Conference Agenda

Askja volcano is situated in the Northern Volcanic Zone in Iceland, and comprises both a central volcano and a fissure swarm covering an area of approximately 190 x 20 km. The central volcano includes a series of nested calderas formed during previous plinian eruptions. Historic eruptions have comprised both basaltic effusive eruptions and silicic explosive eruptions although the former are more common. Eruptions occur on average three times per century. The last eruption at Askja occurred in 1961. This was predominantly effusive and produced a lava field of approx. 0.1 km³. The last plinian eruption to occur here was in 1875. This major event formed the most recent caldera, which is now filled with lake Öskjuvatn (~200 m deep).

At the beginning of August 2021, inflation was detected at Askja volcano, on a continuous GNSS station located to the west of Öskjuvatn (OLAC) and on interferograms generated using data from four separate Sentinel-1 tracks. At the time of writing (March 2023) deformation is continuing, with ~ 50 cm of uplift measured at GNSS station OLAC. Ground deformation measurements at Askja commenced in 1966 with levelling observations, and since this time additional ground monitoring techniques have been employed, including GNSS and Satellite interferometry (InSAR) to detect long-term changes. Ground levelling measurements undertaken between 1966-1972 revealed alternating periods of deflation and inflation. Measurements from 1983-2021 detailed persistent subsidence of the Askja caldera, decaying in an exponential manner. Shortly after the onset of unrest, three additional GNSS stations were installed at Askja and campaign measurements undertaken in summer 2021 and 2022. GNSS time series and InSAR decomposition results indicate that the observed deformation results from upward and lateral migration of magma, potentially feeding multiple shallow sources. This presentation will provide an overview of the GNSS and InSAR observations to date and present the latest geodetic modelling results (including Finite Element models) which describe the best-fit source for the observed deformation. Future eruptive scenarios will also be discussed, based on the location of the intruded magma and historic activity.

Mauna Loa volcano, Hawaii, erupted in November 2022 for the first time since 38 years. The eruption was preceded by >20 years of magma intrusion into its dike-like magma body. In this presentation we (1) discuss how magma intrusion prior to the eruption changed the state of stress in the volcanic edifice priming the volcano for an eruption from the northeast rift zone, (2) use geodetic data to derive a magma source model for the processes of the 2022 eruption and for the rapid re-inflation following the eruption, and (3) evaluate stress changes due to the 2022 events along the rift zone and along the sub-horizontal decollement faults along the base of the volcanic edifice. Did the 2022 eruption increase the failure stress under the eastern flank as did previous eruptions form the northeast rift zone?

The region of Chiles and Cerro Negro volcanoes, located on the Colombian-Ecuadorian border, has experienced unrest since 2013. More recently, since May 2022, the area has shown renewed activity with thousands of seismic events recorded per day, and ground deformation velocities up to 9 cm/yr, the largest recorded in the region to date.

At present, the source of unrest is not well-constrained. Preliminary modeling of GNSS and InSAR data show that the deformation could be consistent with the intrusion of a dike. On the other hand, seismic depths do not suggest shallowing of events or strong alignments to mirror a dike emplacement. New deformation data from Sentinel-1 and the expanded GNSS network could help to better bound the deformation source.

Seismic Activity
Since May 2022 geophysical networks of the Instituto Geofísico (Escuela Politécnica Nacional, Quito) and the Colombian Geological Survey have recorded two main seismic swarms in the Chiles-Potrerillos volcanic region. The epicenters of the first swarm were initially located on the southern flank of Chiles volcano, later migrating southastward to the Potrerillos volcanic plateau. Manual and automatic counts yielded a peak of 2000 earthquakes per day between May and December 2022. During the same timeframe, about 1500 earthquakes were located at depths of 2 – 15 km below sea level in the Potrerillos zone, south of Chiles. Volcano–tectonic earthquakes typify the swarms. The largest earthquake recorded in the Chiles swarm occurred on June 21, 2022, and had a magnitude of 4.3 MLv. On July 25, 2022, in the Potrerillos swarm, an earthquake of 5.6 (Mw) ruptured a blind E-W strike-slip fault at 5.0 km depth. This earthquake was widely felt and caused landslides and ground fracturing, as well as ground displacement.

The second swarm began abruptly on March 9, 2023, and is ongoing. The epicenters for this swarm have been located exclusively on the southern flank of Chiles. Manual and automated counts tally over five thousand events per day. Occasionally events were felt in nearby communities. Overall, since May 2022, approximately 11500 seismic events were recorded in at least two seismic stations on the southern flank of Chiles volcano, with depths of 4 - 7 km below the summit (4600m).

Ground deformation
We detect no appreciable ground deformation until May 2022, as registered by two principal continuous GNSS stations. Then, a northward trending motion with an initial deformation velocity of 4 cm/yr began. The station most affected, CHLS, lies near the SE base of Chiles volcano and is strongly trending northward, and more so with the actual swarm´s onset. Deformation continues to the present with increased deformation velocities up to 9 cm/yr being recorded at two new GNSS stations, TITS and TOAL, installed in August 2022 on the Potrerillos plateau.

Processing of Sentinel-1 ascending and descending tracks and their decomposition into vertical and E-W components shows time-series results that are very similar to GNSS data. Particularly on the vertical component, we observe deformation with a velocity of 5-6 cm/yr, at the southern foot of Chiles and at Potrerillos plateau.

The Chiles-Potrerillos area is crossed by several sets of NE-SW trending transcurrent faults, some of which are Holocene-age. The July 25 2022 earthquake had a pronounced and uncommon E-W rupture with an offset of 0.31 m, as seen in InSAR results.

Preliminary Modeling

Preliminary modeling of GNSS and InSAR data show that the deformation could be consistent with the intrusion of a dike. Unfortunately, the sources inferred by modeling these two data sets are somewhat different and may be compromised by the then lack of GNSS data for the Potrerillos Plateau.

While we suspect that a dike could be intruding, seismic depths do not suggest shallowing of events or strong alignments to mirror dike emplacement. Alternatively, the excitation of a buried hydrothermal system from stress transfer could cause the persistent uplift, and be the driver of unrest in earlier unrest episodes: 2013-2015, 2018-2019, July 2022, and in the present

Harmony has been selected by ESA as the 10^th Earth Explorer Mission, with an expected launch date in 2029. Comprising of two satellites carrying passive (receive-only) Synthetic Aperture Radar (SAR) instruments as well as Thermal-Infrared Spectrometers (TIR), Harmony will operate in tandem with Sentinel-1 and monitor changes in the Earth’s surface and cryosphere, as well as monitor ocean surface conditions.

Harmony will revolutionise the way we measure the rapid topographic changes associated with volcanic eruptive activity. More than 800 million people across the world live within 100km of a volcano and monitoring is key to mitigating the threat of volcanic eruptions to human life. Maps of surface displacement and topographic change are vital for understanding the geometry and activity of underlying magma storage areas and the stability of steep volcanic edifices. Harmony will provide such high temporal-resolution views of topographic change and yearly DEM updates at actively erupting volcanoes. This will improve the modelling and forecasting of volcanic dome growth, collapse, and emplacement of volcanic flows, all of which can pose significant threat to nearby populations.

As part of the Harmony science studies, we have investigated a number of recent volcanic eruptions and the associated topographic change with the aim of providing both ground-truth topographic change measurements as well as simulating the resolving capabilities of Harmony. Interferometric pairs of TanDEM-X high-resolution SAR images were processed to produce high-resolution digital elevation models (DEM) which were used to assess topographic change after volcanic eruptive activity. These SAR images were then subsampled to simulate the imaging resolution achievable by the Harmony mission, and the subsampled imagery was in turn similarly processed to produce digital elevation models and topographic change maps indicative of Harmony’s capabilities.

Our case studies include a) the 2020-2021 eruption of St. Vincent La Soufriere, where the explosive eruption completely destroyed the lava domes and significantly changed the crater topography; b) the Reventador volcano in Ecuador, which has been continuously erupting since 2008, producing tens of lava flows; and c) the June 2018 explosion of the Fuego volcano in Guatemala, which caused pyroclastic density currents that destroyed the nearby town of San Miguel Los Lotes. These three case studies present variations in both local terrain and volcanic hazard that can provide a broad picture of Harmony’s resolving capabilities.

These studies have allowed us to better understand and quantify the effects of topography on the resolution and accuracy of Harmony’s interferometric measurements. In areas of steep terrain (such as most volcanos), layover and shadow artefacts often occur. Layover manifests as compression/foreshortening along steep slopes facing the radar line of sight, while shadowing refers to the complete lack of a return signal from parts of the terrain obscured from the radar beam’s illumination. These lead to artefacts and erroneous data appearing in the generated interferograms, which in turn complicate and introduce errors during the unwrapping process (whereby phase cycles in the interferogram are re-interpreted as continuous phase change directly translatable to displacement in physical units). Identifying and modelling the effects of local topography on the produced interferograms, as well as identifying the optimal processing and unwrapping techniques for such locales have allowed us to identify areas of particular challenge for Harmony.

Additionally, the Harmony measurements simulated during this project via sub-sampled TanDEM-X data have also been compared to the Harmony end-to-end system simulator developed by the German Aerospace Agency (DLR); this has proven beneficial to the validation and further development of both systems. Finally, the full-resolution TanDEM-X -generated topographic change maps themselves serve as valuable input to the DLR simulator, as it is dependent on ground-truth data of topographic change in order to simulate Harmony measurements.

Volcanoes experience some of the largest and most dynamic topographic changes on Earth, spanning a wide range of spatial and temporal scales. Processes include creating topography through eruption of new lava flows or domes, or removing topography by explosive eruptions, caldera and sector collapse. Volcanic eruptions can also change topography by depositing ash or pyroclastic flows or even between eruptions through remobilization of deposits from lahars and landslides. The highly variable nature of these processes in space and time places stringent constraints on the spatiotemporal requirements for topography data products.

We present an overview of our project to quantify the volcano topography change observational needs for a future NASA Surface Topography Vegetation (STV) observing system. We focus on three linked topics relating topographic change to hazard assessment and mitigation: 1) dynamical models of volcanic eruptions; 2) the effects of topography quality on lava flow dynamics, slope instability and pyroclastic flows; 3) an assessment of candidate observing methods and architectures based on the data analyses and simulations in topics (1) and (2). Study locations and events include: 1) caldera collapse and large basaltic lava flows: 2018 Kīlauea, Hawaii ; 2) intermediate to silicic composition lava flows and dome eruptions: Cordón Caulle and Nevados de Chillán, Chile; Ibu, Indonesia; La Soufrière, St. Vincent, West Indies; and Great Sitkin, Alaska; 3) slope stability: Sinabung, Indonesia; and 4) pyroclastic flows: Fuego, Guatemala. Datasets include TanDEM-X (TDX), GLISTIN-A, EarthDEM, Planet Labs, Pleiades, and locally acquired lidar and photogrammetry.

Dynamic volcano source modeling will build from existing models for connected caldera collapse and lava effusion (Roman and Lundgren, 2021) and lava dome extrusion (Delgado et al., 2019).

Flow simulations will use existing software for lava and pyroclastic flows (e.g., VolcFlow, Kelfoun and Vallejo-Vargas, 2016; DOWNFLOW, Favalli et al., 2005), dome and slope instability (Scoops3D; Reid et al., 2015) as well as advanced physics-based dynamic models. We will use flow thickness to constrain lava flow forecasts, where both the spatial quality and temporal sampling affect model predictions.

We will present published (Lundgren et al., 2019) and preliminary results from the NASA airborne responses to the 2018 Kīlauea and 2022 Mauna Loa eruptions where we acquired single-pass Ka-band bistatic radar using the GLISTIN-A synthetic aperture radars (SARs) onboard the NASA G-III jet. These observations will be used to guide simulation studies on topography resolution on lava flow models.

References:

Delgado, F., Kubanek, J., Anderson, K., Lundgren, P. and Pritchard, M., 2019. Physicochemical models of effusive rhyolitic eruptions constrained with InSAR and DEM data: A case study of the 2011-2012 Cordon Caulle eruption. Earth and Planetary Science Letters, 524, p.115736.

Favalli M., Pareschi M. T., Neri A., Isola I., 2005, Forecasting lava flow paths by a stochastic approach. Geophys. Res. Lett., 32.

Kelfoun, K., and Vallejo-Vargas, S. V. (2016). VolcFlow capabilities and potential development for the simulation of lava flows. Geological Society, London, Special Publications, 426(1), 337-343.

Lundgren, P. R., Bagnardi, M., & Dietterich, H., 2019. Topographic changes during the 2018 Kīlauea eruption from single‐pass airborne InSAR. Geophysical Research Letters, 46. https://doi.org/10.1029/2019GL083501

Reid, M. E., Christian, S. B., Brien, D. L. and Henderson, S. T., 2015. Scoops3D: software to analyze 3D slope stability throughout a digital landscape (N. 14-A1). US Geological Survey.

Roman, A., and Lundgren, P., 2021, Dynamics of large effusive eruptions driven by caldera collapse, Nature, 592, 392-396. https://doi.org/10.1038/s41586-021-03414-5.