Electromagnetic wave speed in polar ice: Validation of the CMP technique with high resolution dielectric-profiling and gamma-density measurements

• Main Authors: Eisen, Olaf, Nixdorf, Uwe, Wilhelms, Frank, Miller, Heinrich
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: 2002
• Subjects: Dronning Maud Land; Annals of Glaciology; Antarc*; Antarctica; ice core
• Online Access: https://epic.awi.de/id/eprint/3624/; https://epic.awi.de/id/eprint/3624/1/Eis2000a.pdf; https://hdl.handle.net/10013/epic.14204; https://hdl.handle.net/10013/epic.14204.d001

The accuracy of the traveltime-velocity and traveltime-depth profile derived from ground-penetrating radar (GPR) common-midpoint (CMP) surveys at different frequencies is investigated for the first time ever by direct comparison with the profile calculated from high resolution dielectric-profiling (DEP) ice core data.In addition, we compare two traveltime profiles calculated from ice core density data by means of different dielectrical mixture models with the DEP based profile.CMP surveys were carried out at frequencies of 25, 50, 100 and 200 MHz near the new European deep drilling site DML05 in Dronning Maud Land, Antarctica, during the 1998/99 field season.An improved scanning capacitor for high resolution DEP and a Gamma-densiometer for density measurements were used to determine thecomplex dielectric constant and the density at 5 mm increments along the ice core B32, retrieved in 1997/98 at DML05.The comparisons with DEP and density based velocity series show that the CMP velocity series are slightly higher, but asymptotically approach the core based velocities with depth.Root-mean-square differences of the DEP velocity series range between 8% for the 25 MHz CMP and 2% in the case of the 200 MHz survey.Density based velocities differ from the DEP velocities by less than 1%.The traveltime-depth series calculated from the interval velocities show a better agreement between all series than the velocity series.Differences are between 5.7 and 1.4% for the 25 and 200 MHz CMP measurements, and less than 0.6% for the density data.Based on these comparisons we evaluate the accuracy with which the depth of electromagnetic reflectors observed in common-offset profiles can be determined and discuss reasons for the observed differences between CMP- and core based profiles.Moreover, we compare the errors determined from the field measurements with those estimated from GPR system characteristics to provide a measure that can be used to estimate the accuracy of GPR analyses for the planning of GPR campaigns.Our results show ...