Volcanic Signals Research Articles

How a non harmonic-like tremor at a volcanic lake could be caused by sustained wind: A case of study of Taupō volcano

Taupō volcano lies underneath the largest lake in New Zealand. It undergoes unrest roughly every ten years, has minor eruptions every few hundred years, and has supereruptions. Understanding volcanic seismic signals provides insights into the volcano dynamics and helps to anticipate eruptions. Harmonic and non-harmonic volcanic tremor are common signals that indicate fluid mobilisation in volcanic systems. We observed a signal that resembled non-harmonic tremor during 2019 at Taupō volcano. Careful time-frequency analysis of seismic data and weather data from around the lake revealed that the signals were not magmatic in origin, but were most likely from lake microseisms caused by special conditions of the wind, which generates wind-driven waves in the lake. The frequency range of lake microseisms at Taupō starts and ends with a dominant frequency of 0.7–1.0 Hz, and during the maximum energy peak, the dominant frequency is of 0.50–0.56 Hz. Such signals may be common in caldera lakes, which need to be distinguished from tremor caused by volcanic activity, so as not to exaggerate the probability of eruptions.

Mercury enrichments as a paleo-volcanism proxy: Sedimentary bias and a critical analysis across the end-Triassic

Mercury (Hg) anomalies in sedimentary rocks have been increasingly used in paleoclimatology studies for tracing volcanic signals, as Hg emissions from volcanic activity can cause contemporaneous sedimentary Hg enrichment. However, non-volcanic sedimentary controls on Hg have clear potential to mask these signals. These factors include host phase variability linked to environmentally controlled sourcing and settling changes, and/or variable preservation conditions associated with weathering, oxidation and diagenesis. Such factors can limit the efficacy of Hg as a paleo-volcanism proxy. In this study, sedimentary effects on Hg concentration within a complex depositional system in southwest England (St. Audrie's Bay) across the end-Triassic have been analyzed, together with published data from coeval end-Triassic sections globally – an interval of time coeval with the Central Atlantic Magmatic Province (CAMP). Our statistical analysis of Hg and associated geochemical data highlights significant fluctuations in sedimentary Hg due to relative supply differences in Hg and host phases, as well as the changing types and preservation conditions of host phases. End-Triassic sections globally show a consistent undersupply of Hg relative to organic matter across the end-Triassic mass extinction (ETME). To better assess the magnitude and significance of possible Hg enrichments in sedimentary rocks, we present a statistical method for quantifying Hg anomalies to robustly distinguish Hg variations linked to host phase/depositional changes from paleo-volcanism. Our method supports the existence of transient but asynchronous Hg anomalies linked to volcanism from the CAMP across the end-Triassic in most global sections, albeit not in the St. Audrie's Bay section.

Explosive eruption style modulates volcanic electrification signals

Volcanic lightning detection has proven useful to volcano monitoring by providing information on eruption onset, source parameters, and ash cloud directions. However, little is known about the influence of changing eruptive styles on the generation of charge and electrical discharges inside the eruption column. The 2021 Tajogaite eruption (La Palma, Canary Islands) provided the rare opportunity to monitor variations in electrical activity continuously over several weeks using an electrostatic lightning detector. Here we show that throughout the eruption, silicate particle charging is the main electrification mechanism. Moreover, we find that the type of electrical activity is closely linked to the explosive eruption style. Fluctuations in the electrical discharge rates are likely controlled by variations in the mass eruption rate and/or changes in the eruption style. These findings hold promise for obtaining near real-time information on the dynamic evolution of explosive volcanic activity through electrostatic monitoring in the future.

Changing magma recharge/discharge dynamics during the 2020–22 lava fountaining activity at Mt. Etna revealed by tilt deformation and volcanic tremor

Mt. Etna exhibited 62 lava fountaining events between December 13, 2020 and February 21, 2022. We analyzed tilt deformations and volcanic tremor amplitude time series, to characterize both eruptions and the preceding preparatory phases in terms of magnitude and speed of development of the volcanic phenomena, as well as to reconstruct the processes that took place inside the plumbing system and drove this intense period of activity. Based on deflation amplitudes associated with lava fountains and according with other retrieved parameters (i.e., magnitude of inflations, inflation and deflation velocities and volcanic tremor amplitudes), three periods have been distinguished. Period I displays higher values of all the aforementioned parameters, interpreted as conspicuous volumes of volatile-rich magma transferred towards the surface. Period II shows lower values evidencing lack of important new injections of magma from depth, whereas period III reveals a general increasing trend possibly related to gas flushing from magma residing in deeper portions of the plumbing system. Detailed elaborations of tilt signals allowed the identification of short-lived inflations accompanying the early stages of lava fountains during period II. Our results reveal significant correlations between amplitude and velocity of tilt and volcanic tremor signals associated with lava fountains and evidence the crucial role of gas in the inflation-deflation cycles.

Ionospheric compensation in L-band InSAR time-series: Performance evaluation for slow deformation contexts in equatorial regions

Multi-temporal Synthetic Aperture Radar Interferometry (MT-InSAR) is the only geodetic technique allowing to measure ground deformation down to mm/yr over continuous areas. Vegetation cover in equatorial regions favors the use of L-band SAR data to improve interferometric coherence. However, the electron content of ionosphere, affecting the propagation of the SAR signal, shows particularly strong spatio-temporal variations near the equator, while the dispersive nature of the ionosphere makes its effect stronger on low-frequencies, such as L-band signals. To tackle this problem, range split-spectrum method can be implemented to compensate the ionospheric phase contribution. Here, we apply this technique for time-series of ALOS-PALSAR data, and propose optimizations for low-coherence areas. To evaluate the efficiency of this method to retrieve subtle deformation rates in equatorial regions, we compute time-series using four ALOS-PALSAR datasets in contexts of low to medium coherence, showing slow deformation rates (mm/yr to cm/yr). The processed tracks are located in Ecuador, Trinidad and Sumatra, and feature 15 to 19 acquisitions including very high, dominating ionospheric noise, corresponding to equivalent displacements of up to 2 m. The correction method performs well and allows to reduce drastically the noise level due to ionosphere, with significant improvement compared with a simple plane fitting method. This is due to frequent highly non-linear patterns of perturbation, characterizing equatorial TEC distribution. We use semivariograms to quantify the uncertainty of the corrected time-series, highlighting its dependence on spatial distance. Thus, using ALOS-PALSAR-like archive, one can expect a detection threshold on the Line-of-Sight velocity ranging between 3 and 6 mm/yr, depending on the spatial wavelength of the signal to be observed. These values are consistent with the accuracy derived from the comparison of velocities between two tracks in their overlapping area. In the case studies that we processed, the time-series corrected from ionosphere allows to retrieve accurately fault creep and volcanic signal but it is still too noisy for retrieving tiny long-wavelength signals such as slow (mm/yr) interseismic strain accumulation.

Gliding tremors associated with the 26 second microseism in the Gulf of Guinea

A location in the Gulf of Guinea, which emits monochromatic seismic waves at 26-second period, seemingly continuously, was identified in the 1960s. However, the origin of these seismic waves remains enigmatic to date. Here we use three-component data from two seismic arrays in Africa, as well as additional seismic data compiled from around the world, to investigate the tremors. We identify frequency glides accompanying the previously known 26 s microseism which start at the same frequency and originate in the same, fixed location in the Gulf of Guinea. The stable characteristics of the tremors, their low frequency range, the implied large spatial scale, and the decades-long timescales where this phenomenon seems to have been active, all point towards a gap in our understanding of long period oceanic and volcanic signals. Since tremor is an important tool to monitor volcanoes, understanding this phenomenon may affect future forecasting of volcanic activity.

Distributed dynamic strain sensing of very long period and long period events on telecom fiber-optic cables at Vulcano, Italy

Volcano-seismic signals can help for volcanic hazard estimation and eruption forecasting. However, the underlying mechanism for their low frequency components is still a matter of debate. Here, we show signatures of dynamic strain records from Distributed Acoustic Sensing in the low frequencies of volcanic signals at Vulcano Island, Italy. Signs of unrest have been observed since September 2021, with CO2 degassing and occurrence of long period and very long period events. We interrogated a fiber-optic telecommunication cable on-shore and off-shore linking Vulcano Island to Sicily. We explore various approaches to automatically detect seismo-volcanic events both adapting conventional algorithms and using machine learning techniques. During one month of acquisition, we found 1488 events with a great variety of waveforms composed of two main frequency bands (from 0.1 to 0.2 Hz and from 3 to 5 Hz) with various relative amplitudes. On the basis of spectral signature and family classification, we propose a model in which gas accumulates in the hydrothermal system and is released through a series of resonating fractures until the surface. Our findings demonstrate that fiber optic telecom cables in association with cutting-edge machine learning algorithms contribute to a better understanding and monitoring of volcanic hydrothermal systems.

Temporal variations of surface mass balance over the last 5000 years around Dome Fuji, Dronning Maud Land, East Antarctica

Abstract. We reconstructed surface mass balance (SMB) around Dome Fuji, Antarctica, over the last 5000 years using the data from 15 shallow ice cores and seven snow pits. The depth–age relationships for the ice cores were determined by synchronizing them with a layer-counted ice core from West Antarctica (WAIS Divide ice core) using volcanic signals. The reconstructed SMB records for the last 4000 years show spatial patterns that may be affected by their locations relative to the ice divides around Dome Fuji, proximity to the ocean, and wind direction. The SMB records from the individual ice cores and snow pits were stacked to reconstruct the SMB history in the Dome Fuji area. The stacked record exhibits a long-term decreasing trend at -0.037±0.005 kg m−2 per century over the last 5000 years in the preindustrial period. The decreasing trend may be the result of long-term surface cooling over East Antarctica and the Southern Ocean and sea ice expansion in the water vapor source areas. The multidecadal to centennial variations of the Dome Fuji SMB after detrending the record shows four distinct periods during the last millennium: a mostly negative period before 1300 CE, a slightly positive period from 1300 to 1450 CE, a slightly negative period from 1450 to 1850 CE with a weak maximum around 1600 CE, and a strong increase after 1850 CE. These variations are consistent with those of previously reconstructed SMB records in the East Antarctic plateau. The low accumulation rate periods tend to coincide with the combination of strong volcanic forcings and solar minima for the last 1000 years, but the correspondence is not clear for the older periods, possibly because of the lack of coincidence of volcanic and solar forcings or the deterioration of the SMB record due to a smaller number of stacked cores.

Volcanism and carbon cycle perturbations in the High Arctic during the Late Jurassic – Early Cretaceous

Large perturbations in the global carbon cycle recorded as carbon-isotope (δ13C) excursions (CIEs) in both organic carbon and carbonate records have been linked to volcanism during the emplacement of Large Igneous Provinces (LIPs). This link is based primarily on the purported temporal coincidence between CIEs and LIP emplacement. Mercury (Hg) concentration in sedimentary rocks has been used as a regional to global tracer of large-scale volcanic activity, yet few studies have been undertaken on Upper Jurassic – Lower Cretaceous sediments from Boreal localities compared to those for Tethyan (northern mid-latitude) successions. This has limited our understanding of the regional-to-global spatial impact of volcanic activity during this period. This study examines the Hg record as a proxy for volcanism, and the δ13C records from organic matter (δ13Corg) of CIEs from the uppermost Jurassic to Lower Cretaceous (Callovian – Aptian) successions from Axel Heiberg and Spitsbergen in the Canadian Arctic and Svalbard archipelagos, respectively. This interval includes three regional- to global CIEs. These sections show no significant variation in the ratio of Hg to total organic carbon (TOC) across the Boreal-wide Volgian negative CIE (Volgian Isotopic Carbon Excursion, “VOICE”), which has not been associated with LIP volcanism. The examined successions spanning this interval all show some influence from changing environmental or post-burial parameters, however, which could have (partially) overprinted a volcanic signal. Despite some problems in stratigraphically constraining the Weissert Event, increased Hg/TOC ratios are observed across this interval, which may be partially driven by volcanism associated with the emplacement of the Paraná-Etendeka Traps. A spike in Hg/TOC is observed immediately prior to the negative peak of the Aptian Oceanic Anoxic Event (OAE1a) CIE, supporting recent evidence of a pulse of High Arctic Large Igneous Province (HALIP) volcanic activity preceding this oceanic anoxic event.

Understanding Model‐Observation Discrepancies in Satellite Retrievals of Atmospheric Temperature Using GISS ModelE

AbstractWe examine multiple factors in the representation of satellite‐retrieved atmospheric temperature diagnostics in historical simulations of climate change during the satellite era (specifically 1979–2021) using GISS ModelE contributions to the Coupled Model Intercomparison Project (Phase 6) (CMIP6). The tropospheric and stratospheric trends in these diagnostics are affected by greenhouse gases (notably carbon dioxide and ozone), coupling with the ocean, volcanic aerosols, solar activity and compositional and dynamic feedbacks. We explore the impacts of internal variability, changing forcing specifications, composition interactivity, the quality of the stratospheric circulation, vertical resolution, and possible impacts of the mis‐specification of volcanic aerosol optical depths. Overall temperature trends throughout the satellite period are well captured, but discrepancies at all levels exist and have multiple distinct causes. We find that stratospheric comparisons (using Stratospheric Sounding Unit (SSU) retrievals and successor instruments) are most affected by variations in the representation of ozone depletion and feedbacks, followed by the volcanic signals. Tropospheric skill (using the Microwave Sounding Unit (MSU) retrievals) is affected by the trends in ocean heat uptake and tropospheric aerosols, but also by the representation of stratospheric processes through the impact of the Brewer‐Dobson circulation on the height of the tropical tropopause. We demonstrate that no single factor is the dominant cause of the discrepancies and that almost all observations lie within the broad envelope of structural uncertainty.

Large-scale demonstration of machine learning for the detection of volcanic deformation in Sentinel-1 satellite imagery

Radar (SAR) satellites systematically acquire imagery that can be used for volcano monitoring, characterising magmatic systems and potentially forecasting eruptions on a global scale. However, exploiting the large dataset is limited by the need for manual inspection, meaning timely dissemination of information is challenging. Here we automatically process ~ 600,000 images of > 1000 volcanoes acquired by the Sentinel-1 satellite in a 5-year period (2015–2020) and use the dataset to demonstrate the applicability and limitations of machine learning for detecting deformation signals. Of the 16 volcanoes flagged most often, 5 experienced eruptions, 6 showed slow deformation, 2 had non-volcanic deformation and 3 had atmospheric artefacts. The detection threshold for the whole dataset is 5.9 cm, equivalent to a rate of 1.2 cm/year over the 5-year study period. We then use the large testing dataset to explore the effects of atmospheric conditions, land cover and signal characteristics on detectability and find that the performance of the machine learning algorithm is primarily limited by the quality of the available data, with poor coherence and slow signals being particularly challenging. The expanding dataset of systematically acquired, processed and flagged images will enable the quantitative analysis of volcanic monitoring signals on an unprecedented scale, but tailored processing will be needed for routine monitoring applications.

The Dome Fuji ice core DF2021 chronology (0–207 kyr BP)

Precise ice-core chronologies are essential for identifying the timing and duration of polar climatic changes as well as their phasing with the changes in other parts of the globe. However, existing ice-core chronologies beyond the last 60 kyr show relatively large disagreements with each other and with U–Th chronologies of speleothems. Here, we constructed new ice and gas age scales for the Dome Fuji (DF) core (DF2021) over the last 207 kyr by combining a Bayesian dating model and firn densification model, constrained by various types of chronological and glaciological information including new δO2/N2 age markers, precise synchronization to other high-quality chronologies (volcanic, cosmogenic, and CH4 signals), and high-resolution δ15N of N2 (reflecting past firn thickness). The new chronology is tightly constrained by synchronization to other well-dated records for the last 60 kyr, whereas it is independent from other chronologies for the older period. For the last 60 kyr, the DF2021 chronology agrees with the layer-counted ice core chronologies (GICC05 and a part of WD2014) and U–Th chronologies of speleothems within ∼200 years. For the period 60–130 kyr BP, the timing of all Dansgaard-Oeschger warming events on DF2021 agree with those of corresponding events in the U–Th dated Chinese or European speleothems mostly within 1 kyr (well within 2σ uncertainty of DF2021). The excellent agreement suggests high accuracy of our chronology, and supports the assumption of negligible phasing between the past local summer solstice insolation and δO2/N2 fractionation at bubble close-off (the basis for constructing the δO2/N2 age markers). Between 130 and 207 kyr BP (penultimate glacial period), there is a lower degree of similarity between the variations in atmospheric CH4 and speleothem calcite δ18O than in the last glacial period, making the age comparison challenging. The comparison of DF2021 with 9 U–Th dates at 7 abrupt events shows the mean difference of −0.2 ± 0.8 kyr, which is within the DF2021 uncertainty (on the order of 1.5 kyr). The DF2021 chronology agrees with the AICC2012 chronology within 2 kyr except between ∼103 and 128 kyr BP where AICC2012 is likely too young by up to ∼4 kyr. By analyzing the lag between δ18O of O2 (δ18Oatm) on DF2021 and 65°N summer solstice insolation, the relatively large error in AICC2012 is found to originate in three δ18Oatm age markers assuming a constant lag, which are off by more than 3 kyr. The phasing between δ18Οatm and orbital forcing identified in this study may be useful for future ice core dating, especially where other age constraints are weak or lacking.

Denudation of the Cordillera and intraplate belt in central Patagonia inferred by detrital multi-dating of foreland basin deposits

The evolution of central Patagonia is associated with episodic shortening and extension that have greatly affected the topography of the Cordillera and intraplate belt. The San Jorge Basin is a site of sediment accumulation in the foreland that is surrounded by igneous and broken foreland relief. The latter originated from episodic deformation and reactivation of inherited structures associated with a period of slab shallowing that allowed the far-field transmission of Andean stresses to the foreland. Thus, due to of its location, the San Jorge Basin provides an exceptional opportunity to study the denudation of both Cordilleran and intraplate topography during the Cenozoic, particularly during the late Eocene–early Miocene interval of mild deformation. In this study, we use a single-grain geochronological approach combining apatite fission tracks and UPb dating on apatite, along with maximum deposition ages obtained from UPb zircon dating for Neogene foreland basin deposits, to distinguish between two distinctive sediment source regions in central Patagonia during the Cenozoic, despite the persistent volcanic signal. A compilation of previously published cooling ages combined with our new data define: (i) a dominant local source from the northern broken foreland from the late Eocene until the early Miocene; and (ii) a widespread source in the Cordillera during the Miocene based on a very uniform thermochronological signal observed throughout the entire foreland. Therefore, this study provides new insights into the variation in sediment sourcing in the central Patagonian foreland. This variation is primarily controlled by the decrease of post-orogenic erosional processes during a period of relative tectonic quiescence following the highly active Early Cretaceous–middle Eocene time interval.

Drivers of Last Millennium Antarctic Climate Evolution in an Ensemble of Community Earth System Model Simulations

Improved understanding of the drivers of climate variability, particularly over the last millennium, and its influence on Antarctic ice melt have important implications for projecting ice sheet resilience in a changing climate. Here, we investigated the variability in Antarctic climate and sea ice extent during the last millennium (850–1850 CE) by comparing paleoenvironmental reconstructions with simulations from the Community Earth System Model Last Millennium Ensemble (CESM-LME). Atmospheric and oceanic response to external forcing in CESM-LME simulations typically take the form of an Antarctic dipole: cooling over most of Antarctica and warming east of the Antarctic Peninsula. This configuration is also observed in ice core records. Unforced variability and a dipole response to large volcanic eruptions contribute to weaker cooling in the Antarctic than the Arctic, consistent with the absence of a strong volcanic signal in Antarctic ice core records. The ensemble does not support a clear link between the dipole pattern and baseline shifts in the Southern Annular Mode and El Niño-Southern Oscillation proposed by some paleoclimate reconstructions. Our analysis provides a point of comparison for paleoclimate reconstructions and highlights the role of internal climate variability in driving modeled last millennium climate evolution in the Antarctic.

Volcanic stratospheric injections up to 160 Tg(S) yield a Eurasian winter warming indistinguishable from internal variability

Abstract. Early observational and modeling work suggested that low-latitude volcanic eruptions, comparable to the one of Pinatubo in 1991 or Krakatau in 1883, cause substantial surface warming over the northern continents at mid-latitudes in winter. The proposed mechanism consists of the formation of an anomalously strong Equator-to-pole temperature gradient in the stratosphere due to the presence of volcanic aerosols in the tropics, which are accompanied by an acceleration of the stratospheric polar vortex, which then shifts the Northern Annular Mode into a positive phase, resulting in warming surface temperatures over Eurasia. However, a large body of research in the past decade has shown that, for eruptions such as Pinatubo or Krakatau, no such warming is seen in simulations with more recent climate models which, in general, have much finer vertical and horizontal resolution than the early ones, and which have separated the forced response from the internal variability by using large ensembles of integrations. Since the proposed physical mechanism is sound, it is then possible that eruptions comparable to those of Pinatubo or Krakatau are simply too weak, but even larger ones might indeed be capable of causing Eurasian surface warming in winter. In this study, we explore this possibility using a state-of-the-art, stratosphere-resolving climate model, forced with prescribed aerosols from the Easy Volcanic Aerosol protocol. We consider eruptions with stratospheric sulfur injections of 5, 10, 20, 40, 80, and 160 Tg(S). With 20-member ensembles, we find that with injections of 20 Tg(S) or more – roughly twice the amplitude of the Pinatubo and Krakatau eruptions – our model simulates a winter surface warming over Eurasia, which is statistically significant with a t test given our 20-member ensembles. However, the forced volcanic signal on Eurasian winter surface temperatures is very small, barely exceeding the 1σ range of internal variability for the 160 Tg(S) injection case, and much smaller for smaller eruptions. Most importantly, the number of eruptions needed to establish statistical significance is considerably larger than the number of eruptions known to have occurred in the past 2000 years.

Identifying the Fingerprint of a Volcano in the Background Seismic Noise from Machine Learning-Based Approach

This work is devoted to the analysis of the background seismic noise acquired at the volcanoes (Campi Flegrei caldera, Ischia island, and Vesuvius) belonging to the Neapolitan volcanic district (Italy), and at the Colima volcano (Mexico). Continuous seismic acquisition is a complex mixture of volcanic transients and persistent volcanic and/or hydrothermal tremor, anthropogenic/ambient noise, oceanic loading, and meteo-marine contributions. The analysis of the background noise in a stationary volcanic phase could facilitate the identification of relevant waveforms often masked by microseisms and ambient noise. To address this issue, our approach proposes a machine learning (ML) modeling to recognize the “fingerprint” of a specific volcano by analyzing the background seismic noise from the continuous seismic acquisition. Specifically, two ML models, namely multi-layer perceptrons and convolutional neural network were trained to recognize one volcano from another based on the acquisition noise. Experimental results demonstrate the effectiveness of the two models in recognizing the noisy background signal, with promising performance in terms of accuracy, precision, recall, and F1 score. These results suggest that persistent volcanic signals share the same source information, as well as transient events, revealing a common generation mechanism but in different regimes. Moreover, assessing the dynamic state of a volcano through its background noise and promptly identifying any anomalies, which may indicate a change in its dynamics, can be a practical tool for real-time monitoring.

Fiber-Optic Observation of Volcanic Tremor through Floating Ice Sheet Resonance

AbstractEntirely covered by the Vatnajökull ice cap, Grímsvötn is among Iceland’s largest and most hazardous volcanoes. Here we demonstrate that fiber-optic sensing technology deployed on a natural floating ice resonator can detect volcanic tremor at the level of few nanostrain/s, thereby enabling a new mode of subglacial volcano monitoring under harsh conditions. A 12.5 km long fiber-optic cable deployed on Grímsvötn in May 2021 revealed a high level of local earthquake activity, superimposed onto nearly monochromatic oscillations of the caldera. High data quality combined with dense spatial sampling identify these oscillations as flexural gravity wave resonance of the ice sheet that floats atop a subglacial lake. Although being affected by the ambient wavefield, the time–frequency characteristics of observed caldera resonance require the presence of an additional persistent driving force with temporal variations over several days, that is most plausibly explained in terms of low-frequency volcanic tremor. In addition to demonstrating the logistical feasibility of installing a large, high-quality fiber-optic sensing network in a subarctic environment, our experiment shows that ice sheet resonance may act as a natural amplifier of otherwise undetectable (volcanic) signals. This suggests that similar resonators might be used in a targeted fashion to improve monitoring of ice-covered volcanic systems.

Do Southern Hemisphere tree rings record past volcanic events? A case study from New Zealand

Abstract. Much of our knowledge about the impacts of volcanic eruptions on climate comes from proxy records. However, little is known about their impact on the low to mid-latitudes of the Southern Hemisphere. Using superposed epoch analysis, we investigated whether volcanic signals could be identified in annual tree-ring series from eight New Zealand dendrochronological species. We found that most species are reliable recorders of volcanic cooling and that the magnitude and persistence of the post-event response can be broadly linked to plant life history traits. Across species, site-based factors, particularly altitude and exposure to prevailing conditions, are more important determinants of the strength of the volcanic response than species. We then investigated whether chronology selection impacts the magnitude of post-volcanic cooling in tree-ring-based temperature reconstructions by developing two new multispecies reconstructions of New Zealand summer (December–February) temperature with one reconstruction from the pool of all available chronologies, and the other from a selected subset shown to be sensitive to volcanic eruptions. Both reconstructions record temperature anomalies that are remarkably consistent with studies based on instrumental temperature and the ensemble mean response of climate models, demonstrating that New Zealand ring widths are reliable indicators of regional volcanic climate response. However, we also found that volcanic response can be complex, with positive, negative, and neutral responses identified – sometimes within the same species group. Species-wide composites thus tend to underestimate the volcanic response. This has important implications for the development of future tree-ring and multiproxy temperature reconstructions from the Southern Hemisphere.

Annually Resolved Profiles of δ34S and Sulfate in Shallow Ice Core DF01 (Dome Fuji, Antarctica) Spanning the Nineteenth Century and Their Geochemical Implications

AbstractWe report the results of a high‐sensitivity sulfur‐isotope (δ34S) analysis of samples from shallow ice core DF01 from Dome Fuji, East Antarctica, and present chronological profiles of the data with an annual resolution covering the entire nineteenth century. In the δ34S and non‐sea‐salt (nss)‐sulfate ([SO42−]nss) profiles, the signals of five well‐known nineteenth century volcanic eruptions were confirmed: Krakatau 1883 (and Tarawera 1886), Makian 1861, Cosiguina 1835, Tambora 1815, and Unknown 1809. Due to large influence of sulfate caused by volcanic eruptions in the first half of the nineteenth century, the averages of nss‐sulfur‐isotope composition (δ34Snss) and [SO42−]nss for the second half (+14.6‰ and 67.6 μg/L), calculated after excluding these prominent signals, were found to differ from those of the first half century (+9.1‰ and 84.1 μg/L). We also examined anomalous variations other than the five major volcanic signals in the δ34S and [SO42−]nss profiles based on not only variations in δ34S and [SO42−]nss but also variations in the concentrations of other ions. We then identified the presence of three additional volcanic signals: Babuyan 1831, Galunggung 1822, and an unknown volcanic eruption around 1807. The latter two signals confirmed for the first time in Antarctic ice cores. Other δ34S and [SO42−]nss anomalies may be related to an influx of coastal air masses, a nonstratospheric volcanic eruption near Antarctica or an increase in the relative contribution of a stratospheric component to the precipitation in inland Antarctica. These demonstrate the importance of obtaining the annually resolved, continuous δ34S profile over a certain period.

Toward Autonomous Detection of Anomalous GNSS Data Via Applied Unsupervised Artificial Intelligence

Artificial intelligence applications within the geosciences are becoming increasingly common, yet there are still many challenges involved in adapting established techniques to geoscience data sets. Applications in the realm of volcanic hazards assessment show great promise for addressing such challenges. Here, we describe a Jupyter Notebook we developed that ingests real-time Global Navigation Satellite System (GNSS) data streams from the EarthCube CHORDS (Cloud-Hosted Real-time Data Services for the geosciences) portal TZVOLCANO, applies unsupervised learning algorithms to perform automated data quality control (“noise reduction”), and explores autonomous detection of unusual volcanic activity using a neural network. The TZVOLCANO CHORDS portal streams real-time GNSS positioning data in 1[Formula: see text]s intervals from the TZVOLCANO network, which monitors the active volcano Ol Doinyo Lengai in Tanzania, through UNAVCO’s real-time GNSS data services. UNAVCO’s real-time data services provide near-real-time positions processed by the Trimble Pivot system. The positioning data (latitude, longitude and height) are imported into the Jupyter Notebook presented in this paper in user-defined time spans. The positioning data are then collected in sets by the Jupyter Notebook and processed to extract a useful calculated variable in preparation for the machine learning algorithms, of which we choose the vector magnitude for further processing. Unsupervised K-means and Gaussian Mixture machine learning algorithms are then utilized to locate and remove data points (“filter”) that are likely caused by noise and unrelated to volcanic signals. We find that both the K-means and Gaussian Mixture machine learning algorithms perform well at identifying regions of high noise within tested GNSS data sets. The filtered data are then used to train an artificial intelligence neural network that predicts volcanic deformation. Our Jupyter Notebook has promise to be used for detecting potentially hazardous volcanic activity in the form of rapid vertical or horizontal displacement of the Earth’s surface.