Norsar

On Earth, infrasound can travel vast distances through waveguides formed by changes in the sound speed and wind structure. While propagating, these waves retain valuable information about the source and the propagation medium, making infrasound suitable to monitor explosive events, including from volcanic sources. In contrast to Earth, extreme conditions on the surface rule out conventional ground-based sensors for infrasound recording on Venus. A more practical alternative is to use atmospheric balloons equipped with acoustic sensors. Such balloons can be deployed in the middle atmosphere, where conditions allow for longer mission durations. However, to explore the existence of prospective explosive volcanism on Venus, such deployments will only be valuable if there exist waveguides to duct the infrasound from these sources to the balloons. But do such waveguides exist? To address this question, we performed an extensive set of wave-propagation simulations to map the global morphology of infrasound ducting on Venus, with the Venus Climate Database as the atmospheric model. We considered volcanic sources and assessed how effectively infrasound is guided to balloon altitudes. We find persistent waveguides in both zonal and meridional directions, driven by the strong superrotational and subsolar-to-antisolar winds in the middle and upper atmosphere. We find that these waveguides enable ground-to-balloon propagation, indicating that the conditions are suitable for detecting long-range infrasound. Our discovery strengthens the case for future balloon-based missions to Venus.

Human activity can create acoustic emissions capable of traveling many kilometers. Narrow band, “tonal” signatures are particularly distinct, as there are few processes in nature that produce them (e.g., volcanic tremor). Although stratospheric balloon flights over cities commonly record these signals, the plethora of emitters makes any single source difficult to distinguish. Acoustic sensors aboard the TRANSAT balloon flight from Sweden to Canada captured a multi-hour narrowband signal while crossing northern Norway. This distinct, isolated recording indicates a singular emitter capable of projecting sound nearly 40 km in the air and at least 100 km laterally. We describe the signal properties, calculate the detection range of the floating sensor, and constrain possible locations of the emission source. This is a first step towards the detection, characterization, and geolocation of narrowband acoustic signals from the stratosphere using a single free flying sensor. This has implications for characterizing anthropogenic activity on Earth as well as evaluating volcanic activity during proposed balloon missions to Venus.

Abstract. Landslides in glacial and periglacial environments are increasingly affected by climate change, with rapid failures reported in high mountain regions and the Arctic. The complex mechanisms behind these events are often poorly understood due to a lack of dense in situ data. We investigate two slow-moving landslides in Arctic Norway (70° N), the Jettan and Gámanjunni landslides, located about 10 km apart, with Jettan below and Gámanjunni above the permafrost boundary. Using over a decade of multi-physics observations, including geodetic, borehole, seismic, and hydrological data, we examine surface and subsurface deformation. Both landslides display similar seasonal surface velocity patterns, with peaks in spring and autumn, likely influenced by pore-water infiltration. At Jettan, twelve years of inclinometer data in boreholes reveal a transition from steady state to seasonal deformation in two shear zones. Since 2020, spring accelerations have intensified in years coinciding with deeper snowpacks and associated melt. These observations, together with statistical modeling, suggest that the shear-band interface is becoming increasingly localized and sensitive to pore-water pressure. Conversely, autumn acceleration is not seen in localized shear zones but manifests as distributed volumetric deformation. Seismic velocity variations within the landslide body also exhibit seasonal patterns that correspond with GNSS velocity, interpreted as changes in landslide rigidity due to water infiltration. This integrated analysis of surface and subsurface data offers new insights into the evolving deformation of Arctic landslides, emphasizing the influence of hydrological forcings on both seasonal and long-term deformation processes.

Abstract Distributed acoustic sensing (DAS) enables cost‐effective, dense detection of seismic activity. However, the vast amount of data produced by DAS systems presents a significant challenge for labeling and analysis. Traditional supervised machine learning approaches require extensive labeling, which is time‐consuming and prone to user bias. Our approach meets the challenge of reducing the workload required for annotation, regardless of the size of the data set and without any a priori of the content of the data set, while preserving rare seismic events. We propose a two‐step processing chain. The first step constructs a latent data representation from several hundred features. We compare two approaches: one using signal processing metrics commonly used in seismology (human‐engineered features) and the other using self‐supervised learning with common DAS data representations (image‐BYOL, as known as Bootstrap Your Own Latent). The second step applies unsupervised clustering to reduce the data set. We first apply K‐Means to obtain 5,000 clusters followed by hierarchical clustering, merging them into 500–700 clusters using an inconsistency criterion. This dual‐step approach capitalizes on the computational efficiency of K‐Means and the hierarchical granularity offered by agglomerative clustering. This method is applied to DAS data from two experiments in the Pyrenees with different configurations: a 6‐week measurement on an 800‐m cable at Viella and 19 ten‐minute measurements on a 91‐km cable. For Viella, we detected all events with . The image‐BYOL approach produces more false positives than human‐engineered features. The results highlight the potential of clustering for DAS analysis while emphasizing the need to reduce false positives especially for smaller seismic events.

SUMMARY We present and validate an efficient GPU-accelerated solver for seismic wave propagation in 3-D elastic media. The solver achieves up to a 372$\times$ speedup relative to a CPU implementation and supports forward simulations on grids ranging from 100 million to 1 billion cells. It is based on a velocity-stress, first-order formulation of the elastodynamic wave equation and supports kilometre-scale models with layered isotropic and anisotropic structure. We validate the solver by comparing synthetic seismograms to analytical solutions from a propagator matrix method in axisymmetric media. Simulations include moment-tensor sources for a 2017 nuclear explosion and collapse in North Korea, and a magnitude $\sim$4 earthquake near Linthal, Switzerland (2017 March 6). Anisotropic effects for the Swiss event are modelled using rotated orthorhombic stiffness tensors derived from laboratory measurements of gneiss. Projection onto orthorhombic symmetry enables solver compatibility. We find that anisotropy changes waveform polarity, amplitude and phase at near-source stations. Unscaled laboratory values produce polarity reversals, while velocity-rescaled tensors correct them. These results demonstrate the impact of anisotropy on waveform modelling and indicate that simplified 1-D isotropic models may be insufficient for complex crustal settings. We review how structural effects, including anisotropy and 3-D heterogeneity, contribute to transverse-component energy in the 2017 DPRK explosion and discuss implications for seismic source classification.

Distributed Acoustic Sensing (DAS) enables cost-effective, dense detection of seismic activity. However, the vast amount of data produced by DAS systems presents a significant challenge for labeling and analysis. Traditional supervised machine learning approaches require extensive labeling, which is time-consuming and prone to user bias. To address this, we propose a two-step processing chain. The first step constructs a latent data representation from several hundred features. We compare two approaches: one using signal processing metrics commonly used in seismology (human-engineered features) and the other using self-supervised learning with common DAS data representations (image-BYOL). The second step applies unsupervised clustering to reduce the dataset. We first apply K-means to obtain 5000 clusters, followed by hierarchical clustering, merging them into 500–700 clusters using an inconsistency criterion. This method is applied to DAS data from two experiments in the Pyrenees with different configurations: a 6-week measurement on an 800-m cable at Viella and 19 ten-minute measurements on a 91-km cable. For Viella, we detected all events with Mw >2.0. The image-BYOL approach produces more false positives than human-engineered features. The results highlight clustering’s potential for DAS analysis while emphasizing the need to reduce false positives, especially for smaller seismic events.

The Åknes rockslide is located on the slope of a steeply dipping fjord in Norway in the proximity of urban areas, posing a significant hazard due to its potential to trigger a massive tsunami. This study utilizes data from eight vertically aligned borehole geophones and one broadband seismometer on the surface, collected over a period of approximately 22 months. Previous research has demonstrated that passive seismic monitoring, specifically tracking changes in seismic velocities, can provide precursory indicators of landslide failure. This study aims to assess the potential of this method for monitoring and identifying seasonal patterns in the subsurface properties of the slope. To achieve this, we perform seismic interferometry on various frequency bands to calculate relative seismic velocity changes near the borehole and broadband station.By integrating meteorological data from the study area, we can relate these velocity variations to environmental factors. Our analysis indicates that measurements from borehole sensors demonstrate a positive correlation between temperature and seismic velocity changes during snow-covered months, and a negative correlation during the summer, highlighting the sensitivity of seismic waves to seasonal changes and therefore different environmental regimes. Additionally, results from the broadband sensor reveal a clear decrease in seismic velocities during the melting period, and an increase in seismic velocities with increased precipitation and the reemergence of snow cover, suggesting the seismic velocities being influenced by changes in the water content. These findings advance our understanding of the relationship between calculated relative velocity changes and their connection to complex environmental interactions. This is essential for incorporating seismic velocity monitoring as a tool for assessing the stability of the Åknes slope.

Accurately modeling transmission loss is essential for a variety of applications, such asimproving atmospheric data assimilation for numerical weather prediction, assessing attenuationmaps of sources of interest, or estimating detection capabilities of the International MonitoringSystem infrasound network. However, the high computational cost of numerical modeling solversmakes them impractical for a near-real-time analysis. To address this, a previous study trained aConvolutional Neural Network on regional wind fields, predicting transmission losses in less than 0,05seconds with a mean-squared error of 5 dB. However, this approach uses interpolated atmosphericspecifications and focused only on winds, limiting its applicability for long-range modeling. In thiswork, we develop a convolutional recurrent network to predict transmission losses leveragingrealistic, range-dependent atmospheric specifications combining horizontal winds and temperatures,including small-scale perturbations. The resulting model reaches an error of 4 dB while extendingpropagation range up to 4,000 km and providing epistemic and data uncertainty estimates. Firststudies of such an algorithm on regional scaled events (Tonga-Hunga eruption, Hukkakero explosions,etc.) were performed to further evaluate the model. Predicted attenuation are compared withresults from an alternative regionally fine-tuned neural network. The model also demonstrated itsability to adapt to new frequencies.

Seismic waves can couple to the atmosphere and propagate as acoustic waves, including infrasound at frequency below 20 Hz. Seismically generated infrasound can be recorded by ground-based microbarometers, but also at higher-altitude by pressure sensors carried by balloons. Balloon-borne acoustic observations could be the key to exploring Venus' interior, as surface conditions do not allow for the deployment of seismometers. However, it remains unclear how much information about the subsurface is contained in seismically generated infrasound.   In this contribution we use the recent earthquake-induced acoustic observations from a balloon network on Earth belonging to the Strateole2 campaign, following the 2021 Mw 7.3 earthquake in the Flores Sea, to invert for subsurface velocities. Seismic infrasound signals show body wave arrivals and surface wave dispersion similar to pure seismic signals recorded on the ground. Thus, beyond their detection capability, balloon infrasound also enables the use of classical inversion techniques to retrieve source and subsurface properties. We develop an inversion framework to jointly retrieve earthquake source location and seismic velocities of the subsurface based on arrival time measurements for P, S and Rayleigh waves at multiple balloon stations. We apply this approach to the Flores earthquake using data from four Strateole2 balloons.   The inversion results are the probability density distribution of the seismic source location and of the subsurface velocities in a layered model. Both the resulting location and subsurface model are in good agreement with those obtained from traditional seismic data, confirming balloon seismology as a credible alternative for the seismic exploration of the Earth and other celestial bodies with atmospheres, such as Venus.