Estimation of ground cavity configurations using ground penetrating radar and time domain reflectometry

Won Taek Hong, Jong-Sub Lee

Research output: Contribution to journalArticlepeer-review

8 Citations (Scopus)


Ground cavity configurations, including depth, roof shape, and length, are the main factors affecting the risk of ground sinkholes. In this study, ground penetrating radar and time domain reflectometry are applied to estimate the ground cavity configurations. To accurately estimate ground relative permittivity with depth, a penetrometer incorporated with a time domain reflectometry (PTDR) system is developed. In addition, a new method is established to calculate the coordinates of the reflection points. Experimental studies are conducted on ground models prepared in different soil types with buried objects of various shapes to simulate ground cavities by using circular rubber tubes and rectangular and trapezoidal polystyrene objects. The experimental studies show that the estimated depths of the buried objects are identical to the experimental setup. The estimated roof shapes clearly represent the roof shapes of the buried objects. In addition, the estimated diameters of the rubber tubes and the estimated lengths of the polystyrene objects show good agreement with those of the buried objects. This study shows that the ground penetrating radar survey, PTDR test, and the new method for estimating ground cavity configurations may be effectively used to assess the risk of ground sinkholes.

Original languageEnglish
Pages (from-to)1-19
Number of pages19
JournalNatural Hazards
Publication statusAccepted/In press - 2018 Mar 23


  • Ground cavity
  • Ground penetrating radar
  • Relative permittivity
  • Sinkhole
  • Time domain reflectometry

ASJC Scopus subject areas

  • Water Science and Technology
  • Atmospheric Science
  • Earth and Planetary Sciences (miscellaneous)


Dive into the research topics of 'Estimation of ground cavity configurations using ground penetrating radar and time domain reflectometry'. Together they form a unique fingerprint.

Cite this