Search

Rapid and simple estimation of the mass eruption rate (MER) from column height is essential for real-time volcanic hazard management and reconstruction of past explosive eruptions. Using 134 eruptive events from the new Independent Volcanic Eruption Source Parameter Archive (IVESPA, v1.0), we explore empirical MER-height relationships for four measures of column height: spreading level, sulfur dioxide height, and top height from direct observations and as reconstructed from deposits. These relationships show significant differences and highlight limitations of empirical models currently used in operational and research applications. The roles of atmospheric stratification, wind, and humidity remain challenging to detect across the wide range of eruptive conditions spanned in IVESPA, ultimately resulting in empirical relationships outperforming analytical models that account for atmospheric conditions. This finding highlights challenges in constraining the MER-height relation using heterogeneous observations and empirical models, which reinforces the need for improved eruption source parameter data sets and physics-based models.

Volcano Observatories (VOs) around the world are required to maintain the surveillance of their volcanoes and inform timely civil protection and aviation authorities about impending eruptions. Often, they work through procedures which allow them to respond timely to volcanic crises and provide a service aimed at reducing the potential impact of an eruption. Within the International Airways Volcano Watch (IAVW) framework, VOs are asked to operate a system designed to inform the aviation community about the status of a volcano and the expected threat associated. Despite the IAVW documentation reports the definition of the different color coded levels, operating the Aviation Color Code (ACC) can be a difficult task and VOs might adopt different strategies on how, when and why to change it. Following the two European Volcano Observatories and VAACs meetings, in 2019 and 2021, the European VOs agreed to present an overview on how they work about ACC. This analysis reveals that not all VOs in Europe have the usage of ACC as part of their operations, mainly because of lack of eruptions in recent times. Moreover, those VOs which regularly adopt ACC, operate it differently depending on the frequency of the eruptions, the type of volcano, the historical eruptive activity, the monitoring level, but also the agreement with the local Air Transport Navigation providers. At the same time all VOs change ACC in a very dynamic way to reflect the variability in the volcanic activity, except during decreasing activity and waning phase. This study shows that despite the ACC system is designed to provide a standard, its usage strongly depends on the evaluation of those responding to volcano emergencies. Still the usage of ACC is of relevance for the aviation stakeholders, that encourage its practice in addition to a wider dialogue with the Volcano Observatories in Europe.

Eruption source parameters are required for the initiation of numerical models, and significant work has gone into their estimation for volcanic ash transport and dispersal models. In application of these models, typically a given scenario or range of scenarios is used to inform source parameters and modelling approach. However, increasing computational competencies allow for more complex input parameters to be used with a look towards probabilistic model outputs. Here, we investigate the potential for developing probability density functions for inputs, in particular plume height, for ash-dispersal models using Etna volcano, Italy as a case study. We brought together information from the published record, from eruption databases, and from volcanic ash advisories (VAAs), to inform eruption source parameters. Despite being one of the most well studied and monitored volcanoes in the world, difficulties arose around data availability and bias. Published records tend to have a low temporal resolution and, through analysis of deposits, focus on larger eruptions, while the VAAs capture more recent activity at high temporal resolutions, but which has tended to be at lower levels. Distributions of plume height produced from combining information from these different datasets had long tails to large plume heights, which were difficult to confidently fit with commonly used distributions. Our analysis shows that producing probability density functions of eruption source parameters for a given volcano is nontrivial and limited by uncertainties and bias within eruption data.