GMPV9.4

Many (about 250) volcanoes worldwide are occupied by glaciers. Often glaciers are regarded as problematic for volcano monitoring, since glacier ice potentially masks evidence of volcanic activity. The most devastating volcanic eruptions of the last 100 years involved volcano-glacier interactions. The 1985 eruption of Nevado del Ruiz killed 23000 people, and the 2010 eruption of Eyjafjallajökull led to the closure of many European airports. Therefore, it is imperative to minimize these impacts on society by improving methods for monitoring of glacier-clad volcanoes. Amongst several methods, optical satellite remote sensing techniques are perhaps most auspicious, since they frequently have a relatively high temporal and spatial resolution, and are mostly freely available. They often clearly show the effects of volcanic activity on glaciers, including ice cauldron formation, ice fracturing and glacier terminus changes potentially due to subglacial melt or subglacial dome growth. This study has the objective to link pre-, syn- and post-eruption glacier behaviour to the type and timing of volcanic activity, and to develop a satellite based predictive tool for monitoring future eruptions. Despite several studies that link volcanic activity and changing glacier behaviour, the potential of using the latter to predict the former has yet to be systematically tested. Our approach is to observe how glaciers responded to past volcanic events using mostly, but not exclusively optical satellite imagery, and to build a database of examples for potential automated detection and forecasting on a global scale.

How to cite: Martin, M., Barr, I., Edwards, B., Symeonakis, E., and Spagnolo, M.: Using glaciers to identify, monitor, and predict volcanic activity, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-736, https://doi.org/10.5194/egusphere-egu21-736, 2021.

Localised elevated subglacial or subnivean geothermal activity has the potential to influence the morphology and flow of glaciers. Under conditions where the meltwater produced by these glaciovolcanic interactions is effectively drained away from the geothermal source, glaciovolcanic voids may form. These voids can only exist if the influx of geothermal vapours/gases provides more heat for melting than can be compensated by the inflow of ice. We identify two distinct glaciovolcanic void morphologies: horizontal passageways or chambers beneath the ice/snow, termed caves, and vertical shafts, termed chimneys. Both transient and long-lived caves and chimneys have been observed, with their formation sometimes being precursory or concurrent expressions of volcanic unrest. A better understanding of these features can therefore aid volcano monitoring programs and volcanic hazard assessments. Here we investigate the relationships between glaciological and geothermal conditions and their effects on the formation and evolution of glaciovolcanic caves and chimneys. We adapt existing analytical models, originally developed to describe subglacial hydrology, to derive and balance expressions for the radial melt-opening and creep-closure to find steady-state solutions for cave and chimney geometries. The effects of localised geothermal heat fluxes on fully drained glaciovolcanic voids are further investigated using a numerical full-Stokes ice-flow model. Idealised voids, subject to a prescribed geothermally induced mass balance, are inserted within synthetic glaciers of variable bed slope and thickness. Transient simulations are then used to map out the parameter space that influences the formation and evolution of glaciovolcanic caves and chimneys.

How to cite: Unnsteinsson, T., Flowers, G., and Williams-Jones, G.: Modelling the formation and evolution of glaciovolcanic caves and chimneys., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-6821, https://doi.org/10.5194/egusphere-egu21-6821, 2021.

The chemical and textural characterization of ash layers allows relating them to their volcanic source, provides information regarding an eruptive event and its impact; and pictures more accurate scenarios in case of future activity. Deception Island, located in central Bransfield strait (South Shetland Islands, Antarctica), consists of a horseshoe-shaped composite volcano, whose central part is occupied by a collapse caldera (8.5 x 10 km). It is considered to be among the most active volcanoes in Antarctica and a future eruption is very likely to happen, affecting the military and scientific research stations located nearby. The characterisation of volcanic ash layers found in marine sediment cores outside Deception Island can provide valuable information to: (i) determine the size and explosiveness of past eruptive events, (ii) assess the extent of their related hazards; and (iii) complete the eruption record of the island. Here, we present results of the characterization of the ash layers found on five marine sediment cores (TG-40, 41,43, 48 and 50) drilled proximal to Deception Island (less than 40 km) during the Antarctic Campaign of the MAGIA project (ANT-584/97). The final aim is to trace isochronous tephra horizons between the studied cores and try associating them to their respective eruptive events on the island. First, we carried out a granulometry analysis of each sampled layer and characterized the morphology of the fragments using as parameters: elongation, sphericity, solidity, and length/width ratio. Results obtained indicate that most of the layers are moderate to well sorted coarse ash. Minor amounts of lapilli and fine ash appear in the shallower (0 to 50 cm depth) layers. The granulometry and the morphology indicate that the layers have been reworked by turbiditic currents after the eruption, but not enough to destroy the information necessary for correlation. The petrographical study via optical microscope has highlighted the presence of three different types of volcanic glasses based on: (i) the colour of the ash particles under non-crossed polarized light; (ii) microcrystal content; (iii) texture; and (iv) vesicle abundance. Type 1 glasses, with black colour and generally shard shaped, show a low content in microcrystals and vesicles. Type 2, with brown colour and more spherical shapes, have a higher content in microcrystals and the fragments usually have a fluidal texture; the vesicle abundance is variable. Type 3, with yellow colour and variably shaped, are usually rich in microcrystals and vesicles, and have fluidal texture. In all families, the mineralogy of the microcrystals is mainly plagioclase (90%), pyroxene and olivine. The longest core (TG-48, 120 cm long) contains 15 layers, the deepest ones (113, 115 and 120 cm depth) may be correlated to the ones found in previous studies associated with a period of abundant volcanic activity around 2000 years BP.

This research is part of POLARCSIC and PTIVolcan research initiatives. This research was partially funded by the MINECO grants VOLCLIMA (CGL2015-72629-EXP), POSVOLDEC(CTM2016-79617-P)(AEI/FEDER-UE) and VOLGASDEC (PGC2018-095693-B-I00)(AEI/FEDER, UE). Analyzed tephra samples and sediment cores were provided by the rock repository of the Instituto de Ciencias del Mar del CSIC (ICM-CSIC) (http://gma.icm.csic.es/ca/dades).

How to cite: Polo Sánchez, A., Hopfenblatt, J., Geyer, A., Aulinas, M., Ercilla, G., and Álvarez-Valero, A.: Completing the eruptive record of Deception Island (South Shetland Islands, Antarctica) by characterizing ash layers in proximal marine sediments cores., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-2402, https://doi.org/10.5194/egusphere-egu21-2402, 2021.

Deception Island is the most active volcano in the South Shetland Islands (Antarctica) with more than 20 eruptions in the in the last two centuries, including the 1967, 1969 and 1970 most recent eruptive events, and three episodes of volcanic unrest since 1990 (1992, 1999 and 2014-2015). Since the discovery of Deception island in 1820, the number of scientific bases, touristic activities, and air and vessel traffic in the region, have considerably increased. Only the Antarctic Peninsula region, together with the South Shetland Islands, hosts 25 research stations and 3 summer field camps, which are located inside or within a 150 km radius distance from this active volcano. Nearby, the Palmer Archipelago and the north-western coast of the Antarctic Peninsula are both important tourist destinations exceeding 30,000 visitors per year with a significant increase in vessel traffic during the tourist season. This escalation in the amount of exposed infrastructure and population to a future eruption of Deception Island clearly urges the need to advancing our knowledge of the island’s volcanic and magmatic history and developing improved vulnerability analyses and long-term volcanic hazard assessments. However, past attempts to construct a volcanic hazard map of Deception have always been limited by the lack of a complete eruption record. In this sense, volcanic ash layers found in marine and lacustrine sediment cores, and glaciers outside Deception Island can provide valuable information to: (i) determine the size and explosiveness of past eruptive events; (ii) assess the extent of their related hazards (e.g. ash fall out); (iii) complete the eruption record of the island; and (iv) estimate the island’s eruption recurrence over time.

In this work, we provide a detailed, and up-to-date, revision of the current knowledge on Deception Island’s tephra record.  For this, we have compiled the DecTephra (Deception Island Tephra Record) database, which seeks recording the most relevant information of all up today known tephra layers with Deception Island as presumed source vent. DecTephra database includes 335 tephra layers (including cryptotephras) found in marine/lacustrine sediment and ice cores. For each tephra layer, we have compiled information regarding: (i) location (e.g. latitude, longitude, region) and characteristics of the sampling site (e.g. length of the sediment or ice core); and (ii) tephra characteristics (e.g. age, chemistry, granulometry). The analysis of the information included in this new database shows that Deception Island’s tephras can be observed at numerous proximal (< 150 km) sampling sites distributed all along the South Shetland Islands but also as far as in the Scotia Sea (> 1,000 km) and the South Pole (> 2,900 km). Also, identified isochronous tephra horizons allow defining periods of higher explosive eruptive activity in the island during the Holocene.

This research is part of POLARCSIC and PTIVolcan research initiatives. This research was partially funded by the MINECO projects VOLCLIMA (CGL2015-72629-EXP) and VOLGASDEC (PGC2018-095693-B-I00)(AEI/FEDER, UE). A.P.S is grateful for his JAE_Intro scholarship (JAEINT_20_00670).

How to cite: Hopfenblatt, J., Geyer, A., Aulinas, M., Polo Sánchez, A., and Álvarez-Valero, A.: DecTephra: A new database of Deception Island’s tephra record (Antarctica, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-2075, https://doi.org/10.5194/egusphere-egu21-2075, 2021.

Explosive eruptions in ice-covered volcanoes may deposit large volumes of tephra on the glaciated slopes.  The tephra can influence surface ablation and alter mass balance.  Ice melting by an eruption can change glacier geometry and temporarily alter the flow of outlet glaciers.  Conversely, when assessing the size of past tephra-producing eruptions in an ice-covered volcano the glacier complicates such estimates.  The effects of ice flow, dilation and shear need to be considered.  A tephra layer may get buried in the accumulation area, be transported by glacier flow and progressively removed over years-to-centuries by ice flow, eolian transport of exposed tephras and sediment transport in glacial rivers.  Here we report on a case study from the Mýrdalsjökull ice cap that covers the upper parts of the large Katla central volcano in south Iceland.  Most eruptions start beneath the 300-700 m thick ice cover within the Katla caldera, melt large volumes of ice and cause major jökulhlaups.  They also produce tephra layers that are preserved in soils around the volcano.  The most recent eruption in Katla occurred in October-November 1918, when a large tephra layer was deposited in a 3-weeks long eruption. By using a combination of (1) data obtained at or near the vent area within the SE-part of the Katla caldera in the year following the eruption, (2) mapping of the tephra as exposed at the present time in the ablation areas in the lower parts of the outlet glaciers, and (3) simple models of ice flow based on balance velocities and knowledge of mass balance, we estimate the location of fallout and the original thickness of the presently exposed tephra.  Photos taken in the vent area in 1919 indicate a tephra thickness of 20-30 m on the ice surface proximal to the vents.  The greatest thicknesses presently observed, 30-35 cm, occur where the layer outcrops in the lowermost parts of the ablation areas of the Kötlujökull and Sólheimajökull outlet glaciers.  A fallout location within the Katla caldera is inferred for the presently exposed tephra, as estimates of balance velocities imply lateral transport since 1918 of ~15 km for Kötlujökull, ~11 km for Sólheimajökull and about 2 km for the broad northern lobe of Sléttjökull.  The calculations indicate that ice transport with associated dilation of the glacier through the accumulation areas has resulted in significant thinning.   Thus, the layer that is now 0.3-0.35 m thick in the fastest flowing outlets is estimated to have been four to seven times thicker when it fell on the accumulation area within the ice-filled caldera.  In contrast, changes have been minor in the slowly moving Sléttjökull.  These findings allow for the construction of an isopach map for the glacier.  The results indicate that just under half of the total airborne tephra produced in the eruption fell within the Mýrdalsjökull glacier, with the remaining half spread out over a large part of Iceland.  These methods potentially allow for reconstruction of several tephra layers from ice-covered volcanoes in Iceland and elsewhere. 

How to cite: Gudmundsson, M. T., Larsen, G., Janebo, M. H., Hognadottir, T., and Jonsdottir, T.: Deciphering the fallout of tephra on glaciers in past eruptions, the case history of the Katla 1918 eruption, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-16038, https://doi.org/10.5194/egusphere-egu21-16038, 2021.

Baekdu Mountain is a 2,744 m high stratovolcano, located on the border of China and North Korea. The mountain has a caldera lake, Lake Cheonji, as a result of past volcanic activity. The ice area changes during winter in Lake Cheonji could act as a proxy for volcanic activity monitoring in Baekdu. As Baekdu laid on a political border, remote sensing allows us to quantify attributes of otherwise inaccessible or dangerous places. We assessed changes in winter (October–April) ice area in a high-altitude groundwater-fed caldera lake using Sentinel-1 synthetic aperture radar (SAR) data acquired from 2015 to 2020. To calculate the ice-covered area, 10 gray level co-occurrence matrix (GLCM) texture features were computed from SAR images obtained with VH (vertical transmission and horizontal reception) and VV (vertical transmission and vertical reception) polarizations. A support vector machine (SVM) algorithm was used to classify ice and water pixels from the GLCM layers, and the results from VH and VV imagery were combined to calculate the total area covered by ice. We examined the relationship between ice area and air temperature from the closest weather station, Samjiyeon using fixed period regression. The ice area was inversely proportional to 30-day averaged air temperature and these variables were highly correlated (-0.86). Our results show that there were no significant ice changes during the period, which indicates that there was no significant volcanic activity in Baekdu Mountain during the winters of 2015–2020. This study is expected to be useful for a better understanding of whether and how ice area changes in volcano lakes aid in eruption forecasting.

How to cite: Nur, A. S., Park, S., Lee, S., and Lee, C.-W.: Monitoring Volcanic Activity of Baekdu Mountain based on Ice Area Changes During the Winters of 2015-2020, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-3728, https://doi.org/10.5194/egusphere-egu21-3728, 2021.

The monitoring of seismic activity at active glacier-hosting volcanoes is challenging as volcanic and glacial earthquakes (i.e. icequakes) can have overlapping characteristics (i.e. frequencies, waveform shape and magnitude). Here we present results from the first study to target glacial activity at active ice-covered volcanoes in the Southern Chile. The primary focus so far has been on Llaima volcano, one of the largest and most active volcanoes in the region while hosting >14 km² of glacial ice on the flanks. We use a combination of automatic multi-station event detection and waveform cross-correlation to find candidate repeating icequakes in seismic data from the permanent volcano monitoring network recorded in early 2019. We identified dozens of low magnitude families of repeating seismic events across two months, the largest of which included over 200 events. These findings are comparable to results from analysis of seismic data recorded at Llaima volcano during the same time period in 2015. The persistent, repetitive nature of these events combined with their waveform characteristics and source locations suggest they originated from multiple sub-glacial stick-slip sources around the upper flanks of the volcano. We also deployed a network of seismo-acoustic sensors at Villarrica volcano in early 2020 to record glacial activity in concurrence with the lava lake and strombolian activity at the summit. We conclude that icequakes at Llaima volcano may be more common than previously thought and has implications for how seismic data at ice-covered volcanoes may be used for assessing future volcanic and glacial hazard potential.

How to cite: Lamb, O., Lees, J., Franco Marin, L., Lazo, J., Rivera, A., Shore, M., and Lee, S.: Discriminating glacial and volcanic seismicity at Llaima and Villarrica volcanoes, Chile, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-7940, https://doi.org/10.5194/egusphere-egu21-7940, 2021.