Delle Donne, Dario

Lahars, landslides and debris flows are rapid natural phenomena that can heavily impact on and modify the environment, not only that from which they are triggered but also the one in which they propagate or leave deposits. In particular, lahars can reach significant runout distances from source areas (e.g., several km) and this can mainly depend, among other factors, on the morphology experienced by such propagation. There are cases in the recent history of natural occurrences in which lahars impacted catastrophically on rural and urban settings, such as for example at Nevado del Ruiz volcano (Colombia) in 1985 causing the death of thousands of people living around there. A more recent event occurred on November 26, 2022 at Ischia island (Italy), which is an active volcano particularly subjected to the recurrence of these phenomena. In this case, the emplacement of some lahars caused the death of a few tens of people and the damaging of tens of building, besides the direct impact on local agriculture and tourism. In the nearby Neapolitan volcanic area, several other lahar events occurred in the historical past, not only during but also after or well after explosive eruptions, as the evidence that these phenomena are still to be considered as complex and often unpredictable extreme natural events, also exacerbated by the climate changes, but also that they have some recurrence that cannot be neglected. Such kind of recurrence is mainly related to the local weather, which can even affect the intrinsic behavior of the flows that detach from the source areas and invade the territory. On the other hand, this is not a strictly statistical issue, as there are instrumental measurements that support the fact that heavy rains can exacerbate a landscape already prone to sliding, avalanching, and other catastrophic phenomena. For this, the November 26, 2022 Ischia case study was chosen with the goal of reconstructing the physical features that led to the lahar generation and invasion, which is something that might occur in the future but that should be experienced with a dedicated scientific and territorial consciousness. What was done is an integration of multidisciplinary approaches, corroborated by data from the INGV-OV monitoring network installed on the volcano, capable of detecting the otherwise lost flow timing and dynamical behavior. In particular, the seismic evidence that accompanied the Ischia lahar events, along with the consideration of some lithological features leading to an estimation of flow velocity and dynamic pressure, allow to discriminate multiple lahar pulses over the early morning of November 26, 2022. The main findings of this contribution are that the potential of the Ischia lahars had a sort of recharge timespan which depended on the local weather and lithological features, while the threshold of the lahar trigger depended on the hydrogeological conditions. The seismic reconstruction of the entire event allowed to quantify the first of these two critical issues at Ischia island.

The persistent open-vent degassing of Mt. Etna is often punctuated by monthslong paroxysmal sequences characterized by episodes of violent Strombolian to lava fountaining activity. Understanding these gas-fueled transitions from quiescence to eruption requires routine measurement of gas fluxes. Here, we report SO2 flux measurements, obtained from a permanent UV camera system, collected over a two-year-long period spanning two paroxysmal sequences of Etna’s New South East Crater (NSEC) in December 2020/April 2021 and May/ October 2021. In both cases, SO2 flux increased from ≤ 3250 Mg/day during “ordinary” activity to ≥ 4200 Mg/day. We interpret these distinct SO2 degassing regimes in light of seismic and thermal observations and drawing on numerical simulations of sulfur degassing constrained by parental melt sulfur contents in Etna’s hawaiites. We find that initiation of a paroxysmal sequence results from an approximate doubling of the time-averaged rate of magma supply (and degassing) above the sulfur exsolution level (~150 MPa pressure), to >4m3/s. This corroborates recent models that argue for the triggering of paroxysmal sequences by escalating supply of volatile-rich magma to a reservoir ~3–4 km below the summit region. The non-stationary nature of magma flow and volcanic degassing we identify highlights the need for sustained surveillance to characterize long-term atmospheric budgets of volcanic volatiles

The eruption of the volcano Hunga Tonga-Hunga Ha‘apai on Jan 15, 2022, 04:14:54 UTC, was such energetic that instruments observed different physical phenomena all over the globe. In Italy, the Istituto Nazionale di Geofisica e Vulcanologia (INGV), who is continuously operating different kinds of monitoring networks, as e.g., the Italian Seismic Network (ISN), micro-barometric and infrasonic stations for monitoring the active volcanoes, ionospheric monitoring network (GNSS and ionosonde), recorded seismic, acoustic and electromagnetic signals originated by this exceptional event. The blast wave generated by the volcanic explosion of Hunga Tunga was recorded by the micro-barometric and infrasound stations installed at Phlegrean Fields (PF), at Stromboli volcano and on Mt. Etna. The first arrival was recorded at ~20:00 UTC, after travelling along the “short” great circle (17600 km), was succeeded by a second onset, about 3:40 h later, arriving at PF from the opposite direction. The mean propagation velocity in both directions was calculated as 310 m/s. The stations of the Etna Radio Observatory (ERO) are also equipped with micro-barometers, measuring the atmospheric pressure at a sampling rate of 5 min. The first group of atmospheric shock waves was recorded in the evening of Jan 15, 2022, while 36 hours later the ERO-stations observed a second signal after having completed the second orbit. The magnitude of M5.7 of the Hunga Tonga eruption was strong enough to record core phases (PKIKP, PKP), surface reflection of mantle phases (PP, SS), as well as Rayleigh and Love waves, at many stations of the ISN. The atmospheric waves generated by the eruption generated Travelling Ionospheric Disturbances in the ionosphere detected as disturbances in the Total Electron Content calculated by using GNSS data acquired by the GNSS network of INGV and variations of the ionospheric peak layer parameters (foF2, hmF2), recorded by the ionosonde installed on the Italian territory by INGV.

Open-vent volcanic activity is typically sustained by ascent and degassing of shallow magma, in which the rate of magma supply to the upper feeding system largely exceeds the rate of magma eruption. Such unbalance between supplied (input) and erupted (output) magma rates is thought to result from steady, degassing-driven, convective magma overturning in a shallow conduit/feeding dyke. Here, we characterize shallow magma circulation at Stromboli volcano by combining independent observations of heat (Volcanic Radiative Power; via satellite images) and gas (SO2 , via UV camera) output in a temporal interval (from August 1, 2018 to April 30, 2020) encompassing the summer 2019 effusive eruption and two paroxysmal explosions (on July 3 and August 28, 2019). We show that, during the phase of ordinary strombolian explosive activity that preceded the 2019 effusive eruption, the average magma input rate (0.1-0.2 m3 /s) exceeds the magma eruption rate (0.001-0.01 m3 /s) by ∼2 orders of magnitude. Conversely, magma input and output rates converge to an average of ∼0.4 m3 /s during the summer 2019 summit effusion, implying an overall suppression of magma recycling back into the feeding system, and hence of excess degassing. We find that, during the effusive eruption, the peak in SO2 emissions lags behind the thermal emission peak by ∼27 days, suggesting that magma output, feeding the lava flow field, initially dominates over magma input in the conduit. We propose that this conduit mass unloading, produced by this initial phase of the effusive eruption, leads to an overall decompression (of up to 30 Pa/s) of the shallow plumbing system, ultimately causing ascent of less-dense, volatile-rich magma batch(es) from depth, enhanced explosive activity, and elevated SO2 fluxes culminating into a paroxysmal explosion on August 28. Our results demonstrate that combined analysis of thermal and SO2 flux time-series paves the way to improved understanding of shallow magmatic system dynamics at open-vent volcanoes, and of the transition from explosive to effusive activity regimes.

Improving volcanic gas monitoring techniques is central to better understanding open-vent, persistently degassing volcanoes. SO2 cameras are increasingly used in volcanic gas studies, but observations are commonly limited to one single camera alone viewing the volcanic plume from a specific viewing direction. Here, we report on high frequency (0.5 Hz) systematic measurements of the SO2 flux at Stromboli, covering a 1-year long observation period (June 2017-June 2018), obtained from two permanent SO2 cameras using the same automated algorithm, but imaging the plume from two different viewing directions. Our aim is to experimentally validate the robustness of automatic SO2 camera for volcano monitoring and to demonstrate the advantage of using two co-exposed SO2 camera stations to better capturing degassing dynamics at open-vent volcanoes. The SO2 flux time-series derived from the two SO2 camera stations exhibit good match, demonstrating the robustness of the automatic SO2 camera method. Our high-temporal resolution SO2 records resolve individual Strombolian explosions as transient, repetitive gas bursts produced by the sudden release of over pressurized gas pockets and scoriae. Calculations show that explosive degassing activity accounts for ∼10% of the total SO2 emission budget (dominated by passive degassing) during mild regular open-vent activity. We show that the temporal variations of the explosive SO2 flux go in tandem with changes in total SO2 flux and VLP seismicity, implicating some commonality in the source processes controlling passive degassing and explosive activity. We exploited the spatial resolution of SO2 camera to discriminate degassing at two distinct regions of the crater area, and to minimize biases due by the station position respect to the target plume. We find that the SO2 fluxes from southwest-central (SWCC) and northeast (NEC) crater areas oscillate coherently but those from the NEC are more sensitive to the changes in the volcanic intensity. We interpret this as due to preferential gas/magma channeling into the structurally weaker north-eastern portion of the crater terrace in response to increasing supply rate of buoyant, bubble-rich magma in the shallow plumbing system.

ASHER, a new sensor for the characterization of tephra fallout in real time, was designed and developed for easy field deployment during volcanic eruptions. It can provide information on the accumulation rate of tephra fallout in real time as well as grain-size and settling velocity of falling particles. Particle detection is achieved with a laser barrier, with size and settling velocity being calculated from the amplitude and duration of obscuration peaks. The sampling rate (31,500 Hz), laser thickness (0.5 mm) and operation (ON/OFF state and dual acquisition mode) are adapted to minimize the noise level and allow detection of particles as small as ~100 μm. Additional measurements of weight and level of accumulated material within a removable collector allow broadening of the ASHER operation to accumulation rate from 10−2 to 103 g m-2s-1. Detailed calibration tests were performed in laboratory conditions on single grains of known shape and density along with a high-speed camera to test the capability to measure grain size and terminal velocity, and during two field campaigns at Stromboli and Etna volcanoes to test the operation in the field. Long-term field deployment has shown that combining the optical barrier with an automatic collector allows for a better characterization of tephra fallout, providing an estimate of density, and, therefore, it optimizes sensor operation and minimizes false alerts. Moreover, the low power requirements and onboard processing allows to operate the sensor remotely and solely on solar power in a remote location. Although technical improvements in sensor sensitivity and processing are still possible, the results presented suggest that ground sensors for real-time detection and analysis of tephra could potentially contribute to understanding the dynamics of explosive eruptions and could be successfully integrated into monitoring systems of active volcanoes.

The benign fuming activity of dormant volcanoes is punctuated by phases of escalating degassing activity that, on some occasions, ultimately prelude to eruption. However, understanding the drivers of such unrest is complicated by complex interplay between magmatic and hydrothermal processes. Some of the most comprehensively characterised degassing unrest have recently been observed at La Fossa cone on Vulcano Island, but whether or not these episodes involve new, volatile-rich ascending magma remains debated. Here, we use volcanic gas measurements, in combination with melt inclusion information, to propose that heightened sulphur dioxide flux during the intense fall 2021 La Fossa unrest is sourced by degassing of volatile-rich mafic magma. Calculations using a numerical model indicate observations are consistent with the unrest being triggered by the emplacement of ∼3·106 m3 of mafic magma at ∼4–5 km depth. Degassing of mafic magma is argued as a recurrent driver of unrest at dormant volcanoes worldwide.

Volcano seismicity is one of the key parameters to understand magma dynamics of erupting volcanoes. However, the physical process at the origin of the resulting complex and broadband seismic signals remains unclear. Syn-eruptive very long period (VLP) seismic signals have been explained in terms of the sudden expansion of gas pockets rising in the liquid melt. Their origin is linked to a magma dynamics which triggers the explosive process occurring before the explosive onset. We provide evidence based on acoustic, thermal, and ground deformation data to demonstrate that VLP signals at Stromboli are generated at the top of the magma column mainly after the explosion onset. We show that VLP amplitude and duration scale with the eruptive flux which induces a decompression of 103-104 Pa involving the uppermost ~ 250 m of the feeding conduit. The seismic VLP source represents the final stage of a ~ 200 s long charge and discharge mechanism the magma column has to release excess gas accumulated at the base of a denser and degassed magma mush. The position of the VLP seismic source coincides with the centroid of the shallow mush plug and tracks elevation changes of the magma free surface.

The ordinarily benign activity of basaltic volcanoes is periodically interrupted by violent paroxysmal explosions ranging in size from Hawaiian to Plinian in the most extreme examples. These paroxysms often occur suddenly and with limited or no precursors, leaving their causal mechanisms still incompletely understood. Two such events took place in summer 2019 at Stromboli, a volcano otherwise known for its persistent mild open-vent activity, resulting in one fatality and damage to infrastructure. Here, we use a post hoc analysis and reinterpretation of volcanic gas compositions and fluxes acquired at Stromboli to show that the two paroxysms were preceded by detectable escalations in volcanic plume CO2 degassing weeks to months beforehand. Our results demonstrate that volcanic gas CO2 is a key driver of explosions and that the preparatory periods ahead of explosions in basaltic systems can be captured by precursory CO2 leakage from deeply stored mafic magma.

At open-vent basaltic volcanoes, resolving the activity escalation that heralds larger, potentially harmful eruptions is challenged by the persistent mild ordinary activity, which often masks the precursory unrest signals related to heightened magma transport from depth. Gas (SO2 and CO2) fluxes at surface are controlled by rate of magma transport and degassing within the magma plumbing system, and thus constitute key parameters to infer deep magma budget and dynamics. Here, we use several year-long (2014-present) gas observations at Etna and Stromboli volcanoes, in Sicily, to provide new evidence for the utility of long-term instrumental gas monitoring in real-time detecting the early phase of unrest prior eruption, and for characterizing syn-eruptive dynamics. To this aim, we use information from a gas monitoring network of permanent ultraviolet (UV) cameras and automatic Multi-Gas instruments that, combined with geophysical observations, allow characterizing changes in degassing and eruptive dynamics at high temporal/spatial resolution. Our results show that the paroxysmal (lava fountaining) explosions that periodically interrupted persistent open-vent activity on Etna (during 2014-2020) were accompanied by systematic, repetitive SO2 emission patterns prior, during, and after eruptions. These allow us identifying the characteristic pre- syn- and post- eruptive degassing regimes, and to establish thresholds in the SO2 flux record that mark phases of unrest. On Stromboli, the much improved temporal/spatial resolution of UV cameras allows resolving the escalation of regular strombolian activity, and its concentration toward its North-east crater, that heralds onset of effusive eruptions. During effusive eruption, although magma level drops in the conduit and explosive summit activity ceases, UV camera observations can still detect explosive gas bursts deep in the conduit while no infrasonic activity is detected. Combining the UV camera-derived SO2 fluxes with CO2/SO2 ratio records measured by the Multi-Gas, the CO2 flux can be inferred. We find that such CO2 flux time-series can allow tracking degassing of deeply stored mafic magma months before Stromboli"s eruptions. We finally show that remotely sensed gas emission and thermal activity can be combined together to characterize the dynamics of shallow magmatic system prior to and during unrest, ultimately helping to define timing of magma re-charging events driving the eruptions.