The application of Convolutional Neural Networks to Detect Slow, Sustained Deformation in InSAR Timeseries

Automated systems for detecting deformation in satellite InSAR imagery could be used to develop a global monitoring system for volcanic and urban environments. Here we explore the limits of a CNN for detecting slow, sustained deformations in wrapped interferograms. Using synthetic data, we estimate a detection threshold of 3.9cm for deformation signals alone, and 6.3cm when atmospheric artefacts are considered. Over-wrapping reduces this to 1.8cm and 5.0cm respectively as more fringes are generated without altering SNR. We test the approach on timeseries of cumulative deformation from Campi Flegrei and Dallol, where over-wrapping improves classication performance by up to 15%. We propose a mean-filtering method for combining results of different wrap parameters to flag deformation. At Campi Flegrei, deformation of 8.5cm/yr was detected after 60days and at Dallol, deformation of 3.5cm/yr was detected after 310 days. This corresponds to cumulative displacements of 3 cm and 4 cm consistent with estimates based on synthetic data.

[1]  Claudio Prati,et al.  A New Algorithm for Processing Interferometric Data-Stacks: SqueeSAR , 2011, IEEE Transactions on Geoscience and Remote Sensing.

[2]  J. Kendall,et al.  Geophysical Monitoring of Moisture‐Induced Landslides: A Review , 2019, Reviews of Geophysics.

[3]  Pablo J. González,et al.  Towards InSAR Everywhere, All the Time, With Sentinel-1 , 2016 .

[4]  Matthew Wilks,et al.  Evidence for cross rift structural controls on deformation and seismicity at a continental rift caldera , 2018 .

[5]  Denis Legrand,et al.  Persistent uplift of the Lazufre volcanic complex (Central Andes): New insights from PCAIM inversion of InSAR time series and GPS data , 2014 .

[6]  Tamsin A. Mather,et al.  Applicability of InSAR to tropical volcanoes: insights from Central America , 2013 .

[7]  M. G. Di Giuseppe,et al.  Ground deformation at calderas driven by fluid injection: modelling unrest episodes at Campi Flegrei (Italy) , 2011 .

[8]  Gianfranco Fornaro,et al.  A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms , 2002, IEEE Trans. Geosci. Remote. Sens..

[9]  David Bull,et al.  Application of Machine Learning to Classification of Volcanic Deformation in Routinely Generated InSAR Data , 2018, Journal of Geophysical Research: Solid Earth.

[10]  K. Mogi.  Relations between the Eruptions of Various Volcanoes and the Deformations of the Ground Surfaces around them , 1958 .

[11]  Matthew E. Pritchard,et al.  Global Volcano Monitoring: What Does It Mean When Volcanoes Deform? , 2017 .

[12]  Tom Simkin,et al.  Volcanoes of the World , 2011 .

[13]  G. P. Ricciardi,et al.  Unrest episodes at Campi Flegrei: A reconstruction of vertical ground movements during 1905-2009 , 2010 .

[14]  Zhenhong Li,et al.  Generic Atmospheric Correction Model for Interferometric Synthetic Aperture Radar Observations , 2018, Journal of Geophysical Research: Solid Earth.

[15]  Stefania Usai,et al.  A least squares database approach for SAR interferometric data , 2003, IEEE Trans. Geosci. Remote. Sens..

[16]  Antonio Pepe,et al.  The 2004–2006 uplift episode at Campi Flegrei caldera (Italy): Constraints from SBAS‐DInSAR ENVISAT data and Bayesian source inference , 2008 .

[17]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[18]  H. Zebker,et al.  A new method for measuring deformation on volcanoes and other natural terrains using InSAR persistent scatterers , 2004 .

[19]  Nantheera Anantrasirichai,et al.  A deep learning approach to detecting volcano deformation from satellite imagery using synthetic datasets , 2019, Remote Sensing of Environment.

[20]  Karsten Spaans,et al.  LiCSAR: Tools for automated generation of Sentinel-1 frame interferograms , 2016 .

[21]  Jean-Luc Froger,et al.  Revised interpretation of recent InSAR signals observed at Llaima volcano (Chile) , 2015 .

[22]  Tim J. Wright,et al.  Dike‐fault interaction during the 2004 Dallol intrusion at the northern edge of the Erta Ale Ridge (Afar, Ethiopia) , 2012 .

[23]  Науки о Земле.  Global Volcanism Program , 2010 .

[24]  Olaf Hellwich,et al.  Towards Global Volcano Monitoring Using Multisensor Sentinel Missions and Artificial Intelligence: The MOUNTS Monitoring System , 2019, Remote. Sens..

[25]  Matthew E. Pritchard,et al.  Time-dependent deformation of Uturuncu volcano, Bolivia, constrained by GPS and InSAR measurements and implications for source models , 2017 .

[26]  T. Wright,et al.  Multi-interferogram method for measuring interseismic deformation: Denali Fault, Alaska , 2007 .

[27]  Helen Ashton,et al.  Volcanoes of the World (3rd edition) , 2012 .

[28]  Pierre Briole,et al.  Volcano‐wide fringes in ERS synthetic aperture radar interferograms of Etna (1992–1998): Deformation or tropospheric effect? , 2000 .

[29]  D. Shelly,et al.  Renewed inflation of Long Valley Caldera, California (2011 to 2014) , 2015 .

[30]  Howard A. Zebker,et al.  Phase unwrapping for large SAR interferograms: statistical segmentation and generalized network models , 2002, IEEE Trans. Geosci. Remote. Sens..

[31]  Paul L. Younger,et al.  A meta-analysis of coal mining induced subsidence data and implications for their use in the carbon industry , 2018 .

[32]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[33]  D. Schmidt.  Time-dependent land uplift and subsidence in the Santa Clara Valley , 2003 .