Why study explosive eruptions?

They affect global climate and are hazardous to society

How do explosive eruptions affect climate years after an eruption (Timmreck 2018a)? What is the cumulative effect of eruptions on climate over decades, centuries and millennia (Schmidt et al. 2018)? How will these effects evolve as climate change progresses over the next century (Aubry et al. 2021)? The answers to these questions are crucial for forecasting the effects of climate change, yet uncertainties in how historic and ancient eruptions affected climate remain (Mann et al. 2021). The speed, mass and temperature of volcanic rocks, ash, pumice and gases erupting through a vent of a given or changing geometry during an eruption, known as "eruption source parameters", control the altitude at which volcanic ash-gas clouds spread in the atmosphere (Woods 2010). Whether volcanic ash-gas clouds spread in the troposphere or stratosphere determines an eruption's effect on climate. Volcanic sulfur delivery to the stratosphere can induce a global cooling effect (Fig. 5; Timmreck et al. 2018b) and, in contrast, volcanic halogen (Staunton-Sykes et al. 2021) and water delivery (Joshi and Jones 2009) can induce a global warming effect and the net effect of these three processes remains unclear. The volcano-climate community has demonstrated that uncertainties in the eruption source parameters of historic eruptions are a major obstacle to improved forecasting of volcano-climate effects (Marshall et al. 2021). In addition to climate effects, volcanic ash cloud spreading in the troposphere cause plane engines to fail and ashfall on local communities can damage infrastructure and cause respiratory issues (Fig. 5). Similar to forecasting volcano-climate effects, forecasting volcanic ash cloud hazards accurately is challenging due to the large uncertainties associated with eruption source parameters (Dioguardi et al. 2020).

A key process: Mass partitioning and transport in eruption columns and clouds

During an explosive eruption the erupted mixture of rocks, pumice ash and gas exits the volcanic vent as a high-speed upward flow that is more dense than the surrounding atmosphere (Fig. 6; Woods 2010). Despite the weight of the mixture, its initial momentum carries it upward and the turbulent mixture draws in air (entrainment; Fig. 6d; Morton et al. 1956) that reduces its overall density and can cause it to become buoyant. If this occurs before it loses its upwards momentum, the mixture undergoes a buoyancy reversal and continues to rise driven by its buoyancy to form an eruption column in the Buoyant Plume regime where erupted mass is partitioned into spreading umbrella ash clouds (Fig. 6a & 6d). If the mixture does not entrain sufficient air to reverse its buoyancy, it forms an eruption column in the Total Collapse regime where erupted mass is partitioned to spreading pyroclastic density currents (Fig. 6c & 6f). For many decades, large explosive eruptions have been classified into one of these two regimes, however, recent evidence suggests that most eruptions partition mass to both umbrella clouds and PDCs simultaneously to form an eruption column in the Partial Collapse regime (Fig. 6b & 6e; Neri and Dobran 1994; Kaminski and Jaupart 2001; Rosi et al. 2001; Di Muro et al. 2004; Castruccio et al. 2016). The range of eruption source parameters that define the Partial Collapse regime and the dynamics of entrainment, eruption column rise height, ash cloud spreading, sedimentation, and pyroclastic density currents are poorly understood and represent key knowledge gaps for improving our understanding of large explosive eruptions. This knowledge gap was the main focus of my PhD work at the University of British Columbia (Gilchrist and Jellinek 2021).

A challenge put forth by the volcanology community: A new classification for eruptions

The volcanology community has decided that current eruption classifications are limited in their ability to (Bonadonna et al. 2016; Manga et al. 2017):

1. Consistently classify all styles of explosive eruptions ranging from small puffs of ash to the largest catastrophic caldera-forming eruptions

2. Identify features of eruption deposits that are diagnostic of the eruption style and, in turn, quantitatively constrain eruption source parameters

3. Distinguish eruption styles with relatively constant or time-varying source parameters during the eruption

4. Provide quantitative constraints on the mass of erupted mixtures delivered to spreading ash clouds in the atmosphere and pyroclastic density currents during an eruption

5. Communicate the wide diversity of eruption styles and associated hazards to the public

Accordingly, the community has called for the development of new eruption classifications that can achieve these goals.

My research on mass partitioning in large eruption columns has built a foundation for establishing a new eruption classification scheme where eruption styles are classified on the basis of their strength (initial upward momentum) and the concentration of rocks, pumice and ash in the erupted mixture (particle volume fraction; Fig. 7). However, this foundation is missing a keystone required to classify all explosive eruption styles: a metric for distinguishing eruptions with relatively constant (steady) versus time-varying (unsteady) source parameters. The large eruption styles I have studied are relatively easy to model with analog experiments and computer simulations because their source parameters can be considered steady for the main phase of the eruption. Most eruptions occurring daily on Earth are much smaller and have unsteady source parameters that can change before the erupted mixture reaches its spreading height in the atmosphere or before it collapses back to Earth. The highly time-dependent nature of their source parameters makes them difficult to model with analog experiments and computer simulations. Thus, it is unclear whether these small unsteady eruptions occur in the same regimes and exhibit the same behavior as better understood large eruptions for similar ranges of average eruption source parameter values.

Despite their small size, unsteady eruptions still contribute to Earth's climate and present hazards to aviation and surrounding communities, thus our inability to predict their behavior and associated hazards represents a major knowledge gap in volcanology. I began addressing this knowledge gap during my PhD by discovering a new eruption source parameter metric, the source Pulsation number, that can predict the effects of steady and unsteady eruption source parameters on entrainment and, in turn, mass partitioning between spreading ash clouds and pyroclastic density currents. Currently, I'm working on using this Pulsation number metric with the eruption strength and particle volume fraction metrics, which I used to classify large eruptions, to put forth a new eruption classification scheme that achieves the goals outlined by the volcanology community.