Introduction

The Japanese Antarctic Research Expedition (JARE) has carried out extensive glaciological work in the inland area of east Dronning Maud Land since 1982. During the last two seasons, radio echo-sounding was conducted at the lower altitudes (Reference Nakawo, Narita and IsobeNakawo and others 1984; Reference Nishio, Ohmae and IshikawaNishio and others 1986). They used 60 MHz radar for the ground survey and 179 MHz radar for the airborne survey.

The authors carried out the ground survey by 179 MHz radar at the higher altitudes during the 1984 field season. A-scope data were recorded every 10—20 km along the new traverse route (Fig.1). The object of the observations was to measure the ice thickness along the route and the positions of the internal echoes.

Fig. 1. Inland traverse route during 4 October—5 December 1984.

Radar System

The radar transmits a 1.5 kw (peak) pulse in a repetition frequency of 1 kHz. The receiver has a sensitivity of -104 dBm, whose receiving frequency is 179 MHz ± 2.5 kHz. Six-element Yagi antennae were used as the transmitting and receiving aerials which had linear polarizations. These antennae were set up on the snow, maintaining parallelism to each other at a distance of 3 m during the observations.

Reference Wada, Yamanouchi and MaeWada and others (1982) reported that this radar with a three-element Yagi antenna failed to detect echoes from the bed deeper than 1500 m when the radar was used for airborne surveys. In order to detect minor echoes, we have analyzed the A-scope data by a digital recording system. After having been digitized at 50 ns intervals, 256–512 received echoes were stacked in a memory to eliminate random noises. The A-scope echo shown on the CRT was also recorded as a photograph which was exposed for 20–30 s. This photographic film also represents the optically averaged echo from 20 000–30 000 received signals.

Results and Discussion

The radar system successfully detected all of the echoes from the bed along the route. Fig.2 shows a cross-section of the ice sheet along the traverse route obtained both by the echo-sounding and the surveyed altitudes of the ice surface. The ice thickness was determined by the echo time from the bed, assuming that the radio-wave velocity in the ice is 169 m/μs. The bed elevations in latitudes lower than 72 °S, are 100–200 m a.s.l. In the higher-latitude area, the bed elevations increase to 500—1000 m and the bottom topography is extremely rough. The thickest ice along the route was 2650 m at lat. 75°S. The traverse party of JARE-26 extended this route to lat. 78 °S. in 1985 and reported 3500–4000 m thick ice there (personal communication from K. Kamiyama).

Fig. 2. Cross-section of the ice sheet along the traverse route. Surface elevations of the ice sheet are based on the data of Reference Fujii, Kawada, Yoshida and MatsumotoFujii and others (1986).

The observations on the internal echoes were carried out at Mizuho station (lat. 70 °42 ‘ S., long. 44 °20 ¹E,, 2230 m a.s.l.). Although polarization in the ice sheet, observed when the transmitting antenna was rotated around the fixed receiving antenna R, has been reported in a number of papers (e.g. Reference HargreavesHargreaves 1978, Reference Woodruff and DoakeWoodruff and Doake 1979), few observations on polarization when two aerials are kept parallel to one another have been reported (e.g. Reference BentleyBentley 1975). Furthermore, these reports are based on analyses of the bottom-reflected signals; it seems that there have been virtually no reports on polarization resulting from internal echoes. Since the bottom-reflected echoes are affected not only by the birefringence of the ice but also by the morphology of the glacier bed (Reference BerryBerry 1975), it is necessary to analyze the internal echoes in order to determine the birefringence of the ice body itself. We observed the A-scope data for the internal echoes, when the receiving antenna was rotated around the transmitting antenna, and kept parallel to one another at a distance of 3 m. Measurements were repeated at 22.5° intervals (Fig.3), In order to evaluate the relationship between the polarization and the strain of the ice sheet, the antennae were installed on the snow at the center of a strain-grid which had been set up for observation of the strain-rate of the ice sheet.

Fig. 3. Setting of the antennae. The ice-flow direction is shown at the center. This direction is equal to that of the principal axis of the extending strain of the ice sheet (Reference Naruse and ShimizuNaruse and Shimizu 1978).

Fig.4 shows the noise-reduced A-scope data at the same travel-time (depth) scale which clearly indicates polarization of a radio wave in the ice sheet. The internal echoes from a depth of 1000—1500 m are strongest when the receiving antenna R is situated at east-south-east from the transmitting antenna T (No. 4, 5, 6 in the illustration) or west-north-west (No. 12, 13, 14).

Fig. 4. Internal echoes at 16 different antenna azimuths observed by A-scope data. Echo strengths above the average level of the 16 data sets obtained by A-scope are coloured dark. R and T represent the two parallel antennae, receiving and transmitting aerials, respectively. The observation was carried out in May 1984.

In order to analyze the echo strength with respect to the antenna azimuth, two internal echoes at depths of 500 m (echo A) and 1050 m (echo B) were chosen. Fig.5 shows the results. Both echoes A and B show an elliptical pattern, which confirms the birefringence of the ice sheet. The difference between the maximum and minimum echo strengths is a gain of 7—10 dB, which is a significant value for the receiver sensitivity.

Fig. 5. Echo strength as a function of the direction of the two parallel antennae. The positions of echo A and echo B are shown in Fig.3. The directions of the principal strain axes of the ice sheet observed at the same time are shown at the center of the circle graph.

The A-scope records in Figs.4 and 5 indicate that the internal layers could be most clearly distinguished when the azimuth was perpendicular to the flow line. The internal echo from a depth of 1050 m (echo B) became strongest when the antenna azimuth faced 030° or 210° (R was situated at 120 or 300° (from T), This antenna azimuth is very close to that of the principal compressive strain of the ice sheet, which is perpendicular to the flow line at the observation site. This coincidence of the azimuth may suggest that the birefringence was caused by the ice flow. Reference Narita, Fujii and NakawoNarita and others (in press) have examined the 700 m deep ice core from Mizuho station and have reported that the c-axes of the ice crystals below 400 m depth become aligned along the vertical plane. It is plausible that the Strain history of the ice has caused this orientation fabric, which may determine the birefringence of a polycrystalline ice sheet as Reference HargreavesHargreaves (1978) has argued in a theoretical study.

Another possible cause of this polarization could be explained by the depositional pattern of the internal layers. Internal layers, such as volcanic ash layers, probably have prolonged patch-patterns due to the strain deformation of the ice sheet. The reflection coefficients of these irregular patches would depend on the azimuths of the antennae.

Reference MillarMillar (1981) has analyzed the internal echoes in the Greenland ice sheet and he concluded that the internal echo layering was formed by layers of ice of changed composition, which are probably composed of acidic ice formed after volcanic eruptions. Since we have found a 5 cm thick volcanic ash-laden ice layer at Mizuho station (Reference Narita, Fujii and NakawoNarita and others in press), the A-scope record when the antenna azimuth was perpendicular to the flow line has been compared with the position of the ash-laden ice layer. The result (Fig.6) represents a strong echo up to 20 dB peak at the depth of the volcanic ash-laden ice layer. This coincidence suggests that the echo strength/depth profile corresponds to acidic layers caused by past volcanic activity in east Dronning Maud Land.

Fig. 6. Internal echo and the depth of the volcanic ash layer at Mizuho station. The antennae azimuths were perpendicular to the direction of ice flow. Echo sounding was carried out in December 1984.

Although dating of the ash-laden ice layer at 500 m depth has not yet been completed, from the annual accumulation rate of 7 g/year at Mizuho station (Reference Watanabe, Kato, Satow and OkuhiraWatanabe and others 1978) the age can be estimated as ∼8 × 10³ year B.P. Other internal layers with high reflection coefficients seem to be located at 1000–1500 m depth. The ages of these ice layers can be estimated as between 20 and 40 × 10³ year B.P. These ages of the internal layers seem to coincide with those of volcanic ash bands found in the ice core at Byrd Station (Reference Gow and WilliamsonGow and Williamson 1971); this coincidence may suggest the extensive fall-out of acidic aerosols on the Antarctic ice sheet. Further echo-sounding of internal layers will undoubtedly reveal the areal distribution of these volcanic ash layers during the last glaciation.