Hydrological Sciences—Journal—des Sciences Hydrologiques, 44(1) February 1999 47

Particle size characteristics of suspended sediments and subglacial hydrology of Dokriani Glacier, Garhwal Himalaya, India

RENOJ J. THAYYEN, J. T. GERGAN & D. P. DOBHAL Wadia Institute of Himalayan Geology, Dehradun 248001, India

e-mail : wihg@giasd 101 .vsnl.net.in

Abstract Diurnal and seasonal variations in particle size characteristics of suspended sediments of Dokriani glacier meltwater were studied during the 1994 ablation period (May-October). Diurnal suspended sediment concentration curves exhibit two prominent peaks from May to mid-August and only one peak during the rest of the ablation period. The first peak resulting from dominance of fine and medium sand contributed to the rising limb of the diurnal hydrograph, whereas the second peak was dominated by fine and medium silt which coincides with the diurnal discharge peak. The trends observed on the appearance of silts and sand particles on diurnal scales at various stages of the ablation period suggest that the subglacial drainage system comprises a diurnally-reversing hydraulic gradient between the channels and distributed system along with translatory flow through the distributed system. This study suggests that the subglacial zone is the major source of sediments in the Dokriani glacier meltwaters. However, high intensity monsoonal rainfall has contributed supraglacial sediments and resulted in bimodal distributions. Fines (clay to medium silt) are the dominant classes in the suspended sediments except during the spring event; the supraglacial sediments had coarsely skewed distributions.

Caractéristiques granulométriques des matières en suspension et hydrologie du glacier Dokriani (Garhwal Himalaya, Inde) Résumé Les variations diurnes et saisonnières de la granulométrie des matières en suspension transportées par les eaux de fonte du glacier Dokriani ont été étudiées durant la période d'ablation de 1994 (mai-octobre). La courbe journalière de la concentration des matières en suspension présente deux extremums de mai à la mi-août et un seul extremum durant le reste de la période d'ablation. Le premier extremum résulte de la dominance des sables fins et moyens le long de la branche ascendante de l'hydrogramme journalier, alors que le second résulte de la présence de silts fins et moyens et coïncide avec le maximum journalier de débit. Les tendances observées concernant l'aspect des silts et des sables à l'échelle journalière pour différents moments de la période d'ablation suggèrent que le système de drainage sous-glaciaire subit tous les jours des renversements de gradient hydraulique entre ses chenaux allant de pair avec un flux d'eau transversal. Cette étude suggère que, en ce qui concerne les eaux de fonte du glacier Dokriani, le domaine sous-glaciaire est le principal fournisseur de sédiments. Les fortes intensités des pluies de mousson entraînent toutefois une contribution de sédiments du domaine supra-glaciaire et sont à l'origine de distributions bimodales. Les matières en suspension sont surtout constituées de fines (argiles à silts moyens) sauf au printemps et les sédiments supra-glaciaires sont distribués de façon très asymétrique.

INTRODUCTION

Discharge and suspended sediment transfer characteristics of a glacierized catchment are regulated primarily by input conditions (climate) and routing conditions (Elliston,

Open for discussion until I August 1999

48 Renoj J. Thayyen el al.

1973). The runoff characteristics of the Himalayan catchment are greatly influenced by the monsoonal rainfall from mid-June to mid-September (Thayyen & Hasnain, 1997). Precipitation over the extensive supraglacial debris cover and proglacial zone, especially over the large lateral moraines, enhances the erosion processes in the glaciated region of Himalaya. Within the glacial system, supraglacial and subglacial zones are the source areas of sediments (Gurnell & Fenn, 1984). Therefore the particle size characteristics of suspended sediments at the glacial portal carry the signatures of the supraglacial and subglacial contributions. Glacier channel characteristics play an important role in determining the particle size distribution of suspended sediments from the subglacial zone as well as in modifying the size distribution of suspended sediment produced from the supraglacial zone. Hydraulic characteristics of glacier drainage are determined by the amount of water input derived from snowmelt, icemelt, rainfall and groundwater which vary over the ablation period (Rothlisberger, 1972; Iken, 1981; Hooke et al, 1985; Iken & Bindschadler, 1986; Rothlisberger & Lang, 1987). Therefore, the variability in the hydraulic characteristics of glacier drainage systems and in the particle size characteristics of suspended sediments tend to show diurnal and seasonal variability. Particle size characteristics of suspended sediments in glacier meltwater and their variations through the ablation period are controlled by the size

78°47' 78°50 Fig. 1 Map showing location of Dokriani Glacier.

Particle size characteristics of suspended sediments and subglacial hydrology 49

characteristics of the source sediments, eroding mechanisms and hydraulic character­istics of the flow paths. In this work the diurnal and seasonal variations in particle size distribution of suspended sediments in Dokriani glacier meltwater during the 1994 ablation period (May-October) were studied. Samples representing various stages of suspended sediment transfer processes were analysed in relation to variability in routing conditions and climate.

AREA OF STUDY

Dokriani Glacier is a small Garhwal Himalayan glacier in the Bhagirathi basin, located in Uttarkashi district of Uttar Pradesh. It extends between latitude 30°50'-30°52'N and longitude 78°47'-78°50'E (Fig. 1). The total length of the glacier is 5 km. It originates from Draupadi ka Danda (5716 m a.m.s.l.) and flows in a NNW direction for 2 km; then it turns in a WWN direction and flows another 3 km with an average gradient of 12° and terminates at a height of 3882 m a.m.s.l. The stream emerging from the Dokriani Bamak (Glacier) is known as Din Gad, a tributary of the River Bhagirathi. The total area of the glacier catchment is 9.58 km2, of which 5.76 km2 is covered with glacier ice. Three quarters of the glacier ablation area is covered by thick debris cover. The accumulation area of the glacier is nearly 2.5 km2. This glacier is formed by two cirque glaciers, one originating from Draupadi ka Danda and the other from the western slopes of Janoli peak (6632 m a.m.s.l.) joining near the equilibrium line (4960 m a.m.s.l.).

METHODOLOGY

Observations were carried out from 8 May to 31 October 1994 covering the entire ablation period. A gauging station was established at Din Gad proglacial stream, 600 m downstream from the glacier portal. Discharge measurements and water sampling were carried out at this station. Discharge measurement was carried out by area-velocity method and by establishing a rating curve. Water samples were collected in narrow necked polyethylene bottles by the method described by Ostrem (1975). Meltwater samples were collected at intervals of three hours. Suspended sediments were filtered out from the samples by using pre-weighed 0.45 um Millipore membrane filter papers of 47 mm diameter with the help of a vacuum pump and filtration unit and the filtrate was dried in the open air. Before weighing the sediment, larger grains, if present, were removed manually from the samples. Such problems were encountered in few samples. After drying and weighing, sediments were scraped out from the filter paper. Ashing was not carried out, because of the low amount of sediment and to avoid textural variations of suspended sediments that would have occurred during the ashing. Particle size analyses of selected samples were carried out by Malvern Master Sizer E. The Master Sizer measures the volume percent of the particles by laser ensemble light scattering; it is programmed to convert the result into weight percent by using the density factor.

50 Renoj J. Thayyen et al.

The particle size of the samples analysed was in the range of 0.5-600 um. It is assumed that 600 um is the largest possible size in suspension, considering the hydrological characteristics of Din Gad proglacial stream. The Master Sizer E gives the results in 32 size classes. For the analyses, the results were grouped into six size classes (Table 1) based on the Woodden Worth scale. The statistical parameters were calculated using the method of moments in phi scale.

Table 1 Particle size classes and mean values.

Size class (um)

0.5-7.78 7.78-25.46 25.46-56.09 56.09-123.59 123.59-272.31 272.31-600

Mean (nm)

4.14 16.62 40.72 89.84

198.95 436.15

Size class (<]))

11-7 7-5.3 5.3-4.12 4.12-3.0 3.0-1.88 1.88-0.74

Mean (<|>)

9 6.1 4.71 3.56 2.44 0.81

Specification

Clay-fine silt (A) Fine silt-medium silt (B) Coarse silt (C) Very fine sand (D) Fine sand (E) Medium sand (F)

RESULTS

Diurnal and seasonal variations in particle size characteristics

To study the diurnal and seasonal variations in particle size characteristics of suspended sediments, samples representing various stages of sediment transfer were studied (Fig. 2). The 26 May samples represent the first rise in discharge and suspended sediment concentration and show dominance of fine and medium silt classes (Fig. 3). Discharge and suspended sediment concentration show good positive correlation during the period (0.97). Discharge (Q) and suspended sediment concentration (SSQ again show rising trends from 15 June onwards. On 19 June discharge showed a rising trend throughout the diurnal cycle, whereas the suspended sediment concentration fell after the initial rise (Fig. 4(a)). The percentage of fine and medium sand reduced between 08:00 and 11:00 h and then rose to the higher level by 14:00 h. The suspended sediment concentration curve shows two peaks and recedes after the second peak at 20:00 h. Particles ranging from fine silt to very fine sand (7.78-123.5 um) dominated in the suspended sediments. However fine and medium sand curves peaked at 14:00 h and coincided with the first suspended sediment peak. Diurnal variations in particle size distributions on 19 June are shown in Fig. 4.

A subglacial sediment flushing event was recorded from 29 June to 2 July. This is being interpreted as a spring event (Rothlisberger & Lang, 1987) which is associated with complete breakdown of the subglacial hydrological system (Fountain, 1994). The first sign of the flushing, observed at 08:00 h on 30 June, showed a sharp increase in suspended sediment concentration from the initial level of 4.5 to 23 g l1. The suspended sediment concentration reached a high of 66 g 1"1 by 11:00 h and fell back to the initial levels by 08:00 h on 3 July, 72 h after its

Particle size characteristics of suspended sediments and subglacial hydrology 5 ]

, rfkr. njnMiillJiyJLii nn,nufl,

..h.jtii illililiil Lui lililïllliHh.,4

Date

Fig. 2 Suspended sediment concentration, discharge and rainfall during the 1994 observation period. Small horizontal lines show the periods selected for particle size study,

40.0

o) 20.0 3

s~\

v j

- • -26 May

18 August

Supraglacial

8.99 6.15 2.44 1.31 4.71 3.56 Particle size 0

Fig. 3 Particle size distribution representing initial phase of suspended sediment discharge (26 May), bimodal distributions during the high intensity rainfall (18 August) and supraglacial sediments.

initiation. Discharge variations during the period ranged from 5 to 8 m3 s1 and no correlation existed between discharge and suspended sediment concentration. Figure 5 shows sequential changes in particle size distribution during the flushing event. On 28 June, prior to the flushing, particle size distribution shows dominance of fine and medium silt classes. The distribution pattern started showing charac­teristic deviation on 29 June with the development of bimodal distribution by the increased presence of very fine sand and fine sand. By 30 June, distribution

52 Renoj J. Thayyen et al.

(b)

"5)20.0

1<P

Class

A B C D E F

Mean pm

4.14 16.62 40.72 89.84 198.95 436.15

170.00 170.40 170.80 Julian days

171.20 171.60

Fig. 4 Diurnal variations in (a) discharge and suspended sediment concentrations and (b) particle size classes on 19 June 1994.

characteristics changed completely with strong domination of fine sand and medium sand, and the trend continued until 1 July. Fine and medium sand together contri­buted 79% to the suspended sediment peak (66 g l"1). At 11:00 h on 1 July, sus­pended sediment concentration was 16 g 1"1 and contribution of medium sand alone constituted 72%. After three hours, at 14:00 h, suspended sediment concentration sharply reduced to 8.6 g I1 and the contribution of medium sand reduced to a minimum of 2.74%. During the recessional phase of the event on 2 July, distribution became bimodal with the domination of fine and medium silt and fine sand (Fig. 5). On 6 July, distribution showed dominance of fine and medium silt for the first time

Particle size characteristics of suspended sediments and subglacial hydrology 53

80.0 —,

35

§40.0

80.0

•§40.0

Ï

0.0

8.99 6.15 4.71 3.56 2.44 1.31 Particle size 0

Fig. 5 Particle size distributions showing the various phases of the spring event.

after the flushing event. It was observed that the particle size distribution during the period underwent cyclic variation. Contribution of coarser particles (fine and medium sand) during the flushing event ranged between 87 and 55 %.

Figure 6 shows the diurnal variations in particle size distribution of 5 July representing the post-evacuation period. The correlation between suspended sediment concentration and discharge improved to 0.30. The discharge showed an increasing trend and the sediment transport was also higher than the temporal average. The suspended sediment peak during the period was associated with the peak of medium sand (60%). Analyses of samples after ten days (18, 19 and 20 July) showed a better discharge-suspended sediment concentration relationship (r = 0.56) and a low per­centage contribution of coarse particles (medium sand = 7.0%). Runoff during the period was greatly influenced by precipitation and showed high amplitude diurnal variations in discharge (6.5-13.5 m3 s"1) and suspended sediment concentrations (1.3-9.5 g l"1). Bimodal particle size distribution during these periods suggests that precipitation leads to the contribution of more fine and medium sand to the bulk flow. Medium silt and fine sand were the dominant classes in these bimodal distributions (Fig. 3). On 27 July the discharge-suspended sediment relationship showed further improvement (r = 0.69). There is a reduction in suspended sediment concentration (0.967-2.693 g V) from the earlier high levels, even when the discharge is high (4.52-10.72 m3 s'), which clearly indicates the reduced availability of sediments in the glacial system during the period. Diurnal variations in particle size distribution of 27 July showed that the fine sand and medium sand classes peaked at 14:00 h which was associated with the first suspended sediment peak (Fig. 7). Six hours later, at 20:00 h, a second sediment peak was observed in association with the diurnal discharge

+ - 1/7/94

2/7/94

6/7/94

-e -- % - •

-0

29/6/94

28/6/94

30/6/94

54 Renoj J, Thayyen et ai

(a)

(b)

CO

<u

sz o o to CO

I

Class

A B C D E F

Meanpm

4.14 16.62 40.72 89.84 198.95 436.15

186.00 186.40 186.80 187.20 187.60

Julian days Fig. 6 Diurnal variations in (a) discharge and suspended sediment concentrations and (b) particle size classes on 5 July 1994.

peak. This second sediment peak was the product of finer classes (fine to coarse silt). The percentage of fine and medium sand rose again when discharge fell between 02:00-05:00 h. Observations on 18 August were marked by reduced water flow (2.96-5.5 m3 s4), very low sediment concentrations (0.31-1.56 g l"1) and good correlation between the two (r = 0.87). Medium silt was the dominant size class and fine and medium sand (classes E and F) were the least dominant. The sediment-discharge relationship showed characteristic deviation from the previous observations. The suspended sediment concentration curve has a single peak associated with the crest of the discharge hydrograph (Fig. 8(a)). The percentage contribution of fine and medium

Particle size characteristics of suspended sediments and subglacial hydrology 55

co

(D 8.0

'<0

O to CO

Ciass

A B C D E F

Meanjjm

4.14 16.62 40.72 89.94 198.84 436.15

208.00 208.40 208.80 209.20 209.60

Julian days Fig. 7 Diurnal variations in (a) discharge and suspended sediment concentrations and (b) particle size classes on 27 July 1994.

sand increased at 17:00 h, but was not sufficient to take the sediment concentration to a level to produce a significant peak. The suspended sediment peak at 20:00 h was mainly caused by the increased percentage of fine and medium silt.

The diurnal hydrograph and suspended sediment concentration graph shows better correlation (0.95) on 7 September (Fig. 9). Discharge (2.95 m3 s"1) and suspended sediment concentration (0.177 g l"1) reached a new low during this period. The suspended sediment concentration curve has only a single peak associated with the peak discharge. The sediment peak was associated with higher contributions of silts. The

56 Renoj J. Thayyen et al.

CO

CD 4.0 D) V _

CO

-5 en

b

o CO CO

(b) Class

A B C D E F

Meanjjm

4.14 16.62 40.72 89.94 198.95 436.15

230.00 230.40 230.80 231.20 231.60

Julian days Fig. 8 Diurnal variations in (a) discharge and suspended sediment concentrations and (b) particle size classes on 18 August 1994.

analyses of samples collected on 29 September, 21 and 26 October also showed the same particle size distribution characteristics. Samples collected from various supraglacial environments were analysed to try to understand the distribution characteristics of supraglacial source sediments. All the selected distributions show negative skewness with the domination of fine sand (Fig. 3), which is different from the general distribution characteristics of suspended sediments in the meltwaters, which were positively skewed.

Particle size characteristics of suspended sediments and subglacial hydrology 57

(a)

<D 4 0

Q

O CO CO

(b)

4>

Class

A B C D E F

Meanpm

4,14 16,62 40.72 89.84 198.95 436.15

Julian days Fig. 9 Diurnal variations in (a) discharge and suspended sediment concentrations and (b) particle size classes on 7 September 1994.

Textural characteristics of suspended sediments

Statistical parameters of particle size distribution were calculated using the method of moments. Generally, mean particle size varied between 6.0 and 4.05 § except for the flushing event when mean particle size was 2.8 (j>. There is no specific pattern found in the variation of mean size with discharge and sediment concentration. Sorting of particles in the glacial suspended sediments falls into the categories poorly sorted to very poorly sorted. Standard deviation (a), varies between 1.2 and 2.4. Particles in

58 Renoj J. Thayyen et al.

the range of 3-1.88 § were best sorted and reflected the conditions under which turbulence is just able to support the coarsest size fraction (Passega, 1977; Sly et al., 1983).

DISCUSSION

Variations in particle size distribution and subglacial hydrology

Diurnal and seasonal variations in particle size distributions during the 1994 ablation season show specific trends which are probably related to the glacier drainage characteristics, especially those of subglacial drainage networks. Considering the differences in distribution characteristics, the possibility of dominant supraglacial contribution of suspended sediments to the bulk flow is very low. As pointed out earlier, the size distribution of suspended sediments in meltwater is positively skewed, whereas the supraglacial sediments show negatively skewed distributions. However, bimodal distributions observed during the high intensity rainfall events (Fig. 3) suggest contribution of coarse particles from the surficial sources. Even though the distributions during the flushing event were characteristically closer to the supraglacial samples, the extraordinarily high ionic pulses associated with the event and the extended duration (72 h) as well as absence of high intensity rainfall during the period suggest a subglacial origin (Thayyen, 1997). Humphrey & Raymond (1994) discussed the possibility of sediment storage at the basal zone and related sorting. Their observation suggests that partial settling of sediments may occur in distributed or linked cavity systems when water flows with lower velocities. Coarser particles (0.75-3.0 <|>) from the subglacial zone tend to erode first due to their lower erosion velocity (Hjulstrom, 1939; Sly et al., 1983) than that of fines and they constitute the bulk of the suspended sediments during the flushing period. Low intensity physical activity at the glacial bed during the winter period due to reduced sliding velocity till the spring event (Kamb & Engelhardt, 1987) probably helps the particles to retain their original physical characteristics like in a "passive" englacial system (Small, 1987).

During the 1994 ablation period fine and medium silt dominated the size distribution except during the suspended sediment peaks associated with the spring event. Two distinct peaks were observed during the ablation period to 27 July on diurnal suspended sediment concentration curves. The first peak was observed at 14:00 h associated with the peak of fine sand and medium sand curves. The second peak of suspended sediment concentration always occurred along with the discharge peak at 20:00 h and was associated with larger contribution of fines (fine and medium silt). Another peak of fine and medium sand, which was smaller than the first, was observed between 23:00 and 02:00 h throughout the ablation period. On 18 August a single peak of suspended sediment was observed, even in the presence of the first peak of sand. Diurnal high discharge and associated sediment peaks initially occurred at 20:00 h. As the ablation season progressed the peaks started shifting towards the left. The first peak of sand curve slowly shifted to the right

Particle size characteristics of suspended sediments and subglacial hydrology 59

during the period, from its initial appearance at 14:00 -17:00 h on 7 September, and started contributing to the recession limb of the suspended sediment curve,

The seasonal and diurnal behaviour of particle size distribution and the occurrence of peaks in particular size classes are probably related to the subglacial hydrological processes. Hubbard et al. (1995) suggested the existence of a diurnally shifting hydraulic gradient between a "variable pressure axis" (VPA) and the adjoining distributed system of a deformable glacier bed. When water pressure rises in the channels by early afternoon, the water is driven out laterally from the channel and flows back to it overnight. In-phase variations of suspended sediment concentration with the diurnal hydrograph at the glacier portal suggest the existence of a hydraulic gradient within the distributed system. This may be translatory flow (Gurnell & Fenn, 1984) through an activated gradient within the subglacial aquifer controlled by the bed topography and water pressure variations in the channels. A layer of 6-20 m thick deformable bed beneath the Dokriani Glacier is identified by ground penetrating radar survey (Gergan et al., in press; Gergan, 1996). It has a maximum thickness near the snout and gradually thins down in the upper ablation zone. This subglacial aquifer probably plays an important role in controlling the hydraulic characteristics of Dokriani Glacier. It is proposed that a system exists comprising a diurnally reversing hydraulic gradient between the channels and the distributed system (Hubbard et al, 1995) along with the translatory Darcy flow (Gurnell & Fenn, 1984; Walder & Flower, 1994; Fountain, 1994) through the subglacial aquifer. Fine and medium sand (0.75-3.0 (j>) within the distributed system eroded first due to their low erosion velocity (Hjulstrom, 1939; Sly et al., 1983) by means of the translatory flow during the diurnal rise of discharge. This contributed to the first peak of suspended sediments which appeared between 14:00 and 17:00 h. Diurnal maximum discharges and associated translatory flow brought down maximum fines to the glacier portal and resulted in the second sediment peak. The second and subdued peak of fine and medium sand curves observed between 23:00 and 02:00 h was probably contributed by the reversed hydraulic gradient from the distributed system to the channels. These peaks were consistent throughout the ablation season. As ablation progressed, reduced availability of sediments within the subglacial system resulted in low suspended sediment concentration in meltwater and the disappearance of the first suspended sediment peak. Reduced efficiency of the distributed system during the later part of the ablation period is also evident from the early occurrence of a discharge peak through the fast system (Raymond et al., 1995).

CONCLUSIONS

The relationships between discharge and suspended sediment concentrations on a diurnal scale at different stages of the ablation period show characteristic variations. During the first stage of ablation, to the last week of July, the suspended sediment concentration in meltwaters was controlled more by the internal channel development processes. During this period, variations in suspended sediment concentration occurred independent of discharge variations. After the sediment evacuation

60 Renoj J. Thayyen et al.

processes and the stabilization of glacial drainage systems, the relationship between discharge and suspended sediment concentration improved consistently. Diurnal suspended sediment concentration curves show two distinct peaks till mid August. The first peak is associated with the rising limb of the diurnal hydrograph and the second coincides with the peak discharge. The analysis of diurnal variations in particle size distribution shows that the first peak resulted from the larger contributions of fine and medium sand whereas the second peak was due to the fine and medium silt. The characteristics of diurnal variations in particle size and discharge at different stages of the ablation period suggest that the subglacial drainage system comprised a diurnally reversing hydraulic gradient between channels and the distributed system along with the translatory Darcy flow through the distributed system. The particle size distribution of Dokriani glacier meltwater was dominated by fine and medium silt. However, very high amounts of fine and medium sand were transported during the spring event. The surficial samples show coarse skewed distribution which is characteristically different from the fine skewed distributions of suspended sediments in meltwater and suggests that the subglacial zone is the major source of suspended sediments. However the bimodal distributions associated with the rainfall events suggest that the high intensity rainfall indeed contributes substantial amounts of suspended sediments from the surficial sources.

Acknowledgements The authors are grateful to the Director, Wadia Institute of Himalayan Geology, for assigning the project and providing the facilities. They also acknowledge the comments of reviewers and suggestions by Dr Mark Williams in improving the quality of this paper. Thanks are also due to Department of Science and Technology, Government of India, New Delhi, for funding the project under the Nationally Coordinated Project on Himalayan Glaciers.

REFERENCES

Elliston, G. R. (1973) Water movement through the Gornergletscher. In: Symp. an the Hydrology of Glaciers (Cambridge, UK, 7-13 September 1969). IAHS Publ. no. 95.

Fountain, A. G. (1994) Borehole water level variations and implications for the subglacial hydraulics of South Cascade Glacier, Washington, USA. J. Glacial. 40(135), 293-304.

Gergan, J. T. (1996) Report on scientific expedition to Dokriani Glacier. Submitted to Department of Science and Technology, Govt of India, New Delhi.

Gergan, J. T., Dobhal, D. P. & Rambir Kaushik (in press) Ground penetrating radar ice thickness measurements of Dokriani Bamak (Glacier), Garhwal Himalaya. Current Sci.

Gurneli, A. M. & Fenn, C. R. (1984) Flow separation sediment source areas and suspended sediment transport in a proglacial stream. Catena Suppl. 5, 109-119.

Hjulstrom, F. (1939) Transportation of detritus by moving water. In: Recent Marine Sediments (ed. by P. D. Trask) (Symp. Am. Assoc. Petrol. Geol., Tulsa), 5-31.

Hooke, R. L., Wold, B. & Hagen, J. O. (1985) Subglacial hydrology and sediment transport at Bondhusbreen, Southwest Norway. Geol. Soc. Am. Bull. 96, 388-397.

Hubbard, B. P., Sharp, M. J., Willis, I. C , Nielsen, M. K. & Smart, C. C. (1995) Borehole water-level variations and the structure of the subglacial hydrological system of Haut Glacier d'Arolla, Valais, Switzerland. J. Glacial. 41(139), 572-583.

Humphrey N. F. & Raymond, C. F. (1994) Hydrology, erosion and sediment production in a surging glacier: Variegated Glacier, Alaska, 1982-1983. /. Glacial. 40(136), 539-552.

Iken, A. (1981) The effect of subglacial water pressure on the sliding velocity of a glacier in an idealized numerical model. J. Glacial. 27, 407-421.

Particle size characteristics of suspended sediments and subglacial hydrology (, \

Iken, A. & Bindschadler, R. A. (1986) Combined measurements of subglacial water pressure and surface velocity of Findelengletscher, Switzerland: conclusions about drainage system and sliding mechanism. J. Glaciol. 32(110), 101-119.

Kamb, B. & Engelhardt, H. (1987) Waves of accelerated motion in a glacier approaching surge: the mini-surges of Variegated Glacier, Alaska, USA. J. Glaciol. 33(113), 27-46.

Ostrem, G. (1975) Sediment transport in glacial meltwater streams. In: Glaciofluvial and Glaciolaustrine Sedimentation (ed. by A. V. Jopling & B. C. McDonald), 101-122. Society of Economic Paleontologists and Mineralogists Special Publ. no. 23.

Passega, R. (1977) Significance of CM diagrams of sediments deposited by suspensions. Sedimentology 24, 723-733. Raymond, C. F., Benedict, R. J., Harrison, W. D., Echelmeyer, K. A. & Strum, M. (1995) Hydrological discharges

and motion of Fels and Black Rapids Glaciers, Alaska, USA: implications for the structure of their drainage systems. /. Glaciol. 41(138), 290-304.

Rothlisberger, H. (1972) Water pressure in intra- and subglacial channels. J. Glaciol. 11, 177-203. Rothlisberger, H. & Lang, H. (1987) Glacial hydrology. In: Glacio-Fluvial Sediment Transfer, An Alpine Perspective

(ed. by A. M. Gurnell & M. J. Clark), 207-284. Wiley, Chichester, UK. Small, R. J. (1987). Moraine sediment budgets. In: Glacio-Fluvial Sediment Transfer, An Alpine Perspective (ed. by

A. M. Gurnell & M. J. Clark), 165-197. Wiley, Chichester, UK. Sly, P. G., Thomas, R. L. & Pelletier, B. R. (1983) Interpretation of moment measures derived from water-lain

sediments. Sedimentology 30, 219-233. Thayyen, R. J. (1997) Sediment transfer and hydrochemical studies on Dokriani glacier meltwaters, Garhwal Himalaya,

India. Doctoral Thesis, Wadia Institute of Himalayan Geology, Dehradun, India. Thayyen, R. J. & Hasnain, S. I. (1997) Monsoonal control on discharge characteristics of glacierized Himalayan

catchments. In: Nat. Symp. on Himalayan Glaciers and Snow Cover (3-4 March 1997, New Delhi, India). Extended abstracts, Himalayan Glacier Project, School of Environmental Science, Jawaharlal Nehru University, New Delhi, India.

Walder, J. S. & Flower, A. (1994) Channelized subglacial drainage over a deformable bed. J. Glaciol. 40(134), 3-15. Received 23 July 1997; accepted 2 July 1998