Andrea Bevilacqua

Quantifying volcanic hazard at Campi Flegrei caldera (Italy) with uncertainty assessment: 1. Vent opening maps

Campi Flegrei is an active volcanic area situated in the Campanian Plain (Italy) and dominated by a resurgent caldera. The great majority of past eruptions have been explosive, variable in magnitude, intensity, and in their vent locations. In this hazard assessment study we present a probabilistic analysis using a variety of volcanological data sets to map the background spatial probability of vent opening conditional on the occurrence of an event in the foreseeable future. The analysis focuses on the reconstruction of the location of past eruptive vents in the last 15 ka, including the distribution of faults and surface fractures as being representative of areas of crustal weakness. One of our key objectives was to incorporate some of the main sources of epistemic uncertainty about the volcanic system through a structured expert elicitation, thereby quantifying uncertainties for certain important model parameters and allowing outcomes from different expert weighting models to be evaluated. Results indicate that past vent locations are the most informative factors governing the probabilities of vent opening, followed by the locations of faults and then fractures. Our vent opening probability maps highlight the presence of a sizeable region in the central eastern part of the caldera where the likelihood of new vent opening per kilometer squared is about 6 times higher than the baseline value for the whole caldera. While these probability values have substantial uncertainties associated with them, our findings provide a rational basis for hazard mapping of the next eruption at Campi Flegrei caldera.

Quantifying volcanic hazard at Campi Flegrei caldera (Italy) with uncertainty assessment: 2. Pyroclastic density current invasion maps

Campi Flegrei (CF) is an example of an active caldera containing densely populated settlements at very high risk of pyroclastic density currents (PDCs). We present here an innovative method for assessing background spatial PDC hazard in a caldera setting with probabilistic invasion maps conditional on the occurrence of an explosive event. The method encompasses the probabilistic assessment of potential vent opening positions, derived in the companion paper, combined with inferences about the spatial density distribution of PDC invasion areas from a simplified flow model, informed by reconstruction of deposits from eruptions in the last 15 ka. The flow model describes the PDC kinematics and accounts for main effects of topography on flow propagation. Structured expert elicitation is used to incorporate certain sources of epistemic uncertainty, and a Monte Carlo approach is adopted to produce a set of probabilistic hazard maps for the whole CF area. Our findings show that, in case of eruption, almost the entire caldera is exposed to invasion with a mean probability of at least 5%, with peaks greater than 50% in some central areas. Some areas outside the caldera are also exposed to this danger, with mean probabilities of invasion of the order of 5–10%. Our analysis suggests that these probability estimates have location-specific uncertainties which can be substantial. The results prove to be robust with respect to alternative elicitation models and allow the influence on hazard mapping of different sources of uncertainty, and of theoretical and numerical assumptions, to be quantified.

Temporal models for the episodic volcanism of Campi Flegrei caldera (Italy) with uncertainty quantification

After the large-scale event of Neapolitan Yellow Tuff (~15 ka B.P.), intense and mostly explosive volcanism has occurred within and along the boundaries of the Campi Flegrei caldera (Italy). Eruptions occurred closely spaced in time, over periods from a few centuries to a few millennia, and were alternated with periods of quiescence lasting up to several millennia. Often events also occurred closely in space, thus generating a cluster of events. This study had two main objectives: (1) to describe the uncertainty in the geologic record by using a quantitative model and (2) to develop, based on the uncertainty assessment, a long-term subdomain specific temporal probability model that describes the temporal and spatial eruptive behavior of the caldera. In particular, the study adopts a space-time doubly stochastic nonhomogeneous Poisson-type model with a local self-excitation feature able to generate clustering of events which are consistent with the reconstructed record of Campi Flegrei. Results allow the evaluation of similarities and differences between the three epochs of activity as well as to derive eruptive base rate of the caldera and its capacity to generate clusters of events. The temporal probability model is also used to investigate the effect of the most recent eruption of Monte Nuovo (A.D. 1538) in a possible reactivation of the caldera and to estimate the time to the next eruption under different volcanological and modeling assumptions.

Assessing future vent opening locations at the Somma‐Vesuvio volcanic complex: 2. Probability maps of the caldera for a future Plinian/sub‐Plinian event with uncertainty quantification

In this study, we combine reconstructions of volcanological data sets and inputs from a structured expert judgment to produce a first long-term probability map for vent opening location for the next Plinian or sub-Plinian eruption of Somma-Vesuvio. In the past, the volcano has exhibited significant spatial variability in vent location; this can exert a significant control on where hazards materialize (particularly of pyroclastic density currents). The new vent opening probability mapping has been performed through (i) development of spatial probability density maps with Gaussian kernel functions for different data sets and (ii) weighted linear combination of these spatial density maps. The epistemic uncertainties affecting these data sets were quantified explicitly with expert judgments and implemented following a doubly stochastic approach. Various elicitation pooling metrics and subgroupings of experts and target questions were tested to evaluate the robustness of outcomes. Our findings indicate that (a) Somma-Vesuvio vent opening probabilities are distributed inside the whole caldera, with a peak corresponding to the area of the present crater, but with more than 50% probability that the next vent could open elsewhere within the caldera; (b) there is a mean probability of about 30% that the next vent will open west of the present edifice; (c) there is a mean probability of about 9.5% that the next medium-large eruption will enlarge the present Somma-Vesuvio caldera, and (d) there is a nonnegligible probability (mean value of 6–10%) that the next Plinian or sub-Plinian eruption will have its initial vent opening outside the present Somma-Vesuvio caldera.

The Effects of Vent Location, Event Scale, and Time Forecasts on Pyroclastic Density Current Hazard Maps at Campi Flegrei Caldera (Italy)

This study presents a new method for producing long-term hazard maps for pyroclastic density currents (PDC) originating at Campi Flegrei caldera. Such method is based on a doubly stochastic approach and is able to combine the uncertainty assessments on the spatial location of the volcanic vent, the size of the flow and the expected time of such an event. The results are obtained by using a Monte Carlo approach and adopting a simplified invasion model based on the box model integral approximation. Temporal assessments are modelled through a Cox-type process including self-excitement effects, based on the eruptive record of the last 15 kyr. Mean and percentile maps of PDC invasion probability are produced, exploring their sensitivity to some sources of uncertainty and to the effects of the dependence between PDC scales and the caldera sector where they originated. Conditional maps representative of PDC originating inside limited zones of the caldera, or of PDC with a limited range of scales are also produced. Finally, the effect of assuming different time windows for the hazard estimates is explored, also including the potential occurrence of a sequence of multiple events. Assuming that the last eruption of Monte Nuovo (A.D. 1538) marked the beginning of a new epoch of activity similar to the previous ones, results of the statistical analysis indicate a mean probability of PDC invasion above 5% in the next 50 years on almost the entire caldera (with a probability peak of ~25% in the central part of the caldera). In contrast, probability values reduce by a factor of about 3 if the entire eruptive record is considered over the last 15 kyr, i.e. including both eruptive epochs and quiescent periods.

Assessing future vent opening locations at the Somma‐Vesuvio volcanic complex: 1. A new information geodatabase with uncertainty characterizations

This study presents new and revised data sets about the spatial distribution of past volcanic vents, eruptive fissures, and regional/local structures of the Somma-Vesuvio volcanic system (Italy). The innovative features of the study are the identification and quantification of important sources of uncertainty affecting interpretations of the data sets. In this regard, the spatial uncertainty of each feature is modeled by an uncertainty area, i.e., a geometric element typically represented by a polygon drawn around points or lines. The new data sets have been assembled as an updatable geodatabase that integrates and complements existing databases for Somma-Vesuvio. The data are organized into 4 data sets and stored as 11 feature classes (points and lines for feature locations and polygons for the associated uncertainty areas), totaling more than 1700 elements. More specifically, volcanic vent and eruptive fissure elements are subdivided into feature classes according to their associated eruptive styles: (i) Plinian and sub-Plinian eruptions (i.e., large- or medium-scale explosive activity); (ii) violent Strombolian and continuous ash emission eruptions (i.e., small-scale explosive activity); and (iii) effusive eruptions (including eruptions from both parasitic vents and eruptive fissures). Regional and local structures (i.e., deep faults) are represented as linear feature classes. To support interpretation of the eruption data, additional data sets are provided for Somma-Vesuvio geological units and caldera morphological features. In the companion paper, the data presented here, and the associated uncertainties, are used to develop a first vent opening probability map for the Somma-Vesuvio caldera, with specific attention focused on large or medium explosive events.

Analyzing Complex Models Using Data and Statistics

Complex systems (e.g., volcanoes, debris flows, climate) commonly have many models advocated by different modelers and incorporating different modeling assumptions. Limited and sparse data on the modeled phenomena does not permit a clean discrimination among models for fitness of purpose, and, heuristic choices are usually made, especially for critical predictions of behavior that has not been experienced. We advocate here for characterizing models and the modeling assumptions they represent using a statistical approach over the full range of applicability of the models. Such a characterization may then be used to decide the appropriateness of a model for use, and, perhaps as needed weighted compositions of models for better predictive power. We use the example of dense granular representations of natural mass flows in volcanic debris avalanches, to illustrate our approach.