Satellite-based InSAR images have the potential to detect volcanic deformation prior to eruptions, but while a vast number of images are routinely acquired, only a small percentage contain volcanic deformation events. Manual inspection could miss these anomalies, and an automatic system modelled with supervised learning requires suitably labelled datasets. To tackle these issues, this paper explores the use of unsupervised deep learning on InSAR images for the purpose of identifying volcanic deformation as anomalies. We test three different state-of-the-art architectures, one convolutional neural network (PaDiM - Patch Distribution Modeling) and two generative models (GANomaly and DDPM). We propose a preprocessing approach to deal with noisy and incomplete data points. We further improve the performance of PaDiM by using a weighted distance, assigning greater importance to features from deeper layers. The final framework was tested with four different volcanoes, which have different characteristics and its performance was compared against an existing supervised learning method for volcanic deformation detection. The experiments show that our final anomaly detection outperforms the supervised learning, particularly where the characteristics of deformation are unknown. Our framework can thus be used to identify deformation at volcanoes without needing prior knowledge about the deformation patterns present there.

Estimates of the percentage of moderate to large crustal earthquakes (mainshocks) that have foreshocks (the foreshock rate) vary widely: recent estimates in Southern California using an enhanced catalog range between 19 and 72%. Enhanced catalogs seem to reveal more foreshocks, possibly providing new constraints on nucleation mechanisms, but precise, commonly-accepted foreshock definitions are lacking. To investigate the observed range we quantify the sensitivity of foreshock rates to mainshock selection method, catalog (standard and enhanced), foreshock definition, geographical restriction and magnitude cut-offs. We compare two foreshock definitions: type A - any earthquakes above a magnitude threshold in a space-time window; and type B - an earthquake count in a space-time window that exceeds the 99th percentile of a statistical representation of past seismicity rates (using three distributions: Poisson, Gamma and Empirical). Foreshock rate estimates are increased by (in order of influence): Poisson distribution, type A definition, fixed mainshock selection, and restricting to mainshocks with minimum background rates or spatial completeness magnitudes. Rates are lowered by: magnitude-dependent mainshock selection, Gamma and Empirical distributions, and applying a magnitude cut-off. A large increase in foreshock rate between the standard and enhanced catalog is only observed when using Poisson distributed background rates for type B foreshocks. A lower magnitude of completeness may thus not lead to significantly more mainshocks with detected foreshocks. Our preferred method, using a more robust mainshock selection and quality-controlled data, estimates ~25% of M4+ “mainshocks” in Southern California have foreshocks.

East Africa hosts significant reserves of untapped geothermal energy. Most exploration has focused on geologically young (<1 Ma) silicic caldera volcanoes, yet there are many sites of geothermal potential where there is no clear link to an active volcano. The origin and architecture of these systems is poorly understood. Here, we combine remote sensing and field observations to investigate a fault-controlled geothermal play located north of lake Abaya in the Main Ethiopian Rift. Soil gas CO2 and temperature surveys were used to examine permeable pathways and showed elevated values along a ~110 m high fault which marks the western edge of the Abaya graben. Ground temperatures are particularly elevated where multiple intersecting faults form a wedged horst structure. This illustrates that both deep penetrating graben bounding faults and near-surface fault intersections control the ascent of hydrothermal fluids and gases. Total CO2 emissions along the graben fault are ~300 t d–1; a value comparable to the total CO2 emission from silicic caldera volcanoes. Fumarole gases show δ13C of –6.4 to –3.8 ‰ and air-corrected 3He/4He values of 3.84–4.11 RA, indicating a magmatic source originating from an admixture of upper mantle and crustal helium. Although our model of the North Abaya geothermal system requires a deep intrusive heat source, we find no ground deformation evidence for volcanic unrest nor recent volcanism. This represents a key advantage over the active silicic calderas that typically host these resources and suggests that fault-controlled geothermal systems offer viable prospects for further exploration and development.

Two of the most widely observed co-eruptive volcanic phenomena - ground deformation and volcanic outgassing - are fundamentally linked via the mechanism of magma degassing and the development of compressibility, which controls how the volume of magma changes in response to a change in pressure. Here we use thermodynamic models (constrained by petrological data) to reconstruct volatile exsolution and the consequent changes in magma properties. Co-eruptive SO2 degassing fluxes may be predicted from the mole fraction of exsolved SO2 that develops in magma whilst stored prior to eruption and during decompression. Co-eruptive surface deformation may be predicted given estimates of erupted volume and the ratio between chamber compressibility and magma compressibility. We conduct sensitivity tests to assess how varying magma volatile content, crustal properties, and chamber geometry may affect co-eruptive deformation and degassing. We find that magmatic H2O content has the most impact on both SO2 flux and volume change (normalised for erupted volumes). Our findings have general implications for typical arc and ocean island volcanic systems. The higher magmatic water content of arc basalts leads to a high pre-eruptive exsolved volatile content, making the magma more compressible than ocean island eruptions. Syn-eruptive gas fluxes are overall higher for arc eruptions, although SO2 fluxes are similar for both settings (SO2 flux for ocean island basalt eruptions is dominated by decompressional degassing). Our models are consistent with observation: deformation has been detected at 48% of ocean island eruptions (16/33) during the satellite era (2005-2020), but only 11% of arc basalt eruptions (7/61).

Satellite radar backscatter has the potential to provide useful information about the progression of volcanic eruptions when optical, ground-based, or radar phase-based measurements are limited. However, backscatter changes are complex and challenging to interpret: explosive deposits produce different signals depending on pre-existing ground cover, radar parameters and eruption characteristics. We use high temporal- and spatial-resolution backscatter imagery to examine the emplacement and alteration of pyroclastic flows, lahars, and ash from the June 2018 eruption of Volcan de Fuego, Guatemala, drawing on observatory reports and rain gauge data to ground truth our observations. We use dense timeseries of backscatter to reduce noise and extract deposit areas. Backscatter decreases where six flows were emplaced on 3 June 2018. In Barranca Las Lajas, we measured a 11.9-km-long flow that altered an area of 6.3 km2; and used radar shadows to estimate a thickness of 10.5 +/- 2 m in the lower sections. The 3 June eruption also changed backscatter over an area of 40 km2, consistent with ashfall. We use transient patterns in backscatter timeseries to identify nine periods of high lahar activity in B. Las Lajas between June and October 2018. We find that the characterisation of subtle backscatter signals associated with explosive eruptions is assisted by (1) radiometric terrain calibration, (2) speckle correction, and (3) consideration of pre-existing scattering properties. Our observations demonstrate that SAR backscatter can capture both the emplacement and subsequent alteration of a range of explosive products, allowing the progression of an explosive eruption to be monitored.

Fracturing and gouge formation absorb ≥50% of earthquake energy on low displacement (<1-2 km) faults in isotropic crust. To assess how these processes absorb earthquake energy in anisotropic crust, we performed field and microstructural investigations on the 110 km long, 0.4-1.2 km displacement, Bilila-Mtakataka Fault (BMF), Malawi. Where the fault is parallel to surface metamorphic fabrics, macroscale fractures define a 5-20 m wide damage zone. This is narrow relative to where the BMF is foliation-oblique (20-80 m), and to faults with comparable displacement in isotropic crust (~40-120 m). There is minimal evidence for cataclasis and microfracturing along the BMF; therefore, despite its 110 km length and geomorphic evidence for MW 7-8 earthquakes, widespread fault zone fracturing has not occurred. We attribute lack of damage to fault growth along shallow and deep-seated pre-existing weaknesses. This conclusion implies that earthquake energy dissipates differently along incipient faults in isotropic and anisotropic crust.

We present the Malawi Active Fault Database (MAFD), a geospatial database of 114 active fault traces in Malawi, and in neighboring Tanzania and Mozambique. The MAFD has been developed from a multidisciplinary dataset: high resolution digital elevation models, field observations, aeromagnetic and gravity data, and seismic reflection surveys from offshore Lake Malawi. Active faults longer than 50 km are found throughout Malawi, where seismic risk is increasing due to its rapidly growing population and its seismically vulnerable building stock. The MAFD also provides an opportunity to investigate the population of normal faults in an incipient continental rift. We find that the null hypothesis that the distribution of fault lengths in the MAFD is described by a power law cannot be rejected. Furthermore, a power-law distribution of faults in Malawi is consistent with its thick seismogenic crust (~35 km), and low (<8%) regional extensional strain that is predominantly (50-75%) accommodated across relatively long hard-linked border faults. Cumulatively, the data and inferences drawn from the MAFD highlight the importance of integrating onshore and offshore geological and geophysical data to develop active fault databases along the East African Rift and similar continental settings, both to understand the regional seismic hazard and tectonic evolution.