Elsevier

Geomorphology

Volume 229, 15 January 2015, Pages 30-44
Geomorphology

Assessing sinkhole activity in the Ebro Valley mantled evaporite karst using advanced DInSAR

https://doi.org/10.1016/j.geomorph.2014.07.035Get rights and content

Highlights

  • Three DInSAR displacement maps were compared with a comprehensive sinkhole inventory.

  • Non-inventoried sinkholes were detected using DInSAR maps.

  • Sinkholes previously classified as inactive were revised as active applying DInSAR.

  • Underestimated sinkhole boundaries were refined utilizing DInSAR data.

  • DInSAR maps overlooked 70% of the previously inventoried active sinkholes.

Abstract

Sinkholes in karst areas may cause subsidence damage in transportation infrastructures, demolition of buildings and even the loss of human lives when they occur in a catastrophic way. Differential Interferometry (DInSAR) is a promising technology for detecting and characterizing sinkholes, as well as for reducing the associated risk when combined with other sources of data such as a sinkhole inventory. In this work, the usefulness of InSAR techniques and data for sinkhole risk management has been analyzed through the comparison of three DInSAR-derived velocity maps with a comprehensive sinkhole inventory in the Ebro Valley, NE Spain. The DInSAR maps have contributed to improve the sinkhole inventory in different ways: (1) detection of non-inventoried sinkholes; (2) revision of sinkhole areas previously classified as inactive as active; and (3) refinement of underestimated sinkhole boundaries. The obtained results suggest that DInSAR products are suitable for analyzing active dissolution-induced subsidence. The application of these techniques may help in recognizing and better characterizing previously unknown karst subsidence problems and in preventing personal and property losses. However, the analysis reveals that the available DInSAR maps combined overlook about 70% of the previously mapped active sinkholes mainly due to decorrelation.

Introduction

Subsidence sinkholes are closed depressions commonly generated by subsurface dissolution of soluble rocks and the downward displacement of the overlying materials and the ground surface (Waltham et al., 2005). Three main subsidence mechanisms may operate independently or jointly: sagging, collapse, and suffusion (Gutiérrez et al., 2008a, Cooper and Gutiérrez, 2013). The settlement of the land surface associated with sinkholes, regardless of the subsidence mechanism, may cause severe damage in any human structure borne by the undermined sediments. For the latest reviews on sinkhole related damage, see the introduction section in Guerrero et al. (2008), Table 1 in Gutiérrez et al. (2009) and supplementary material in Galve et al. (2012). Collapse sinkholes related to the failure of cavity roofs may occur catastrophically and cause personal losses. These rapid sinkholes may directly engulf buildings or people (e.g. De Bruyn and Bell, 2001) and lead to fatal accidents when they form in transportation routes. Previous investigations document slow ground subsidence (creep) preceding sudden collapse, suggesting that the occurrence of these highly dangerous phenomena might be anticipated through the detection of precursory displacements (e.g. Ferretti et al., 2004, Nof et al., 2013).

The avoidance of the existing sinkholes and the areas more prone to new sinkhole occurrences is usually the most cost-effective mitigation option (Cooper and Calow, 1998, Gutiérrez, 2010). This preventive strategy is commonly based on the production of cartographic sinkhole inventories and susceptibility and hazard models derived from them (e.g. Galve et al., 2009a, Galve et al., 2011). However, sinkhole mapping is frequently a difficult task due to factors like the obliteration of the geomorphic expression of the depressions by anthropogenic and natural processes. Moreover, there is frequently a high uncertainty regarding the activity of the existing subsidence sinkholes. It is desirable to discriminate which are the active sinkholes and obtain information on their kinematics: (1) deformation regime (continuous, episodic or a combination of both); (2) subsidence rate; and (3) spatial variability of the deformation pattern, including the precise limits of the area affected by ground motion. The assessment of the activity of sinkholes is usually based on qualitative data, like the freshness vs. degradation state of the depressions, the presence of recent surface deformation features or damaged structures. The application of traditional geodetic methods over large areas is not practical due to the dispersed and localized nature of the karst subsidence process.

Since the early 2000s (e.g. Baer et al., 2002), Differential Interferometric Synthetic Aperture Radar (DInSAR) techniques have opened promising and innovative prospects in sinkhole investigation. The use of SAR interferometry for the study of dissolution-induced subsidence was first applied along the Israeli Dead Sea shores (Derauw and Moxhet, 1996a, Derauw and Moxhet, 1996b, Cornet et al., 1997, Derauw, 1999, Baer et al., 2002). Subsequently, DInSAR studies have also been developed in karst areas of Germany, Italy, Jordan, Spain and USA. These works document: (1) gradual settlement in areas affected by sinkholes (Abelson et al., 2003, Al-Fares, 2005, Castañeda et al., 2009, Castañeda et al., 2011, Rucker et al., 2013); (2) displacements preceding the catastrophic collapse of cavities (Ferretti et al., 2000, Ferretti et al., 2004) or human structures underlain by karstified sediments (Closson et al., 2005, Closson et al., 2010); and (3) the relationships between karst subsidence and its triggering factors, such as earthquake shaking (Closson et al., 2010). Gutiérrez et al. (2011) presented a review of the abovementioned works and a research based on the combination of DInSAR displacement maps with conventional geomorphological methods, Ground Penetrating Radar (GPR) and trenching in the Ebro Valley (NE Spain). Dahm et al. (2010) combined geophysical data sets with InSAR-derived subsidence rates obtained by Schäffer (2009) in Hamburg, Germany. Paine et al. (2012) applied ALOS interferograms to guide site-specific gravimetric investigations over large sinkholes (90 to 200 m in diameter) related to deep-seated dissolution of salt in Texas, USA. The subsidence areas and rates (up to 30 mm/yr) generally matched previous data obtained by geodetic surveys. Since 2007, the availability of high resolution data and the improvement of DInSAR techniques and computing processes make advanced-DInSAR (PS, SBAS) a promising technique even in areas that may pose temporal decorrelation limitations. This paper evaluates the capability of SAR data with different wavelengths and advanced DInSAR techniques to overcome some of the constraints identified in previous works related to sinkhole activity: 1) the general limited coherence in agricultural areas and specially the frequent atmospheric noise in the study area (Castañeda et al., 2011), 2) insufficient spatial resolution to detect most of the active karst features (Castañeda et al., 2009); and 3) the restricted measurable deformation range that determines the identification of the active sinkhole according to their subsidence rates. Three DInSAR velocity maps are compared with a comprehensive sinkhole inventory in a sector of the Ebro Valley to analyze (1) the improvements and limitations of the more recently developed DInSAR products; (2) the contribution of the new data in active sinkhole detection and characterization; and (3) the implications for sinkhole risk management. Additionally, areas affected by subsidence identified using DInSAR that were not previously inventoried were surveyed to semi-quantitatively assess sinkhole-related damage on human structures.

Section snippets

Geological setting

The investigation is focused on the mantled evaporite karst of Zaragoza metropolitan area, located in the middle reach of the Ebro Valley, NE Spain (Fig. 1a and b). The bedrock in this sector of the Ebro Cenozoic Basin consists of sub-horizontally lying evaporites of the Zaragoza Gypsum Formation (Quirantes, 1978). Secondary gypsum is the dominant rock type in outcrop and highly soluble halite and glauberite units up to 70 m and 30 m thick, respectively, occur a few tens of meters beneath the

DInSAR processing

Differential Synthetic Aperture Radar Interferometry (DInSAR) is a microwave remote sensing technique that enables measuring surface displacement with sub-centimeter accuracy by subtracting the topographic phase from SAR interferograms (e.g. Massonnet and Feigl, 1998) by assuming that the scattering characteristics of the ground surface (ground nature and geometry) remain undisturbed (Rosen et al., 2000). Advanced Time-Series DInSAR techniques use two main approaches to measure ground

Assessment of DInSAR velocity maps for sinkhole detection

The main results of the comparative analysis between the three DInSAR velocity maps (Fig. 4b, c and e) and the available sinkhole map (Fig. 4a) are featured below. The results are presented according to a multi-scale approach. A general analysis covering the entire study area (Section 4.1) is followed by a site- and sinkhole-specific analysis (Section 4.2). See Fig. 4a for location of analyzed sites and sinkholes.

Detection of active sinkholes using DInSAR techniques

The previously produced and analyzed SBAS-map (Castañeda et al., 2009), with a pixel size of ~ 90 m, is compared with the two new velocity maps (C-SPN- and L-SPN-maps) with higher spatial resolution (Table S1, see online Supplementary material). Both C- and L-SPN-maps have increased the density of measurement points by 23 and 6 times, respectively (Table S2, see online Supplementary material), resulting in the recognition of a higher number of active sinkholes and detection of smaller subsidence

Acknowledgments

We wish to thank Prof. Adrian Harvey, Prof. Jo De Waele and five anonymous reviewers for their helpful comments. This research has been funded by the Spanish national projects CGL2010-16775, AGL2012-40100 and CGL2013-40867-P (Ministerio de Ciencia e Innovación and FEDER), the Regional projects 2012/GA-LC-021 and 2012/GA LC 036 (DGA-La Caixa) and the European Interreg IV B SUDOE project DO-SMS-SOE1/P2/F157. Jorge Pedro Galve has been contracted under the DGA-La Caixa project and the contract of

References (70)

  • O. Pueyo-Anchuela et al.

    Actual extension of sinkholes: considerations about geophysical, geomorphological, and field inspection techniques in urban planning projects in the Ebro basin (NE Spain)

    Geomorphology

    (2013)
  • P. Thierry et al.

    3D geological modelling at urban scale and mapping of ground movement susceptibility from gypsum dissolution: the Paris example (France)

    Eng. Geol.

    (2009)
  • R. Tomás et al.

    Mapping ground subsidence induced by aquifer overexploitation using advanced differential SAR interferometry: Vega Media of the Segura River (SE Spain) case study

    Remote Sens. Environ.

    (2005)
  • M. Abelson et al.

    Collapse-sinkholes and radar interferometry reveal neotectonics concealed within the Dead Sea basin

    Geophys. Res. Lett.

    (2003)
  • P. Acero et al.

    Hydrogeochemical characterization of an evaporite karst area affected by sinkholes (Ebro Valley, NE Spain)

    Geol. Acta

    (2013)
  • R. Al-Fares

    The Utility of Synthetic Aperture Radar (SAR) interferometry in monitoring sinkhole subsidence: subsidence of the Devil's Throat Sinkhole Area (Nevada, USA)

  • M.D. Aurit et al.

    A GIS analysis of the relationship between sinkholes, dry-well complaints and groundwater pumping for frost–freeze protection of winter strawberry production in Florida

    PLoS ONE

    (2013)
  • G. Baer et al.

    The lowest place on earth is subsiding — an InSAR (interferometric synthetic aperture radar) perspective

    Geol. Soc. Am. Bull.

    (2002)
  • G. Benito et al.

    Karst in gypsum and its environmental impact on the Middle Ebro Basin (Spain)

    Environ. Geol. Water Sci.

    (1988)
  • P. Berardino et al.

    A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms

    IEEE Trans. Geosci. Remote Sens.

    (2002)
  • S. Bianchini et al.

    Landslide activity maps generation by means of persistent scatterer interferometry

    Remote Sens.

    (2013)
  • D. Carbonel et al.

    Evaluation of trenching, Ground Penetrating Radar (GPR) and Electrical Resistivity Tomography (ERT) for sinkhole characterization

    Earth Surf. Process. Landf.

    (2013)
  • C. Castañeda

    Basic Terrain-Motion of Spain: Zaragoza

  • C. Castañeda et al.

    DInSAR measurements of ground deformation by sinkholes, mining subsidence, and landslides, Ebro River, Spain

    Earth Surf. Process. Landf.

    (2009)
  • C. Castañeda et al.

    Dedicated SAR interferometric analysis to detect subtle deformation in evaporite areas around Zaragoza, NE Spain

    Int. J. Remote Sens.

    (2011)
  • D. Closson et al.

    Subsidence and sinkhole hazard assessment in the Southern Dead Sea Area, Jordan

    Pure Appl. Geophys.

    (2005)
  • D. Closson et al.

    Salt-dissolution-induced subsidence in the Dead Sea area detected by applying interferometric techniques to ALOS Palsar Synthetic Aperture Radar images

    Geodin. Acta

    (2010)
  • A.H. Cooper

    Subsidence hazards due to the dissolution of Permian gypsum in England: investigation and remediation

  • A.H. Cooper

    The classification, recording, databasing and use of information about building damage caused by subsidence and landslides

    Q. J. Eng. Geol. Hydrogeol.

    (2008)
  • A.H. Cooper et al.

    Avoiding Gypsum Geohazards: Guidance for Planning and Construction

    (1998)
  • Y. Cornet et al.

    Use of ERS tandem data to produce digital elevation models by interferometry and study land movements by differential interferometry in Calabria and Jordan

  • M. Crosetto et al.

    Generation of advanced ERS and Envisat interferometric SAR products using the Stable Point Network technique

    Photogramm. Eng. Remote. Sens.

    (2008)
  • M. Crosetto et al.

    Interferometric SAR monitoring of the Vallcebre landslide (Spain) using corner reflectors

    Nat. Hazards Earth Syst. Sci.

    (2013)
  • T. Dahm et al.

    Combining geophysical data sets to study the dynamics of shallow evaporites in urban environments: application to Hamburg, Germany

    Geophys. J. Int.

    (2010)
  • I.A. De Bruyn et al.

    The occurrence of sinkholes and subsidence depressions in the far West Rand and Gauteng province, South Africa, and their engineering implications

    Environ. Eng. Geosci.

    (2001)
  • Cited by (47)

    • Inferring subsidence characteristics in Wuhan (China) through multitemporal InSAR and hydrogeological analysis

      2022, Engineering Geology
      Citation Excerpt :

      This process influenced the dynamic strength of the soil body, which in turn resulted in karst collapse and subsidence (He et al., 2010). The relationship between land subsidence and sinkhole activity is complex and depends on many factors, including hydrogeological conditions and types of sinkholes (Jia et al., 2018; Galve et al., 2015; Shi et al., 2019; Orhan et al., 2021). For example, Orhan et al. (2021) found that sinkhole activity in the Karapınar region in Turkey was limited to areas with low rates of subsidence and attributed this phenomenon to the strong cohesion of the near-surface rock layer that resisted subsidence.

    • Coastal paleokarst landforms: A morphometric approach via UAV for coastal management (Algarve, Portugal case study)

      2019, Ocean and Coastal Management
      Citation Excerpt :

      Remote sensing technologies like airborne Light Detection and Ranging (LiDAR), Interferometric Synthetic Aperture Radar (InSAR), and other geophysical techniques in use during the last 40 years, have generated large volumes of high-accuracy and high-density topographical measurements, making regular update of sinkhole inventories and automated mapping achievable (Berardino et al., 2002; Lindsay and Creed, 2006; Filin et al., 2011; Kaufmann, 2014; Wu et al., 2016). Some of these technologies produce high-resolution digital elevation data, allowing an entirely new level of detailed delineation and permitting analyses of small-scale geomorphic features (Gutiérrez et al., 2011; Huang et al., 2014; Wu et al., 2014; Galve et al., 2015). These technologies have high costs that are not viable for many countries and cannot be achieved in a regular basis.

    View all citing articles on Scopus
    View full text