Using a bistatic radar echo sounding (RES) system developed for calibration of the RICE particle astrophysics experiment at the South Pole, we have studied radio frequency (RF) reflections off the bedrock. The total propagation time of ~ns-duration, vertically (z-) broadcast radio signals, as a function of polarization orientation in the horizontal plane, provides a direct probe of the geometry-dependence of the ice permittivity to a depth of 2.8 km. We observe clear birefringent asymmetries along z- in the lowest half of the ice sheet, at a fractional level ~0.3%. This result is in contrast to expectations based on measurements at Dome Fuji, for which birefringence was observed in the upper 1.5 km of the ice sheet. This effect, combined with the increased radio frequency attenuation expected near the bedrock, renders the lower half thickness of South Polar ice less favorable than the upper half of the ice sheet in terms of its ultra-high energy neutrino detection potential.
Deep Dive into Radio Frequency Birefringence in South Polar Ice and Implications for Neutrino Reconstruction.
Using a bistatic radar echo sounding (RES) system developed for calibration of the RICE particle astrophysics experiment at the South Pole, we have studied radio frequency (RF) reflections off the bedrock. The total propagation time of ~ns-duration, vertically (z-) broadcast radio signals, as a function of polarization orientation in the horizontal plane, provides a direct probe of the geometry-dependence of the ice permittivity to a depth of 2.8 km. We observe clear birefringent asymmetries along z- in the lowest half of the ice sheet, at a fractional level ~0.3%. This result is in contrast to expectations based on measurements at Dome Fuji, for which birefringence was observed in the upper 1.5 km of the ice sheet. This effect, combined with the increased radio frequency attenuation expected near the bedrock, renders the lower half thickness of South Polar ice less favorable than the upper half of the ice sheet in terms of its ultra-high energy neutrino detection potential.
The response of ice as a function of polarization ("birefringence") is characterized by differences in either wavespeed or absorption along linear (generally orthogonal) axes. Over a km-scale pathlength, in the absence of any preferred in-ice direction, one might expect any time propagation asymmetry at the single-crystal level to be macroscopically mitigated by the randomness of the corresponding single-crystal orientation. In such a case, over a total pathlength l consisting of N unit steps, each characterized by an asymmetry b, the average propagation time along each polarization axis should have a Gaussian distribution, centered at l/c, with width b √ N l/N c. The asymmetry distribution would therefore be a Gaussian of width σ b = b √ 2N l/N c, centered at zero. For 1% birefringence (b = 0.01), l=1000 m, and step sizes corresponding to typical grain sizes (10 -3 m, or N=10 6 ), we expect σ b 0.1 ns. If, however, bulk flow of the ice sheet results in a preferred in-ice direction much longer than typical grain sizes, the propagation time asymmetry can be O(10 ns).
Efforts are underway to use the Antarctic ice sheet as an ultra-high energy neutrino target(DeYoung, T., 2010; Kravchenko, I. and others, 2006;Gorham, P.W., and others, 2009). Neutrino-ice collisions result in the production of charged particles which emanate from the interaction point with velocities approaching the speed-of-light in vacuo: v → c. In a medium with index-of-refraction n > 1, detection of the resulting Cherenkov radiation in either the near-UV (TeV-scale neutrinos) or radio wavelength regime (PeV-scale neutrinos) by a suite of sensors can be used to reconstruct the kinematics of the initial neutrino, provided the absorption and refraction of the original electromagnetic signal due to the intervening ice can be reliably estimated. The RICE experiment (Kravchenko, I. and others, 2006) demonstrated the feasibility of the radiodetection approach over the last decade. The Askaryan Radio Array (ARA) Collaboration(ARA, 2010) seeks to substantially enlarge the current RICE footprint at South Pole by instrumenting an 80 km 2 area over the period 2010-2015. Ice birefringence could result in an initially ns-scale RF pulse being resolved into two components, with a time stagger comparable to the signal duration itself, requiring a trigger system with a correspondingly long signal integration time. Complete characterization of the ice permittivity, as a function of depth, and also polarization is therefore essential in obtaining a reliable estimate of the neutrino detection efficiency.
In particular, as the transmitter-receiver systems were rotated in azimuth, signal drops were observed at angles (2m + 1)π/4 (with m an integer) relative to the presumed horizontal COF-alignment, which were attributed to destructive interference between the ordinary vs. extra-ordinary signals at the broadcast wavelength. Those data (Matsuoka, K. and others, 2003) imply ∼ λ/2 variation over a depth of 1500 meters, which (given that λ ice ∼3 m for 179 MHz in-ice broadcasts) imply a birefringent asymmetry of order 0.1%. The authors note that their measured birefringence is somewhat weaker at 60 MHz, although an authoritative recent study of pure laboratory ice obtained an asymmetry δ ǫ ′ of 1.07±0.23% (Matsuoka, T. and others, 1997) at both 1 MHz and 39 GHz.
In a comprehensive attempt to model the Dome Fuji data, and neglecting any possible tilting of the reflecting internal layers (which must occur coherently over an aeral scale of order the Fresnel zone to be significant), the dependence of radar scattering on density, COF, and acidity effects were assessed (Matsuoka, K. and others, 2004). In principle, the type of scattering can be elucidated on the basis of signal strength: COF and acidity-based layers typically reflect -30 –60 dB of the incident power; density scattering, integrated through a vertical chord is typically a factor 10 larger (in power). To the extent that acidity scattering is a pure conductivity effect, we would expect it to vary as the inverse of frequency. To the extent that density and COF scattering is due to a variation of n(ω), we would expect such scattering to show much weaker frequency dependence, depending on the proximity of Debye resonances. In order to simplify the interpretation of data, the authors made the assumption that one scattering effect was dominant, and that although all types of scattering can result in large cross-polarized signals, only COF produces anisotropic scattering which also gave the observed azimuthal interference patterns.
In 2006, our group previously used time-domain bedrock reflections observed at a site near Taylor Dome (Besson, D.Z. and others, 2008) to estimate a birefringent asymmetry of 0.12%, projected onto the vertical ẑ-axis (perpendicular to the surface). A lack of conclusive ice flow data at Taylor Dome prevented a correlation with the local ice flow direction from being established. A foll
…(Full text truncated)…
This content is AI-processed based on ArXiv data.
