Paraglacial geomorphology

Abstract

Paraglacial geomorphology is the study of earth-surface processes, sediments, landforms, landsystems and landscapes that are directly conditioned by former glaciation and deglaciation. The withdrawal of glacier ice exposes landscapes that are in an unstable or metastable state, and consequently liable to modification, erosion and sediment release at rates greatly exceeding background denudation rates. This paper (1) reviews research on paraglacial processes, landforms and landscape change in a range of geomorphological settings; (2) explores the importance of paraglacial landscape modification and sediment recycling as a component of alternating glacial/nonglacial landscape evolution; (3) assesses the nature and significance of paraglacial facies in Quaternary stratigraphic sequences; and (4) develops a general model of the sequence of paraglacial landscape modification and the changing nature of paraglacial landsystems.

Six paraglacial landsystems are identified: rock slopes, drift-mantled slopes, glacier forelands, and alluvial, lacustrine and coastal systems. Each contains a wide range of paraglacial landforms and sediment facies. Collectively these landforms and sediments (e.g. talus accumulations, debris cones, alluvial fans, valley fills, deltas and coastal barrier structures) can be conceptualised as storage components of an interrupted sediment cascade with four primary sources (rockwalls, drift-mantled slopes, valley-floor glacigenic deposits and coastal glacigenic deposits) and four terminal sediment sinks (alluvial valley-fill deposits, lacustrine deposits, coastal/nearshore deposits and shelf/offshore deposits). Paraglacial sediment stores and sinks may form major sources of readily erodible sediment during the early stages of glacial cycles, leading to high rates of sediment transport during periods of glacier or ice-sheet expansion. Probably because of the limited preservation potential of paraglacial sediments that were subsequently over-run by glacier ice, identification of paraglacial facies in both terrestrial and marine settings has been almost exclusively limited to sequences that post-date the Last Glacial Maximum.

The unifying concept of paraglacial geomorphology is that of glacially conditioned sediment availability. Relaxation of landscape elements to nonglacial conditions operates over timescales of 10¹–>10⁴ years, and is conditioned by both process and spatial scale. Rate of sediment reworking can be described by an exhaustion model. In the case of primary reworking of glacigenic sediment, the rate of reworking declines approximately exponentially through time, though extrinsic perturbation may rejuvenate paraglacial sediment flux long after termination of the initial period of paraglacial adjustment. Landscape-scale (particularly alluvial and coastal) systems may exhibit intrinsically complex responses due to reworking of secondary paraglacial sediment stores. The long relaxation time of such systems implies that many areas deglaciated in the Late Pleistocene or Early Holocene have still not fully adjusted (in terms of sediment supply) to nonglacial conditions.

Introduction

The retreat of glacier ice commonly exposes a landscape that is susceptible to rapid change. Debuttressing of glacially steepened rockwalls may result in slope failure or enhanced rockfall activity; unvegetated drift-mantled slopes are vulnerable to rapid reworking by debris flows, snow avalanches and slopewash; glacier forelands are exposed to wind erosion and frost action; and rivers entrain and redistribute large amounts unconsolidated glacigenic sediment that is subsequently redeposited in a variety of terrestrial, lacustrine and marine sediment sinks. Such accelerated geomorphic activity is termed ‘paraglacial’, a term first introduced by Ryder (1971a), Ryder (1971b) to describe alluvial fans in British Columbia that had accumulated through the reworking of glacial sediment by rivers and debris flows following Late Wisconsin deglaciation. The concept was subsequently formalised by Church and Ryder (1972, p. 3059), who defined ‘paraglacial’ as referring to ‘…nonglacial processes that are directly conditioned by glaciation’, adding that ‘…it refers both to proglacial processes, and to those occurring around and within the margins of a former glacier that are the direct result of the former presence of ice’. They also identified a ‘paraglacial period’ as the time interval over which paraglacial processes operate. Church and Ryder (1989) later emphasised the generality of their definition, specifying that it is applicable to all periods of glacier retreat (not merely Late Pleistocene deglaciation), and that the ‘paraglacial period’ is not restricted to the closing phases of glaciation but may extend well into the ensuing nonglacial interval.

The paraglacial concept as conceived by Church and Ryder (1972) introduced no new or distinctive geomorphological processes, but emphasised the relatively rapid adjustment of deglaciated landscapes to nonglacial conditions through the enhanced operation of a wide range of subaerial processes—such as slope failure, debris flow and fluvial reworking of sediment—that operate in a wide range of environments. The essence of the concept is that recently deglaciated terrain is often initially in an unstable or metastable state, and thus vulnerable to rapid modification by subaerial agents. Effectively, then, the ‘paraglacial period’ is the period of readjustment from a glacial to a nonglacial condition, as ‘…fluvial, slope and aeolian systems relax towards a nonglacial state’ (Benn and Evans, 1998, p. 261). As will become evident in the review that follows, different elements of paraglacial landsystems relax at very different rates: steep, sediment-mantled hillslopes, for example, may achieve stability within a few centuries of ice retreat, whereas large fluvial systems may still be reworking glacigenic sediment more than 10,000 years after deglaciation. It follows that the ‘paraglacial period’ not only varies markedly in different landscape contexts, but also with spatial scale.

Quaternary geomorphologists were initially slow to appreciate the versatility of the paraglacial concept and particularly its relevance to the understanding of Holocene landforms and landscape evolution. For nearly two decades after its initial formulation, only a handful of studies made explicit reference to the idea. Most of these concerned paraglacial reworking of glacigenic sediment in the North American Cordillera, mainly in the context of alluvial fan formation, the development of terraced valley fills, enhanced sediment yield and patterns of lacustrine sedimentation (e.g. Roed and Waslyk, 1973; Slaymaker (1977), Slaymaker (1987); Slaymaker and McPherson, 1977; Luckman, 1981; Jackson et al., 1982; Hammer and Smith, 1983; Leonard (1985), Leonard (1986); Clague, 1986; Kostachuck et al., 1986). The past 12 years, however, have witnessed a remarkable flourishing of research devoted to paraglacial processes, paraglacial sediment transport and paraglacial landscape modification (Fig. 1). Several trends are evident in these more recent studies. First, there has been a marked diversification in the geomorphic contexts in which the paraglacial concept has been explicitly employed, including studies of rock-slope failure and enhanced rockwall retreat following deglaciation, paraglacial adjustment of drift-mantled hillslopes, aeolian and periglacial modification of glacier forelands and research on the nature and evolution of ‘paraglacial’ coasts where reworking of glacigenic sediments in estuarine and littoral environments has influenced coastal morphology and sediment transport. A second development has been a focusing of research on present-day paraglacial processes and landsystems, particularly in areas deglaciated over the last 200 years as glaciers retreated from their Little Ice Age maxima, thus providing analogues for rapid landscape adjustment during and following widespread deglaciation in Late Pleistocene and Early Holocene times (Fig. 1B). Thirdly, the paraglacial concept has been employed as a framework for research across a wide range of contrasting deglacial environments, ranging from tectonically stable mountains such as those of Scandinavia to tectonically active mountains such as those of the Karakoram Himalaya, and from cold, arid Antarctic environments to mid-latitude coasts. Finally, there has been a growing appreciation of the palaeoenvironmental significance of paraglacial facies in Quaternary stratigraphic successions.

The aims of this paper are fourfold: (1) to provide a comprehensive review of paraglacial research across a wide range of geomorphic contexts, ranging from rock-slope instability through the reworking of glacigenic drift by mass-movement, fluvial and aeolian processes to the deposition of reworked sediment in terrestrial, lacustrine and coastal environments; (2) to explore the significance of paraglacial landscape modification and sediment recycling in the overall context of alternating glacial/nonglacial landscape evolution; (3) to assess the nature and significance of paraglacial facies in Quaternary sediment sequences; and (4) to provide a general model of the sequence of paraglacial landscape modification, and the changing nature of resulting paraglacial landsystems. In recognition of the widening context in which the paraglacial concept has been employed, particularly over the last decade, the working definition of ‘paraglacial’ adopted here is:

nonglacial earth-surface processes, sediment accumulations, landforms, landsystems and landscapes that are directly conditioned by glaciation and deglaciation.

This revised definition retains the essence of the concept as defined by Church and Ryder (1972) whilst recognising that the use of the term has extended from a descriptor of processes, to a descriptor of resultant landforms, sediment facies and landscapes or landsystems. It is common, for example, to find references to ‘paraglacial debris cones’, ‘paraglacial alluvial fans’, ‘paraglacial landslides’ and ‘paraglacial coasts’, terms which demonstrate that the current use of the term is rather more eclectic than that originally envisaged. Certain uses of the term are, however, excluded from consideration here. Glacio-isostatic rebound, though a consequence of glacial thinning and retreat, is not considered a paraglacial process on the grounds that it is an indirect tectonic response rather than a process operating at the earth's surface, though of course glacio-isostatic uplift may influence various paraglacial processes, notably incision of valley fills and marine deltas. The notion that the term ‘paraglacial’ may also encompass edaphic and phytological processes on recently deglaciated terrain (Matthews, 1992; Matthews et al., 1998) is also largely excluded from this review, which focuses specifically on geomorphic processes and consequences.

The first and most extensive part of this review considers the operation of paraglacial processes in a range of geomorphological contexts: (1) adjustment of rock slopes; (2) adjustment of sediment-mantled slopes; (3) paraglacial modification of glacier forelands, including aeolian processes; (4) paraglacial fans, debris cones and valley fills; (5) sediment yield and fluvial sediment transport; (6) lacustrine sedimentation and (7) paraglacial coasts and glaciated shelves. The review then considers the representation of paraglacial sediments in the Quaternary stratigraphic record, and concludes by proposing a general model for paraglacial modification and the evolution of paraglacial landsystems.