomas et al (2018) constriction in relation to the volume erupted; this is tested at the node Evac_constrict.…”

“…Uncertainty distribution for drawdown depth (a total of 50,000 Monte Carlo samples are distributed into the depth ranges shown on the x-axis). The top of the upper reservoir is assumed at an approximate depth of 5 km (following Thomas et al, 2018). see Ogburn and Calder (2017) for a detailed review of several accepted empirical, statistical or physical computer codes; where computational speed is critical, the emulation approach of Spiller et al (2014) could be helpful in the present context).…”