1.5 2 2.5 Elevation (km) 1598 1596 1594 North (km) 1592 1590 1588 East (km) 762 760 758 756 4. InSAR Modeling of the January – March 2014 event 4.1 Mixed Boundary Element Method (MBEM) [5] • The shape, depth and overpressure of potential sources of the observed deformation (Fig. 7) were obtained with a 3D-MBEM model combining the Direct and Displacement Discontinuity Methods [5] • Noisy areas were masked out (Fig. 7) • These bodies as well as topography were modeled with triangular elements [7] • Initial models include a single dike, breaching the surface along a linear fissure connecting the 2014 eruptive vents. [1] Assumptions: The medium is linearly elastic, homogeneous, isotropic. A Young’s Modulus of 50 GPa and a Poisson’s ratio of 0.25 are used for all depths [5] 4.2 Monte-Carlo Inversion [6] The Monte-Carlo inversion calculates the fit between the observed InSAR data and the LOS displacement models produced through MBEM and constrains the best fitting model that minimizes the following misfit function: Χ " = ( & − ( ) * , -. ( & − ( ) where & and ( are vectors of observed and modeled LOS displacements and C D is the covariance matrix [6] Fig. 7: Wrapped RSAT2 interferogram spanning January 8 th 2014 - March 21 st 2014, with mask, left, and without, right. The surface expression of the dike is traced in green. 755 756 757 758 759 760 761 East (km) 1587 1588 1589 1590 1591 1592 North (km) 4.3 Preliminary Inversion Results A joint NA+MBEM approach [7] was used to invert for the dip, bottom elevation, bottom length and distance to top of a preliminary single dike, providing a 60% misfit. Fig. 8 shows the best-fitting preliminary dike result. Fig. 9 depicts the model residuals. Fig. 9: LOS modeled deformation and residual signal not explained by this model (60% misfit) Fig. 8: Best-fitting dike with dip = 105, bottom elevation = 1628 m above sea-level, bottom length = 0.065 and distance to top = 741 m . The topographic mesh for Pacaya is overlain in grey. One dike alone does not fit the InSAR data satisfactorily – more sources should be investigated 755 756 757 758 759 760 761 East (km) 1587 1588 1589 1590 1591 1592 North (km) Investigating Deformation Behavior at Pacaya Volcano, Guatemala, through InSAR Time-series Analysis and 3D Mixed Boundary Element Modeling Judit Gonzalez Santana 1 , Christelle Wauthier 1,2 [email protected] 1 Department of Geosciences and 2 Institute for CyberScience, The Pennsylvania State University Key Questions: • Is surface deformation at Pacaya continuous or episodic? • Are there temporal links between deformation and magmatic events? • What are the potential sources of surface deformation? Fig. 1: Location and geological setting of Pacaya Volcano (red triangle). Pacaya lies at the intersection of the extensional GCG (Guatemala City Graben) (8 mm/yr E-W extension) and the right-lateral JFZ (Jalpatagua Fault Zone) (10-14 mm/yr right-lateral motion) and South of the MFZ (Motagua Fault Zone) and PFZ (Polochic Fault Zone) 1. Geological Setting and Motivation • Pacaya is an active basaltic stratovolcano with an unstable SW flank [1], located in Guatemala (Fig.1) • Overall stress regime is ENE - WSW directed tension. [2] • 9000 people live <5km away from the active cone [2] • SW flank collapse produced a 0.65 km 3 debris avalanche between 0.6 - 1.6 ka [2] • An eruption in 2010 was accompanied by 4m of flank displacement [4] • RSAT2 radar satellite data from January-March 2014 show flank displacement. 92˚W 91˚W 90˚W 89˚W 88˚W 13˚N 14˚N 15˚N 16˚N 17˚N 18˚N Pacaya Guatemala Mexico Belize Honduras El Salvador PFZ MFZ JFZ GCG Carribean Plate Cocos Plate North American Plate Middle America Trench 3. InSAR Time-series Analysis Interferometric Synthetic Aperture Radar (InSAR) uses the difference in phase between two radar images to determine cm scale surface deformation. InSAR images show the change in distance of the Earth’s surface in the Line Of Sight (LOS) of the satellite recorded in between the two image acquisitions (Fig.2) Jul 14 Feb 15 Aug 15 Mar 16 Sep 16 Apr 17 Nov 17 Date -250 -200 -150 -100 -50 0 50 100 150 200 250 Perpendicular Baseline (m) Baseline plot - RSAT2 (Asc) Sep 10 Apr 11 Oct 11 May 12 Nov 12 Jun 13 Jan 14 Jul 14 Feb 15 Aug 1 Date -200 -150 -100 -50 0 50 100 150 200 250 Perpendicular Baseline (m) Baseline plot - RSAT2 (Des) Fig. 2: How InSAR works (by Conway) Fig. 3: Plot of temporal and spatial baselines < 200 days and < 300 m. Fig. 4: Plot of temporal and spatial baselines < 180 days and < 150 m. Datasets: • 35 descending scenes (September 2010 – July 2015) (Fig.3) • 52 ascending scenes (June 2014 – November 2017) (Fig. 4) Acknowledgements and References: This work was supported by NASA grant NNX16AK87G issued through the Science Mission Directorate’s Earth Science Division. RSAT-2 SAR data was provided by The Canadian Space Agency, the European Space Agency (Third Party Mission Projects #16819 and #28777) and the Committee on Earth Observation Satellites (CEOS) Volcano Pilot Program. The 3D-MBEM code was developed by Cayol and Cornet (1997), the NA inversion by Sambridge (2001) and coupled into a NA+MBEM approach by Fukushima et al. (2005). SBAS scripts were developed and provided by Dr. Susanna Ebmeier. 1. Wnuk and Wauthier (2017) Surface deformation induced by magmatic processes at Pacaya Volcano, Guatemala revealed by InSAR. J. Volcanol. Geotherm. Res., 344, 197-211 2. Schaeffer et al. (2013) An integrated field numerical approach to assess slope stability hazards at volcanoes: the example of Pacaya, Guatemala. Bull. Volcanol., 75:720 3. Wunderman and Rose (1984) Amatitlan, an actively resurging cauldron 10km south of Guatemala City. J. Geophys. Res., 89:8525-8539 4. Schaeffer et al. (2017) Three-dimensional displacements of a large volcano flank movement during the May 2010 eruptions at Pacaya Volcano, Guatemala. Geophys. Res. Lett., 44, 135-142 5. Cayol and Cornet (1997) 3D mixed boundary elements for elastostatic deformation field analysis. Int. J. Rock Mech. Min. Sci., 34(2):275-287 6. Sambridge (1999) Geophysical inversion with a neighbourhood algorithm – I. Searching a parameter space. Geophys. J. Int., 138, 479-494 7. Fukushima and Cayol (2005) Finding realistic dike models from interferometric synthetic aperture radar: The February 2000 eruption at Piton de la Fournaise. J. Geophys. Res., 110, B03206 8. Hooper (2008) A multi-temporal InSAR method incorporating both persistent scatterer and small-baseline approaches, Geophys. Res. Lett., 35, L16, 302 5. Current and Future Work: • SBAS time-series analysis of the ascending data – implementing atmospheric corrections to obtain a clearer signal: • Addition of further potential sources of deformation with more complex geometries to the MBEM model – eg. detachment fault on the SW flank. • InSAR modelling of 2010 deformation events using ALOS-1 and UAVSAR data • Examining more recent surface deformation with Sentinel-1 • Pixel offset tracking for the observed deformation events in 2010 and 2014 2. What is InSAR? Preliminary SBAS processing was performed for the descending dataset. Fig. 5 shows the cumulative ground displacement between 26 th September 2010 and 9 th April 2015 relative to the reference point (red circle). Fig. 6 shows the cumulative LOS displacement at the yellow square labelled in Fig. 5 relative to the reference point. Flank motion between 2010 and 2015 appears to follow a semi-continuous LOS subsidence trend, before and after the 2014 eruptive event Fig. 6: Cumulative LOS displacement plot at the yellow square relative to the red circle in Fig. 5, showing ~4.23 cm/yr subsidence between 2010 and 2015. The red box shows the dates over which the 2014 deformation event modeled in section 3 was recorded. Fig. 5: Map of the cumulative ground displacement between September 2010 and April 2015 relative to the reference point (red circle). The orange square is the location chosen for the time-series plot in Fig. 6 Cumulative LOS displacement Jan-10 Jan-11 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Year -25 -20 -15 -10 -5 0 Cumulative LOS displacement (cm) LOS