Annals of Glaciology 20 1994 £; International Glaciological Society

Flow regime of the Lambert Glacier- system, : structural evidence from Landsat Imagery•

MICHAEL J. HAMBREY School of Biological and Earth Sciences, Liverpool John lvIoores University, Liverpool L3 3AF, England

JULIAN A. DOWDESWELL Scott Polar Research Institute, Universi~y of Cambridge, Cambridge CB2 IER, England

ABSTRACT. High-resolution visible and near-infrared satellite imagery provides a means of investigating the structural glaciology, and in turn the dynamics, of large ice masses. The Lambert Glacier-Amery Ice Shelf system is one of the largest ice drainage basins in Antarctica and has previously yielded conflicting evidence concerning its dynamic behaviour: either that the system has a propensity for surging or that it has a constant flow regime. Digital manipulation of Landsat imagery allows analysis of the structure of the glacier system, showing longitudinal foliation, medial moraines and crevasse patterns that provide no evidence of surging behaviour during the residence time of ice in the glacier system.

INTRODUCTION THE LAMBERT GLACIER-AMERY ICE SHELF SYSTEM The possibility of disintegration of large parts of the Antarctic ice sheet, either through the removal of ice The Lambert Glacier-Amery Ice Shelf system is a shelves (which hold back the marine-based ,Vest composite feature, comprising eight structurally defined Antarctic ice sheet) under the influence of global ice streams, [or example, Lambert, l\1ellor and Fisher warming or as a result of surging of the major ice Glaciers, which feed into the Amery Ice Shelf (Figs 2 drainage basins, has been suggested on a number of and 3). A number of other ice streams join the main trunk occasions (Wilson, 1964; Hollin, 1969; Hughes, 1975; glacier along its length, notably Charybdis Glacier from Mercer, 1978). Supporting evidence of surging has been the west, and a large icc stream off l\1awson Escarpment offered for the Lambert Glacier-Amery Ice Shelf [rom the east (Fig. I). Data on ice thickness, velocity, ice system (Fig. 1), which drains about one-fifth of the discharge and the grounding-line position have been East Antarctic ice sheet (Budd and McInnes, 1978; provided by Morgan and Budd (1975), Budd and others Derbyshire and Petersen, 1978; Allison, 1979; Well- (1982), McIntyre (1985a) and Hambrey (1991). man, 1982). However, other authors have argued A model of surging of the system (Budd and others, against surging [or this system (Robin, 1979, 1983; 1978) suggested a surge periodicity o[ the Lambert McIntyre, 1985a; Radok and others, 1987). Glacier-Amery Ice Shelf system of 23 000 years, with The purpose of this paper is to examine the major. surges lasting 250 years and velocities reachi~g structural evolution of the Lambert Glacier-Amery several kilometres per year. It was further suggested that Ice Shelf system in order to assess whether the system similar behaviour might be representative of other exhibits surging on a cyclical basis or has a constant Antarctic drainage basins. flow regime. Features such as foliation, medial Allison (1979) calculated the state of balance of the moraines and crevasse patterns, observable in Landsat system from mass-balance and ice-flux data, and found images, may be used to determine the mode of flow o[ that the mass flux for the interior part of the Lambert the major ice streams and outlet glaciers of the Glacier basin was far in excess of output into the main Antarctic ice sheet. These structures are most obvious channel of Lambert Glacier. He regarded this imbalance where the snow cover has been removed by ablation, as providing evidence for the build-up of ice prior to a revealing bare glacier ice, as in the case of the Lambert surge. On the other hand, Robin (1979, 1983) reached Glacier system. These structures are described from an the alternative conclusion that the apparent imbalance analysis of digital Landsat imagery and are discussed in may be due to strong basal melting. In a re-assessment of the context of possible past instabilities in the flow of the area of the interior basin from Landsat imagery, the system. McIntyre (1985a) reduced its size from 1.09 to 0.90

401

Downloaded from https://www.cambridge.org/core. 29 Sep 2021 at 14:03:46, subject to the Cambridge Core terms of use. Hambr~y and Dowde.w;ell: Flo 11.) regime ~[ Lambert Glacier-Amery lee Shelf system, Antarctica

PRYDZ BAY

KEY

Rock outcrops • Structurally defined /' flow unit boundaries , Extrapolation of flow unit ,." boundaries Inferred ice flow /' directions B.L. Beaver Lake d Area covered by Fig. 2

FLOW UNITS

1 NW Amery Ice Sheet 2 Charybdis Glacier 3 From E Prince Charles , Mountains 4 5 6 Lambert Glacier 7 From NE of Mawson Escarpment PRINCE NE Amery Ice Sheet

CHARLES MOUNTAINS 100. km

Fig. 1. lviajor flow units of the Lambert Glacier-Ame~y lee Shelf .~ystem derived from the structural jJattem revealed in satellite images (after Hambrq, 1991). The grounding line isfrom Budd and others (1.982). The location of Figure 2 is shown. The location (d the J~ambert Glacier-i1me~y lee Shel[rystem {('ithin Antarctica is shown in the inset; dashed lines represent ice divides (after Drew~y) 1983).

x]O fi km-,? and funher suggested that the accumu]atton. METHODS rate was much lower than previously estimated; this had the effect of reducing the mass flux from the interior basin Enhancement of digital satellite imagery allows specific to a fig-ure that almost matched the input into Lambert surface features relating- to structure and topog-raphy to be Glacier. Subsequent correspondence (Allison and others, analysed in derail (cf. Dowdeswcll and McIntyre, 19871. ]985; Mcintyre, ]985b) served to emphasize the Digital data from three Landsat :Ylultispectral Scanner disagreement and pointed to a need for better field (:YlSS) scenes, covering the four spectral bands with data, especially concerning accumulation rates, or [or wavelengths between 0.5 and 1.1 11m (path/row 134/112 alternative methods of analysis. imaged on 20 February 1974; path/row 135/111 and 135/ .vlost recently, an attempt has been made using a 112, both imaged on 16 :Ylareh 1973), were manipulated theoretical modelling approach to determine which on a Sun-based image-processing system in Cambridge, Antarctic drainage basins \overeprone to surging (Radok and structural analysis took place on enhanced images and others, ]987). The models demonstrated that the both on screen and in the [arm of hard copy. major Antarctic ice streams, including the Lambert In the absence of measurements of velocity and strain Glacier-Amery Ice Shelf system, arc unlikely to surge. rate, the pattern o[ice structures may be used to infer flow

402

Downloaded from https://www.cambridge.org/core. 29 Sep 2021 at 14:03:46, subject to the Cambridge Core terms of use. Hambre)! and Dowdeswell: Flow regime of Lambert Glacier~Amer'y Ice Shelf j)istem, Antarctica

'i,;":'.~.• •• • I •..•. • '. ' "" KEY

• Rock outcrops Thin Cloud ~ Foliation

@. Folded foliation

~\ Discrete crevasses

Inferred bedrock step .,'

50 I

Fig. 2. Structural map from satellite image/I, illustrating the dominant foliation trend, medial moraines and discernihle crevasse patterns in the rep,ion where Lamhert, AIelloI' and Fisher Glaciers mel}!,e.

dynamics and particle paths through ice masses (Rey- imagery and airborne altimetric data from the head of nolds, 1988; Reynolds and Hambrey, 1988; Hambrey, , that the surface expression of foliation may_ 1991:. Despite the order-of~magnitude larger scale, there have amplitudes in excess of 10 m, is a close correspondence between the crevasse and foliation patterns in large ice streams, such as the Lambert Glacier system, and those in valley glaciers in STRUCTURE OF LAMBERT GLACIER AND ITS mountain regions (cf. Hambrey and :\,hiller, 1978). TRIBUTARIES Landsat images, which clearly depict the major structures, cover Lambert, l'viellor and Fisher Glaciers, Foliation and the main Lambert Glacier system north of the confluence of these ice streams (Fig. 1), areas which are The dominant feature on digitally enhanced Landsat critical for examining structural evidence for a variable imagery of the Lambert Glacier Amery Ice Shelf system is flow regime. Satellite imagery is probably recording hoth a longitudinal structure or foliation (Fig, 2). Such structures, the gross textural contrasts between different flow units which in Antarctic ice masses generally are parallel to the and topographic variations. Dowdeswell and iVid ntyre margins of the flow unit, have previously been referred to as (1987) have shown, through comparison of Landsat flowlines ~Crabtree and Doake, 1980; Swithinbank, 1988),

403

Downloaded from https://www.cambridge.org/core. 29 Sep 2021 at 14:03:46, subject to the Cambridge Core terms of use. Iiambrq and Dowdeswell: Flo/v regime of l~amhert C;lacier-Ame~y Ice Shelf .~J!stem,Antarctica

Fig. 3. Digital(J! enhanced Landsat iVISS image oj the Fig. 4. Digital(y enhanced Landsat 1\ISS image of ice confluence region of l~ambert, ;VIellor and Fisher Glaciers. .flow around Clemence ;Hassif, located ill Figure 1. The The image is Fom jJath/row 134/ ll2, acquired on 20 image is pom jJath/row 135/ Ill, acquired on 16 March Februa~y 1974, and is ajJjJroximate(J! 70km across. A 1974, and is ajJjJroxirnate(J! 40 km acruss. Ice fluw is marked break of slope, seaward of the confluence and towards the north-northeast (tof) of image) in the zone to afJInoximatel;' 70 km across, is marked ~y arrows. Ice flow the left uf Clemence AIassif. Various Jeatures are labelled in the main channel is towcirds the north-northeast. i.e. the and discussed in the text. tOjJ~lthe image.

flow stripes of uncertain origin (Casassa and others, 1991) or Although thc foliation remains parallel to the overall flow traces originating from former shear margins (Merry direction of flow, diverging flow around Clemence Massif and Whillans, 1993!. However, the structures are not always to the north of Mawson Escarpment (Fig. I) deforms the parallel to current l10w direction and, from ground longitudinal foliation into an arcuate or transverse trend observations elsewhere, it appears that they are the surface as it turns northeast and flows around the eastern side of expression of the three-dimensional structurc, longitudinal this nunatak (Fig. 4:. initially, the arcs are expressed at foliation (Reynolds and Hambrey, 19881. Longitudinal the surface as waves a few kilometres apart (Fig. 4. foliation commonly is parallel to medial moraines and is location A). After passing through a snow-covered area therefore useful in delineating individual flow units. showing relatively little definition, this icc turns northwest In the Lambert Glacier-Amery lee Shelf system, the and the arcs, now as transverse foliation, re-emerge. The foliation and moraine pattern indicate that, as flow units arcs beeome progressively tighter downstream as iee flow merge, their width is considerably reduced (Figs I 3) and is constrained by a major ice stream (flow unit 7). The the foliation is intensified. structure herc is dearly picked out by meltpools on the ice

404 Downloaded from https://www.cambridge.org/core. 29 Sep 2021 at 14:03:46, subject to the Cambridge Core terms of use. Hambrq and DowdesweLl: Flow regime of Lambert Glacier-Ame'(y lee Shelf 5.-ystem,Antarctica

surface (Fig. 4, location B). As the ice rejoins the main Several ice-surface features, exhibited on enhanced trunk glacier (Fig. 4, location C), the arcuate foliation Landsat imagery of the area around Clemence Massif becomes increasingly attenuated. Transposition of this (Fig. 4), suggest that the ice may be afloat at this location, structure to a new longitudinal foliation takes place in the considerably further south of the grounding-line position zone of convergence. To the west of Clemence .YIassif, the of Budd and others (1982). These features (Fig. 4) longitudinal foliation of the Lambert Glacier flow unit inelude: (i) a very flat feature about 6 km in length also becomes deformed, here forming faint drag-like folds occupying a depression and having the form of an ice- (Fig. 4, location D). covered lake, which may be connected to sub-ice-shelf waters; (ii) two elliptical depressions 3- 4 km in diameter, Crevasses similar to features termed "ice dolines" by Mellor (1960) and found on several Antarctic icc shelves (Mellor and Crevasses are widely distributed in the three ice streams, McKinnon, 1960; Swithin bank, 1988); these features Fisher, S1ellor and Lambert, that feed into the main have been interpreted as melt lakes that drain period- trunk glacier called Lambert Glacier. In the first two ice ically through fractures to the sea beneath (Swithinbank, streams, longitudinal zones of closely spaced transverse 1988); (iii) large areas of icc-surface lakes, indicating the crevasses several kilometres wide are separated by very low ice-surface slopes found in this area. Floating icc relatively crevasse-free zones of similar width (Fig. 3). around Clemence .Ylassif would imply that the ice shelf These zones are parallel to the dominant longitudinal extends at least 400 km inland from the terminus of the structure and are present across the entire width of each Amery Ice Shelf on the east side of Lambert Glacier. ice stream. Tn Lambert ice stream, crevasse fields are more extensive and chaotic, and the zones, although still recognizable in enhanced images, are more diffuse. DISCUSSION Soon after the three ice streams have merged, the ice passes through a transverse, concave down-g'lacier zone of Does the Lambert Glacier-Amery Ice Shelf system disturbed flow (arrowed in Figure 3). Here, the crevasses surge? are locally more intense and the longitudinal structures

are slightly bent. These perturbations ~re probably Structural patterns l!1 surge-type glacier systems are related to a bedrock step or riegel and the ice surface distinctive and differ from those in glaciers of the non- has the appearance of steepening through this zone. surging type (~1eier and Post, 1969; Post, 1972). Folded Down-glacier, the crevasses gradually die out and, as and truncated foliation, and tear-drop-shaped moraines the ice enters the narrowest part of the Lambert Glacier are typical. If the Lambert Glacier system, and in trough, few remain, allowing meltwater streams to flow particular Fisher Glacier, had surged as previously considerable distances and lakes to form on the glacier supposed (Wellman, 1982) within the time it takes for surface. Only at the margins of the main trunk glacier, all its ice to reach the ice-shelf edge, surge-related where small tributaries descend from the Prince Charles structures should be clearly visible in the satellite I\!Iountains and I\!lawson Escarpment, are marginal imagery. Evidence of surging in earlier times would crevasses developed. have been preserved down-glacier if any of these flow units had surged. Velocity data suggest a residence time Boudinage of ice passing along the centre line is of the order of 1000 years, a figure likely to be an order-of~magnitude Boudinage structures on a scale of a few metres, resulting grea ter near the margins. None of this ice shows evidence from differences in ductility within a multi-layered of former surging behaviour. Rather, we conclude that sequence, arc common in foliated glacier ice (Hambrey the structures indicate a constant flow and constant and Milnes, 1975). deformation regime through time (Fig. 4·). The zone between the lVlellor and Lambert flow units Crevasse patterns may also be used to infer whether a west of Clemence Massifdisplays a string ofmega-boudins glacier is building up to a surge condition. During the (Fig. 4, line EE'). On average, these are 2 km wide, quiescent phase of a surge-type glacier, crevasses thinning to 1 km at the necks or breaking into discrete disappear as the ice stagnatcs. As the ice at the head of blocks, and are of the order of 5 km long. These mega- the glacier builds up to a new surge and becomes mor.e boudins appear to originate to the west of ~tawson active, it becomes progressively more crevassed. There is a Escarpment to the south, and become more pronounced clear demarcation between heavily crcvassed ice and ice adjacent to Clemence Massif. Their development is which remains relatively slow moving (Lawson, 1989). probably related to longitudinal extension of the ice, first There is no such demarcation in Lambert Glacier, further where it becomes strongly channelized and sub-sequently supporting the hypothesis that the (low regime is constant. as it approaches the grounding zone. Mega-boudins are This flow rcgime appears to have been maintained even also evident at the boundary between the Fisher and though the ice thickness and mass balance may have ~Ie]]or flow units east and northeast of Fisher .\'lassif. changed considerably during this period.

Other surface features Implications

Grounding lines or zones are often associated with a break Tht' method of analysing Ice structures usmg Landsat in ice-surface slope and have been identified in a number imagery, which has been tested alongside ground of places around Antarctica (e.g. Swithinbank, 1988). observations elsewhere, has served to demonstrate that

405

Downloaded from https://www.cambridge.org/core. 29 Sep 2021 at 14:03:46, subject to the Cambridge Core terms of use. Hambr~y and Dowdeswell: Flow regime ~f Lambert Glacier-Ame~y Ice Sheff ~ystem, Antarctica

one of the major drainage basins of has Hambrey, M.J. 1991. Structure and dynamics of the Lambert Glaeier- not surged for several thousand years. In contrast, Amery lee Shelf system: implications Jar the origin of sediments. In Barron, J. and others. 1991. Proceedings oj the Oceal1 imbalances in flow have been documented for Ice Drilling ProKrum, Scientific Results, 119, 61 75. Ocean Drilling Stream B, which feeds the Ross Ice Shelf from the West Program, College Station, Texas. Antarctic ice sheet (Bindschadler and others, 1987; Hambrey, :V1.J. and A. G. Milnes. 1975. Boudinage in glacier ice - some Whillans and others, 1987). Surging is further suggested examples. ]. Glaciol., 14(72), 383-393. Hambrey, M.J. and F. Moller. 1978. Structures and ice deformation in by the deformed nature of "flow stripes" (which we the \Vhite Glacier, Axel Heiberg Island, "Iorthwest Territories, interpret as foliation) in the Ross Ice Shelf (Casassa and Canada. ]. Clucio!., 20~82), 41-66. others, 1991). Hollin, J. T. 1969. Ice-sheet surges and the geological record. Can. ]. Satellite imagery offers the prospect of determining, Earth Sci., 6, 903 910. Hughes, T. 1975. The \Vest Antarctic ice sheet: instability, disinte[(- from structural patterns, the flow regimes of many other ration, and initiation of ice ages. Rev. Geophys. Space I'hys., 13:4'1, parts of the Antarctic ice sheet, and establishing which, if 502-526. any, part has a propensity for surging. In ice masses Lawson, \V. 1989. The structural evolution of the Variegated Glacier, outside Antarctica, such phenomena are exccptional, and Alaska. (Ph.D. dissertation, University of Cambridge.) ~lcIntyre, N. F. 1935a. A re-assessment of the mass balance of the t-here is little reason to suppose that the Antarctic ice sheet Lambert Glacier drainage basin, Antarctica. J. Glaciol., 31!] OJ', is any different. 34-38. McIntyre, "I. F. 1985h. Reply to "On re-assessrnent of the mass balance of the Lambert Glacier drainage basin, Antarctica". ]. Glaciol., 31( l(9), :,81. ACKNOWLEDGEMENTS Meier, M. F. and A. S. Post. 1969. What are glacier surges? Can.]. Earth Sci., 6(4), 807-817. We thank the B. B. Roberts Fund of the Scott Polar 1\1ellor , 1\1. 1960. Correspondence. Antarctic icc terminology: icc Rcscarch Institute for the financial support which facili- dolines. Polar Rec., 10(64),92. Mellor. :VI. and G. McKinnon. 1960. The Amery Ice Shelf and its tated this work, and M.J. Sharp, 1. M. Whillans and two hinterland. Polar Rec., 10:64), 30-34. anonymous referees for comments on versions of this paper. Mercer, J. H. 1978. \Vest Antarctic ice sheet and CO2 greenhouse efIect: a threat of disaster. }(utllle, 271~56431, 321-325. Merry, C.J. and 1. M. Whillans. 1993. Ice-flow features on Ice Stream B, Antarctica, revealed by SPOT HR V imagery.]. Glaciol., 39,] 331, REFERENCES 515-527. Morgan, V.1. and W. F. Budd, 1975. Radio-echo sounding of the Allison,!. 1979. The mass budget of the Lambert Glacier drainage Lamhert Glacier basin. ]. Glaciol., 15(73), 103-11]. basin, Antarctica. J. Glaciol., 22(87), 223;!:>.'i. Post, A. 1972. Periodic surge origin of Ji)lded medial moraines on Bering Allison, 1. F., "I. \Y. Young and T. :Vledhurst. 1985. On re-assessment of Piedmont Glacier, Alaska. ]. Glacial., 11 (62), 219 ·226. the mass balance of the Lambert Glacier drainage basin, Antarctica. Radok, U., D. Jenssen and B. McInnes. 1987. On the surging patmtial of .J. Glaciol., 3W09), 378-381. /Jolar ice streams. Antarctic surges - a rlear and present danger' \Vashington, Ilindschadlcr, R. A., S. i\. Stcphenson, D. R. :VlaeAycal and S. Shahtaie. DC, U.S. Department of Energy. (DOE/ERj60197-H1.1 1987. Icc dynamics at the mouth of Ice Stream B, Antarctica. Reynolds, J.M. 1988. The structure of Wordie Ice Shelf, Antarctic ]. Geo/JIZYJ. Res., 92(B9), 8885-8894. Peninsula. Brit. Antaret. Surv. Bull., 80, 57 64. Budd, \V.I'. and B.J. McInnes. 1978. Modelling surging glaciers and Reynolds, J. M. and M.J. Hambrey. 1988. The structural glaciolog, of

periodic surging of the Antarctic icc sheet. 111 Pittock, A. B., L. A. George V I Icc Shelf; Antarctica. Brit. Antaret. Surv. Bull., 79. 79-95. Frakes, D. Jenssen, J. A. Peterson and J. \V. Zillman., cds. Climatic Robin, G. de Q ]979. Formation, flow and disintegration of icc shelves. dwnge and ;;ariabiliU: a southern /Jmpecti;;e. Cambridge, Cambridge ]. Glaciol., 24(90;, 259 271. Cniversity Press, n8-2:B. Robin, G. dc Q ]983. Coastal sites, Antarctica. In Robin, G. de Q, ed. Budd, \Y. F., :V1.J. Corry and T. H.Jacka. 1982. Results lI'orn the Amcry The climatic record in /Jolar ice sheets. Cambridge, Cambridge University Ice Shelf Project. Ann. Giliciol .. 3, 36-41. Press, 118 122. Casassa, G., K. C.Jezek,J. Turner and 1. M. Whillans. 1991. Relict flow Swithinbank, C. 1988. Satellite imagc arlas of glaciers of (he world. stripes on the Ross Icc Shelf. Anl1. Glaciol., 15, 132-138. Antarctica. C.S. Ceol. Suru. Prof. Pap. 1386-B. 1-138. Crabtree. R. D. and C. S. M. Doake. 1980. Flow lines on Antarctic ice Wellman, P. 1982. Surging of Fisher Glacier, East Antarctica: evidence shelves. I'olar Rec.. 20: 12'1',31 37. from geomorphology. ]. Glaciol. 28(98;, 2328. Dcrbyshirc, E. and J. A. Peterson. 1978. A photo-geomorphic map of the \Vhillans, I. M., J. Bolzan and S. Shabtaic. 1987. Velocity of Ice Streams 1\1t. 1\Ienzics nunatak, , Australian Hand C, Antarctica. ]. Ceopkys. ReJ., 92(B9), 8895-8902. Antarctic Territory. Z. Gle/scherkd. Gladulgeol., 14, 17-26. Wilson, A. T. 196,1. Origin of ice ages: an ice shelf theory for Pleistocene J)owdcswdl, J. A. and N. F. Mcintyre. 1987. The surface topography of glaciation. Nature, 201(491."),147-149. large icc masses fi'om Landsat imagery.]. Glacio!., 33(] n), ]6-23. Drewry, D.J. 1983. The sudace of the Antarctic ice sheet. In Drewry, D.J., d. ~lntarctica: glaciological and geophysical Jolio. Cambridge, Tlte accuracy of references in the text and in this list is the l;ni,ersity of Cambridge. Smtt Polar Research Institute, Sheet 2. resjJonsibility of the authors, to whom queries should be addressed.

406

Downloaded from https://www.cambridge.org/core. 29 Sep 2021 at 14:03:46, subject to the Cambridge Core terms of use.