Ground penetrating radar ice thickness measurements of Dokriani bamak (glacier), Garhwal Himalaya

J. T. Gergan, D. P. Dobhal and Rambir Kaushik

Wadia Institute of Himalayan Geology, Dehradun 248 001, India

Ice thickness measurements of Dokriani bamak (glacier) were carried out with a ground penetrating radar (GPR) pulse EKKO IV Sensor Software, Canada. This is the first attempt to estimate the ice thickness of an Indian Himalayan glacier by GPR. Profiling by GPR was carried out at a central frequency of 12.5 MHz with 2 m step size. Three distinct reflection patterns were observed; (i) glacier ice (GI), (ii) subglacial lodgment till (SLT) produces complex series of reflections; and (iii) bedrock (BR). Crevasses are distinguished by the laterally compressed synclinal hyperbolas stacked one over the other. The glacier ice thickness measured by GPR ranged from 15 m to 120 m. Ice thickness calculated theoretically and by GPR were very close. The computed total volume of glacier ice was 283.6  106 m3.

In November 1995, ground penetrating radar (GPR) profiling was carried out to measure ice thickness of Dokriani bamak (bamak meaning glacier in local dialect of the area). Pulse EKKO IV radar of Sensor Software, Canada, was used to carry out the profiling. GPR sounding was carried out from the snout, the accumulation zone of the glacier. Nearly 30% to 40% of the glacier could not be approached by the GPR profiling team as the glacier is intensely dissected by crevasses. Therefore, special efforts were made to measure the glacier ice thickness at the base of the icefall and above it, to have a realistic estimate of ice thickness of the glacier as far as possible. Profiling along the central line and across the glacier has provided us with a fair amount of data to understand the morphological feature of the bedrock. The GPR profiling of subglacial lodgment till and bedrock profile has facilitated in the interpretation and understanding of the subglacial hydrology and sediment transfer. With intensive GPR profiling of the glacier, it would be possible to estimate the volume of glacier ice with a higher degree of confidence.

Dokriani bamak is one of the valley glaciers in Bhagirathi river basin, that has nearly 282 glaciers. The Dokriani bamak glacier lies to the west of famous Gangotri glacier, the source of Ganga River (Figure 1). It extends from 30 50 to 30 52  N and 78 47 to 78 50  E. The total length of glacier is 5 km. It has a catchment area of 15.1 sq km, out of which 5.76 sq km is covered with ice, 3.94 sq km is covered by permanent snowfield, making a total glacierized area of 9.70 sq km. The remaining 5.40 sq km area is nonglacierized.

Dokriani bamak is formed by two cirques glacier, one on the northern slopes of Draupadi Ka Danda (5716 m) and second on the western slopes of Janoli (6632 m). The glacier flows in the direction of NNW for 2.0 km till the base of the ice fall from where it turns towards WWN direction and flows for 3.0 km with an average gradient of 12 and terminates at a height of 3882 m (m.s.l). The glacier is lying over a thick layer of lodgment till and it flows between two lateral moraines which were formed during the past glaciation.

Dokriani bamak glacier was selected for GPR studies, as it is easily accessible throughout the year. GPR survey was carried out in the month of November, so as to exploit the general absence of surface melt water and the danger of hidden crevasses are minimum during this month. Furthermore, the study was facilitated owing to the permanent glaciological field research station of Wadia Institute located near this glacier.

170.gif (25990 bytes)

Pulse EKKO IV ground penetrating radar is a modular designed lightweight portable system, made up of (i) Radar control system; (ii) Console; (iii) Transmitter and receiver assembly. GPR profiling was carried out in reflection mode with a central frequency of 12.5 MHz. The two antennas for transmission and reception were placed parallel to each other with a spacing of eight meters. The transmitter and the receiver were connected to a console unit of 2.4 ns. This in turn was connected to a laptop computer to control the operation of pulse EKKO IV. The computer was used to collect and store the data. In order to minimize the lateral reflection from valley wall–bedrock ice interface, the antennas were oriented parallel to the survey line1. More than 5 km of glacier was profiled by GPR along the central longitudinal profile and transverse profiles with a step size of 2 m. In the present work, profiles for only 4.5 km have been discussed. Network of the GPR profile lines over the glacier (Figure 1) were established. A longitudinal profiling was carried out for a length of 2.346 km from the snout of the glacier to the upper reaches of ablation zone (3940 to 4680 m). Five transverse profiles across the central line were undertaken in the ablation zone and two profiles in the accumulation zone (Figure 2 a and b). In the accumulation zone, GPR profile was not carried out extensively due to the presence of large and deep crevasses.

GPR profile data collected was processed by Pulse EKKO IV software. GPR data of Dokriani glacier shows three types of prominent reflection patterns: (i) The topmost zone has very few reflections and appears to be almost transparent; this transparent-looking zone is typical of presence of massive glacier ice (GI) (Figure 3). In this zone the only prominent reflections are from englacial boulders and water/air pocket in glacier ice.

171.jpg (18588 bytes)

One of the most prominent reflections in GI is from the crevasses, they are reflected as laterally compressed and vertically stacked hyperbolas (Figure 3). (ii) Just below the glacier ice there is a zone with very complex series of reflections with a number of flattened hyperbolas. This zone is identified as the subglacial lodgment till (SLT). SLT is thick in the lower reach of the ablation zone of the glacier. The average thickness of SLT is 20–25 m below the ablation zone near the snout. SLT thins down in the higher reaches of the ablation zone and it is almost absent in some of the parts of the profile from the accumulation zone. The calculation of the SLT thickness from GPR profile is difficult at present, as the reflections from SLT-bedrock (BR) interphase are not very clear. Due to the scattering of radar energy caused by the melt water and number of boulders in the SLT, a complex pattern of reflections at the interphase of SLT and BR is produced. Right below the SLT, there is a zone with regular pattern of reflections, which are from the massive BR.

The hyperbolas produced by the reflection from englacial boulders air/water pocket (Figure 3) in the GI or SLT are flatter compared to hyperbolas produced by crevasses. They are by and large compressed with steeply sloping limbs (Figure 3). The reflection from large crevasses continues through the SLT, while from the smaller crevasses they do not continue till the SLT. The hidden crevasses (HC) have similar reflection patterns as open crevasses, but they do not continue till the surface of GI (Figure 3). The depressions in SLT and BR formed by the subglacial pool (SGP) have synclinal reflection pattern that slopes towards the centre from the side of subglacial depression, giving the appearance of half-cut inverted parabola at the edge of the depression. The sediments in the sub glacial pool have regularly layered pattern of reflections. There are three such depressions which have been identified along the longitudinal GPR profile (Figures 3 and 4). The glacier ice thickness of Dokriani bamak varies between 15 and 120 m. SLT is 20–25 m below the ablation zone of the glacier, which thins down to 6 m in the accumulation zone2. Mathematical formula used to estimate glacier ice thickness in order to cross check the GPR results and to calculate the ice thickness in inaccessible area of the glacier is2:


where t b is the shear stress on the glacier bed, which is equal to the plastic yield stress of ice and is usually taken as 1bar (1 bar is equal to 10dyne/cm2) (ref. 2), r is the ice density (0.9 gm/cm3), and f is the slope of the glacier surface (slope of the glacier is based on the Survey of India contour map, 1995), g is the gravity acceleration and considered as 978.3 gm/cm3 (ref. 3). F is the shape factor of the valley (obtained from a factor W). Ice thickness was calculated by assuming that the glacier cross-section is a semi-ellipse in shape with length of 5.0 km varying in width from 80 to 500 m. W is half of the valley width divided by thickness of the glacier ice along the central line obtained by GPR profiling. W is calculated at different cross-sections of the glacier which varies between 0.5 and 0.7 as Dokriani bamak is assumed as a semi-ellipse, shape factor (F) for semi-ellipse is 1 (ref. 4).

The glacier ice thickness calculated by mathematical method of Dokriani glacier varied from 25 to 135 m. While the minimum thickness was obtained near the snout (<  25 m), the maximum thickness of 135 m was in the middle of accumulation zone (5000–5100 m.a.s.l.). The

172.jpg (123169 bytes)


173a.gif (16334 bytes)

173b.jpg (11233 bytes)

ice thickness between an altitude of 4500 and 4800 m on the eastern flanks of the valley was only 25 m; this is due to the steep slope in the bedrock. The ice thickness calculated by GPR survey near the snout was 15–25 m. Whereas the ice thickness measured manually in 1995 near the center point of snout ice wall was 20 m, the maximum thickness obtained by GPR survey was 120 m (in accumulation zone). This shows that the mathematically computed ice thickness varies from the ice thickness derived from GPR profile by  (5 – 15 m). GPR profiling and mathematically computed ice thickness have been complied and a glacier ice thickness (isopack) map of Dokriani bamak has been prepared (Figure 5). The total volume of the glacier ice calculated from isopack map is 283.6   10m3 (of w.equi).

Figure 4 shows the glacier features identified by the GPR profiling along the central line of the glacier. Three subglacial pools (SGPs) and many open and hidden crevasses have been identified. It also shows that the SLT at lower reaches is thick which thins at the higher altitude.

GPR studies of Dokriani glacier have shown that it is a convenient and economical method to calculate glacier ice thickness of temperate Himalayan glaciers. Besides this, it has also been demonstrated that it is a good tool to detect the presence of subglacial pools, hidden crevasses, englacial boulders and air and water pockets within the glacier ice. However, more intensive studies are required to develop methodologies to profile the glacier with thick supraglacial cover and to demarcate the boundary between the valley wall and glacier ice. Moreover, there are many other problems to be investigated such as the reflection pattern from the interphase of bedrock subglacial lodgment till.

  1. Anna, A. P. and Cosway, S. W., Extended Abstracts of the 62nd Annual International Meeting of the Society of Exploration Geophysicist, New Orleans, 25–29 October 1992.
  2. Nye, J. F., Nature, 1952, 169, 529.
  3. Bahuguna, V. B., Report of scientific studies along with finding based on Geodetic and Geophysical observations carried out by Survey of India in Chhota Shigri glacier, Multi-disciplinary glacier expedition to Chhota Shigri glacier, July–August 1987, Technical Report No. 2, 1988, pp. 264–348.
  4. Paterson, W. S. B., The Physics of Glaciers, Pergamon Press, Toronto, 1st edn, 1969.

ACKNOWLEDGEMENTS. We thank Director, Wadia Institute of Himalayan Geology for assigning the project and providing the facilities. We also thank Department of Science and Technology, Governtment of India, New Delhi, for funding the project.

Received 30 April 1998; revised accepted 22 April 1999