A comprehensive, multi-model exploration of flow scenarios for supporting probabilistic hazard assessment of pyroclastic density currents: application at Fuego volcano (Guatemala)
This study presents a comprehensive multi-model framework using TITAN2D and Volcflow to assess pyroclastic density current hazards at Fuego volcano, revealing that while topography primarily controls flow pathways, model-specific rheological assumptions significantly influence inundation footprints and mobility, thereby establishing a transferable approach for propagating source and structural uncertainties into probabilistic hazard assessments.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Volcanoes that never truly sleep pose a unique challenge to the people living in their shadow. Unlike a volcano that erupts once and then sleeps for centuries, these persistent giants produce frequent, smaller explosions that can suddenly turn into fast-moving rivers of hot gas, ash, and rock. Scientists call these rivers pyroclastic density currents. They are terrifyingly efficient at destroying everything in their path, often traveling dozens of kilometers from the summit. For communities living on the slopes of such volcanoes, knowing where these flows might go is a matter of life and death. However, predicting their path is incredibly difficult because the ground is uneven, the flows behave in complex ways, and the conditions that start them can vary wildly. To make safe decisions about evacuations or long-term settlement, scientists need more than a single guess; they need a clear picture of all the possible ways a flow could behave, including how far it might travel and where it might spill out of its natural channels.
A team of researchers recently tackled this problem at Fuego volcano in Guatemala, a steep, active mountain that has produced devastating flows in the past, including a catastrophic event in 2018 that claimed hundreds of lives. The team did not rely on just one computer program to predict the future. Instead, they ran thousands of simulations using two different, widely trusted computer models that calculate how these hot, heavy flows move over real terrain. They wanted to see if the models agreed on the big picture, even if they disagreed on the details. To do this, they created a massive library of scenarios. They simulated flows ranging from very small to enormous, lasting from seconds to many minutes, and they varied the internal "friction" of the flows to see how that changed their movement. They ran these simulations down six major ravines that cut through the volcano's slopes, generating a total of 3,600 different flow scenarios.
The results revealed a reassuring consistency in the broad strokes. Both computer models agreed on the outer limits of where the flows could possibly reach. More importantly, they both identified the exact same spots along the ravines where the ground changes shape in a way that forces the flow to spill out of its channel and spread across the surrounding land. This suggests that the shape of the landscape itself is the most powerful factor controlling where these dangerous currents go. If the ground has a sharp bend or a sudden flattening, the flow will likely spill over, regardless of which computer model is used. This gives emergency planners confidence that they can identify the most vulnerable areas based on the topography alone.
However, the two models told very different stories about what happens once the flow spills out. One model produced flows that were long and narrow, staying tightly packed within the river valleys for much of their journey. The other model produced flows that were shorter but much wider, spreading out like a thin sheet across the land as soon as they left the main channel. The difference comes down to how each model handles the resistance the ground exerts on the moving rock. One model assumes that as the flow gets thinner, it slows down much faster, keeping it confined. The other model assumes the flow keeps moving as long as the ground is steep enough, even if the layer of rock is very thin. This means that while both models agree on where a flow might spill, they disagree on how much land it will cover once it does.
The researchers also discovered that the size of the flow matters in a specific way. They found a transition point around a volume of 10 million cubic meters. Below this size, the flows tended to stay in the channels. Above this size, the flows in one of the models began to spread out rapidly, turning into thin sheets that could cover vast areas of land far from the main ravine. This finding helps explain why some eruptions cause localized damage while others blanket entire regions. Furthermore, the team used advanced statistical tools to map out exactly what combination of flow size, speed, and duration would be needed to reach specific villages located near the riverbanks. They found that the time it takes for the flow to start moving can change the outcome, even for locations ten kilometers away from the summit.
Ultimately, this work does not provide a single, perfect prediction of the next disaster. Instead, it provides a robust framework for understanding the range of possibilities. By running thousands of scenarios and comparing two different ways of thinking about the physics, the researchers have shown that while the details of a flow's behavior depend on the tools used to measure it, the fundamental risks are dictated by the shape of the land. This approach allows scientists to move beyond simple guesses and toward a more complete, probabilistic understanding of volcanic hazards. For the communities living on the slopes of Fuego, this means that future safety maps can be built on a foundation that acknowledges uncertainty rather than ignoring it, ensuring that decisions about where to live and how to evacuate are based on the full spectrum of what the mountain might do.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.