Articles | Volume 14, issue 2
Nat. Hazards Earth Syst. Sci., 14, 247–257, 2014
https://doi.org/10.5194/nhess-14-247-2014
Nat. Hazards Earth Syst. Sci., 14, 247–257, 2014
https://doi.org/10.5194/nhess-14-247-2014

Research article 14 Feb 2014

Research article | 14 Feb 2014

Modeling of fast ground subsidence observed in southern Saskatchewan (Canada) during 2008–2011

S. V. Samsonov1, P. J. González2, K. F. Tiampo2, and N. d'Oreye3,4 S. V. Samsonov et al.
  • 1Natural Resources Canada, 588 Booth Street, Ottawa, ON K1A0Y7, Canada
  • 2Department of Earth Sciences, University of Western Ontario, London, Ontario, Canada
  • 3European Center for Geodynamics and Seismology, Rue Josy Welter 19, 7256 Walferdange, Grand-Duchy of Luxembourg
  • 4National Museum of Natural History, Dept. Geophysics/Astrophysics, Rue Josy Welter 19, 7256 Walferdange, Grand-Duchy of Luxembourg

Abstract. Fast ground subsidence in southern Saskatchewan (Canada) between the city of Saskatoon and Rice Lake was observed with the RADARSAT-2 interferometric synthetic aperture radar (InSAR) during 2008–2011. We collected 23 ascending Multi-Look Fine 3 Far (MF3F) and 15 descending Standard 3 (S3) RADARSAT-2 images and performed time-series analysis utilizing Small Baseline Subset (SBAS) and Multidimensional SBAS (MSBAS) methodologies. We observed two well-defined circular regions located a few kilometers apart and subsiding with the nearly constant rate of about 10 cm yr-1. MSBAS decomposition revealed the presence of both vertical and horizontal ground displacements. For further analysis we selected two highly coherent interferograms spanning from November to December 2009 until April 2010 thanks to particularly favorable ground conditions that displayed superior coherence. We performed modeling and inversion assuming spherical and sill source models in order to determine the source location, depth and strength. The sill source model produced the smallest residual of 0.7 cm yr-1 applied to ascending interferograms and 0.9 cm yr-1 applied to descending interferograms. A residual of 1.0 cm yr-1 was achieved with the sill model when both ascending and descending interferograms were used. This model suggested sources located at 1.3 and 1.2 km depth with radius of 1.0 and 1.3 km for eastern and western areas, respectively. The spherical model suggested slightly shallower sources located at 0.9 and 0.8 km. We could not precisely identify the cause of this deformation, but the observed subsidence rate and source depth suggest mining-related origin. Topographic changes produced by this subsidence rate over a long time may produce shallow groundwater redistribution and flooding of agricultural lands.

Download
Altmetrics
Final-revised paper
Preprint