Search

Geophysical surveys in the Lakes Tazenat, Aydat, Chambon and Lacassou combined with a multiproxy study of sediment cores highlight Late Holocene abrupt environmental changes in the northern part of the French Massif Central (FMC). Dating landslide-induced dust layers and subaqueous slope failures impacting lake basin sedimentary fills suggest regional triggering by palaeo-earthquakes (cluster of events) either in the Puy-de-Sancy volcano area, or along the regional Limagne fault. In the present study, a cluster of palaeo-earthquakes occurring in the Sancy area between AD 1243 and 1270 are documented in the Lakes Montcineyre, Chauvet, Pavin and Guery, as well as by the formation of Lake Lacassou following the last emplacement of the Dent du Marais landslide in the Chaudefour glacial valley around AD 1250 ± 30. Former cluster of earthquakes occurring near the Limagne fault between AD 580 and 650 are evidenced in Lake Tazenat and in maar Lake Pavin. Lake Pavin present lake-level was similarly favoured by the cluster of earthquakes between AD 1243 and 1270 and the occurrence of a slump and an outburst flood event resulting in the bypassing of a palaeo-outlet. Holocene earthquakes near the Limagne fault are suggested in maar Lake Tazenat by coeval mass wasting deposits (MWDs) around 2250 ± 50 cal. BP and around 2730 ± 30 cal. BP. A landslide-induced dust layer dated between 2760 and 2520 cal. BP at the transition from paludal to lacustrine deposits in the Chaudefour glacial valley is suggesting that the formation of Lake Chambon resulted from a first emplacement of the Dent du Marais landslide. Radionuclides and radiocarbon dating in Lake Tazenat suggest variable sedimentation rates over the last millennia due to: (i) floods during the Little Ice Age, (ii) coeval MWDs at the basin edges and an erosive turbidite in the central basin between AD 575 and 625, and (iii) lake eutrophication since AD 1945. More gravity cores are needed to directly sample and date the youngest generation of coeval MWDs detected on seismic profiles in order to test the impact in this lake of the cluster of major historical earthquakes in the FMC between AD 1450 and 1490 near the Limagne fault.

In this study, a glacial lake-outburst flood was documented in the immediate foreland of the maximum phase of the Wartanian (Warthian, Late Saalian, MIS 6) Glaciation at Adamow in south central Poland. The structural and textural analyses of the sediments, as well as palaeohydraulic analysis, were carried out in the proximal part of the outwash plain where a fragment of the floodwater drainage system comprising buried erosional channels have been documented. Their fill reached a thickness of up to 11 m and was composed of poorly sorted, massive gravels with boulders, as well as horizontally, sub-horizontally, low-angle, and cross-stratified gravels. The coarse sediments were found to be inversely and normally graded. These sediments constituted the records of two subsequent flooding episodes. In the channel fill, the lenticular accumulations of gravel were identified and interpreted as buried large gravel dunes, while exhibiting a thickness of up to 2.5 m. Traces of sediment diapirism at the contact of channel fill and underlying sediments were also documented. Deformation was observed to originate from liquefaction caused by the rapid depositional loading of flood deposits. The example used in this study exhibited that the impact of glacial lake-outburst floods in shaping the marginal zones and foreland of ice sheets in the North European Plain was higher than that considered earlier. This example also showed a different case of glacial outburst flood in the foreland of the Middle Pleistocene ice sheets of Europe than that already known, which includes the discharge of sub- or englacial meltwater reservoirs emanated from a large proglacial ice-dammed lake.

Landslide damming of valleys may have a significant effect on the evolution of fluvial landscapes. The southeastern Tibetan Plateau is highly prone to landslide damming due to its deeply incised valleys and young, active geology. Here we present a sedimentological investigation of a landslide-dammed lake in the upper reach of Lulang River on the eastern Himalayan syntaxis. Detailed interpretation of lacustrine varve, lakeshore, and subaqueous delta sediment shows that Lulang landslide-dammed lake (LLDL) had a surface elevation of 3360 m a.s.l., an area of 33 x 10(6) m(2), and a volume of 6.5 x 10(2) m(3). Radiocarbon dating of six organic clay lacustrine samples demonstrate that LLDL formed before 24.2 ka cal. BP and persisted for at least 8.8 ka. Sediment infill reached full capacity soon after 15.4 ka cal. BP and fluvial conditions then prevailed. Sedimentary evidence and morphological analysis of cross sections along the LLDL outlet channel revealed a larger, earlier lake with a total volume of 4.3 x 10(9) m(3) and water surface elevation at 3570 m a.s.l. that failed catastrophically; the resulting outburst flood drained 98.5% of the impounded lake. The remnant lake maintained a relatively stable water level of 3360 m a.s.l. for its lifetime of at least 8.8 ka. Long profile and steepness index analysis indicate that the location of the catastrophic dam failure and subsequent long-term blockage is associated with a knickpoint. The origin of the knickpoint may have been polygenetic, however, the landslide damming seems to have increased its distinctiveness and persistence. The relatively wide valley and low channel slope characteristic of reaches upstream of LLDL is likely to have resulted from glacial activity, rather than backwater aggradation which had only a local influence. (C) 2019 Elsevier B.V. All rights reserved.

The first regional model for the deglacial history of the Beaufort margin slope, in the western Arctic, is presented. The conceptual model was developed using new high-resolution CHIRP seismic reflection, multibeam bathymetry, and sediment core data acquired along the margin. This synthesis provides important constraints on sediment source and dispersal patterns for the last deglaciation. The continental slope from Barrow Canyon to the Mackenzie Trough is characterized by thick Holocene sediments mostly sourced from Barrow Canyon and continental discharge. This acoustically transparent, fine-grained unit overlies acoustically laminated sediments sourced from the Mackenzie River and ice rafting. The deglacial history of the margin from Mackenzie Trough to the Amundsen Gulf is characterized by many ice rafting and meltwater discharge events. Ice rafted debris layers were deposited around ∼14.6 ka and ∼14.1 ka and likely document enhanced ice shedding events from the Amundsen and M'Clure ice streams as they retreated. There are three inferred meltwater discharge events, all sourced from the Mackenzie region. The oldest discharge event had two phases, one sometime between ∼14.5 ka and ∼14.2 ka and another between ∼14.0 ka and ∼13.0 ka. This discharge deposited finely laminated sediments more than 7 m thick sourced from proglacial lakes in the area. Following this was a flood event starting at ∼12.94 ka, which generated high amplitude reflectors, deposited coarse debris, and caused a lowering in the δ18O record. This is possibly a major outburst flood from glacial Lake Agassiz with an additional component of Agassiz meltwater that was diverted from its southern drainage pathway down the Mississippi to the Gulf of Mexico. Finally, a third discharge event was initiated by ∼11.3 ka, which deposited coarse laminated sediments focused in the Mackenzie Trough. Timing of these second and third events correlates with the onset of the Younger Dryas cold period and the preBoreal oscillation event, respectively, suggesting that fresh water discharge may have reached a tipping point, leading to those climate events.

Petrographic and geostatistical analysis of linear dune forms on the Hanford Site, south-central Washington, have resulted in data supporting reinterpretation of depositional environments from eolian to primary glaciofluvial with post-depositional eolian modification. Miocene basalt flows of the Columbia River Basalt Group form the bedrock across the region overlain by Pliocene fluvial deposits from the preglacial Columbia River. Between ~21,000 and 12,000 YBP, glacial outburst floods scoured eastern Washington forming the Channeled Scablands. In the Pasco Basin, floodwaters were ponded behind flow restrictions resulting in deposition of coarse-grained gravel and sand sheets capped by fine-grained cyclic Touchet beds in higher elevation peripheral canyons. Outflow of ponded water mobilized the upper few meters of sediment forming a series of giant current ripples or megaripples. Modification of the surfaces of the megaripples under the influence of the Holocene wind regime has produced an eolian surficial signature. Subsurface excavations in these dune forms revealed the primary glaciofluvial provenance. Sedimentologic and petrographic evidence supports a general winnowing of fine-grained sediments from west to east under the current wind regime and includes both quartz and heavy minerals. These processes have resulted in the formation of an active dune field along the western bank of the Columbia River that overlies flood deposits locally.

Glacier recession and formation of glacial lakes along with outburst flood hazards are increasing in response to changing climate in the Himalayan region. This study describes evolution of Dalung and Padam glaciers and their associated lakes employing multi-temporal datasets of Landsat (MSS, TM, ETM+), Indian Remote Sensing LISS-III, Google EarthTM, ASTER GDEM and Survey of India topographic maps. The Dalung Glacier (DG) and Lake (DL) are present in Suru River catchment whereas the Padam Glacier (PG) and Lake (PL) are present in Tsarap River catchment of Zanskar basin, western Himalaya. From 1962 to 2015; the DG and PG have shrunk at different rates. The DG has shrunk by 2.6 ± 0.56 km2 (19.2 ± 4.1%) at a rate of 0.05 ± 0.01 km2 a−1 whereas, the PG has shrunk by 3.4 ± 0.65 km2 (12.7 ± 2.4%) at a rate of 0.06 ± 0.01 km2 a−1. The frontal recession was recorded as 1.240 ± 0.02 km (15.7 ± 2.0%) at a rate of 20 ± 0.4 m a−1 for DG and 2.690 ± 0.5 m (21.3 ± 2.1%) at a rate of 50.7 ± 0.1 m a−1 for PG. Consequently, the proglacial glacial lakes have formed indicating an expansion of 20 ± 3.4% (DL) and 58 ± 4.5% (PL) during 1990–2015. Using integrated criteria of standard indices for the identification of potentially dangerous glacial lakes (PDGLs), the lakes were found to be vulnerable and have a potential to cause breaching in the coming years. In order to provide inputs for the emergency action plan, one dimensional MIKE 11 Model was used to generate the peak flood hydrographs at the lake and other vulnerable sites downstream. Highest peak discharge of 3361.7 m3 s−1 (DL) and 4210.3 m3 s−1 (PL) were estimated at the lake sites which has reduced to 1520.9 m3 s−1 and 1722.4 m3 s−1 respectively at the outlet sites. The reduced discharge at downstream sites still bears the potential to create dangerous flood hazards.

The last deglaciation of the Laurentide Ice Sheet in northern Quebec and Labrador led to the formation of several glacial lakes that drained into Ungava Bay and the nearby Labrador Sea. Assessing the potential impact of their drainage on ocean surface conditions and climate, however, is limited by the few existing age constraints on ice retreat and associated evolution of these lakes. Here we report 21 10Be ages from shorelines and an outburst flood landform formed by Lake Naskaupi, one of the largest glacial lakes in this region. The results indicate that the lake drained from its full extent at 8300 ± 300 a, suggesting that it may have contributed to the freshwater forcing initiated by the drainage of Lake Agassiz-Ojibway that caused the 8.2-ka cold event. Additionally, the results provide important constraints on the position of the ice margin of the Labrador Sector during its retreat across this region.

High resolution digital terrain models (DTMs) generated from airborne LiDAR data and supplemented by field evidence are used to map glacial landform assemblages dating from the last glaciation (Midlandian glaciation; OI stages 2–3) in the central Irish Midlands. The DTMs reveal previously unrecognised low-amplitude landforms, including crevasse-squeeze ridges and mega-scale glacial lineations overprinted by conduit fills leading to ice-marginal subaqueous deposits. We interpret this landform assemblage as evidence for surging behaviour during ice recession. The data indicate that two separate phases of accelerated ice flow were followed by ice sheet stagnation during overall deglaciation. The second surge event was followed by a subglacial outburst flood, forming an intricate esker and crevasse-fill network. The data provide the first clear evidence that ice flow direction was eastward along the eastern watershed of the Shannon River basin, at odds with previous models, and raise the possibility that an ice stream existed in this area. Our work demonstrates the potential for airborne LiDAR surveys to produce detailed paleoglaciological reconstructions and to enhance our understanding of complex palaeo-ice sheet dynamics.

Glacial lake outburst floods (GLOFs) have recently become one of the primary natural hazards in the Himalayas. There is therefore an urgent need to assess GLOF hazards in the region. Cirenmaco, a moraine-dammed lake located in the upstream portion of Zhangzangbo valley, Central Himalayas, has received public attention after its damaging 1981 outburst flood. Here, by combining remote sensing methods, bathymetric survey and 2D hydraulic modeling, we assessed the hazard posed by Cirenmaco in its current status. Inter-annual variation of Cirenmaco lake area indicates a rapid lake expansion from 0.10 ± 0.08 km2 in 1988 to 0.39 ± 0.04 km2 in 2013. Bathymetric survey shows the maximum water depth of the lake in 2012 was 115 ± 2 m and the lake volume was calculated to be 1.8 × 107 m3. Field geomorphic analysis shows that Cirenmaco glacial lake is prone to GLOFs as mass movements and ice and snow avalanches can impact the lake and the melting of the dead ice in the moraine can lower the dam level. HEC-RAS 2D model was then used to simulate moraine dam failure of the Cirenmaco and assess GLOF impacts downstream. Reconstruction of Cirenmaco 1981 GLOF shows that HEC-RAS can produce reasonable flood extent and water depth, thus demonstrate its ability to effectively model complex GLOFs. GLOF modeling results presented can be used as a basis for the implementation of disaster prevention and mitigation measures. As a case study, this work shows how we can integrate different methods to GLOF hazard assessment.

Unusual channel-amphitheatre landforms are present in Late Pleistocene–early Holocene, subaqueous fan and delta deposits in the glacial Lake Fraser basin, central British Columbia. The lake formed during the decay of the last Cordilleran Ice Sheet and drained ~11,500 years ago during a large outburst flood. The fronts of a delta and two subaqueous fans consisting of silt to fine sand are marked by branching networks of wide, nearly flat channels that terminate upstream in digitate, steep-walled amphitheatres. We propose that these channel-amphitheatre landforms formed by liquefaction flowslides that were induced by the rapid drawdown of glacial Lake Fraser during the outburst flood. Similar geomorphic forms, which we believe also to be associated with rapid drawdowns of large Late Pleistocene–early Holocene lakes, occur elsewhere in North America. A recent tailings dam failure and an intentional breaching of a 100-year-old hydroelectric dam provide insights into the processes responsible for the landforms. By using geomechanical analysis, we show how rapid lake drawdown can trigger liquefaction flowslides in deposits of silt to fine sand. The novelty of our approach lies in combining geomechanical reasoning with geomorphic analogues to understand histories of ancient glacier-dammed lakes and of the glacial lake outburst floods that are sourced from them.

This study reveals that a distinct damming incident occurred during a prehistory period (approximately 30 ka BP) in the upper Minjiang River, at the southeastern margin of the Tibetan Plateau, which formed a giant dammed lake. It can be inferred that paleoseismic induced major landslides were the direct cause of the river damming. The dammed lake breached and caused a large outburst flood at about 27 ka BP. Between the section of the prehistoric ancient dam-body and its downstream areas (5 km), an abundance of diamicts consisting of a poor-sorted wide range of particle sizes, including gravel, sand, and minor silt and clay, have been observed. The outburst deposits in the Diexi area maybe divided into two periods: the thick outburst deposits represented the earlier and largest dammed-break event, and the deposit terrace was due to the subsequent and smaller dammed-break event. The sedimentary structures of the outburst deposits included imbrication, cavitation, matrix, and massive structures, as well as gravel support-stacked structures. Furthermore, a special sedimentary structure became evident, consisting of a rhythmite interbedded structure composed of a coarser gravel layer, and a finer sand and gravel layer, with no sand lens formed within the layers. This special structure can potentially provide an important original criterion for the outburst deposits in this area. The two types of facies associations have been recorded which indicate contrasting sediment transport regimes. The characteristics of boulder facies were most likely associated with hyperconcentrated flood flows. However, the finer-grained Gm/Gfm/GRm facies, which overlaid the Bcm facies, were determined to represent the waning stage of the flooding events. The delicate structures, as well as the absence of coarse components in the Sh facies, indicated an episode of accumulations, which was subsequent to the deposition of the underlying Bcm facies. The results of this study potentially provide important implications for the interpretation of the diamicts in the Diexi area. The grain-size frequency curves of the outburst deposits in the Diexi area were found to all be bimodal (a main peak and an inconspicuous sub-peak). These findings indicated a poorly sorted sedimentary structure. It was found that, from the upstream segment to the downstream segment, the gravel sediment became gradually lessened, while the fine-grained sediment increased. Additionally, the outburst flood hydrodynamic intensity was found to have gradually become weakened.

Alpine glaciers and permafrost are sensitive to climate change, and their degradation due to annual temperature increasing has already induced many mass movements such as debris flows. On September 4 of 2007, July 25–31 and September 5–8 of 2010, three large debris flows took place in Tianmo gully, a left-bank tributary of Parlung River, southeast Tibetan Plateau. The debris flows blocked the river, and a section of highway 450 m long including a bridge 76 m long were destroyed by the subsequent outburst flood. This paper, based on post-event investigations, witness accounts, news reports and satellite image interpretation, systematically analyses the geo-environments, climate conditions and sources of these debris flow events. Their differences in flow process, surge numbers and velocity suggest that they resulted from different geohazard processes, cascades and initiation mechanisms. The 2007 debris flow originated from a bare rock/moraine collapse next to the cirque due to strong alternation of wet and dry conditions. The mass moved down the gully with velocity of 30–40 m/s estimated by back-calculation using superelevation and run-up. It entrained moraine, avalanched snow-ice and water from the channel, and transformed into high-speed debris flow, crossing the Parlung River and finally depositing onto the highway. The other two events in July and September of 2010 initiated from two channel-bank landslides triggered by melt-water and concentrated rainfall, which dammed the channel. The subsequent landslide dam failure generated debris flows over several days, with velocities of 12–14 m/s, that temporally blocked Parlung River. The initiation mechanisms of most large debris flows recorded in Parlung region are similar to the events in Tianmo gully, originating either from rock/moraine avalanches or from the collapse of a landslide dam. This implies that the periglacial degradation of bedrock and moraine is the key process to be monitored and assessed under climate warming. The paper applies the geohazard mechanisms and cascade to hazard monitoring in order to protect the existing Sichuan-Tibet highway and the forthcoming railway.

Outburst floods originating in moraine-dammed lakes represent a significant geomorphological process as well as a specific type of threat for local communities in the Cordillera Blanca, Peru (8.5°–10° S; 77°–78° W). An exceptional concentration of catastrophic floods has been reported from the Cordillera Blanca in the first half of 20th Century (1930s–1950s), leading to thousands of fatalities. The main objective of this paper is to provide a revised and comprehensive overview of geomorphologically effective floods in the area of interest, using various documentary data sources, verified by analysis of remotely sensed images (1948–2013) and enhanced by original field data. Verified events (n = 28; 4 not mentioned before) are analysed from the perspective of spatiotemporal distribution, pre-flood conditions, causes, mechanisms and geomorphological impacts as well as socioeconomical consequences, revealing certain patterns and similar features. GLOFs are further classified according to their magnitude: 5 extreme events, 8 major events and 15 minor events are distinguished, referring to the quantified geomorphological and socioeconomical impacts. Selected moraine dams and flood deposits are dated using lichenometric dating. Special attention is given to moraine dam breaches - the most frequent type of water release with the most significant consequences. Selected major events and their consequences are studied in detail in a separate section. Finally, a general schematic model of lake formation, growth and post-flood evolution reflecting initial topographical setting and glacier retreat is introduced and the utilization of the obtained results is outlined.

The environmental challenges posed by global warming in the Himalayan region include early and rapid melting of snow and glaciers, creation of new lakes, and expansion of old ones posing a high risk of glacial lakes outburst flood (GLOF) hazard for downstream communities. According to various elevation ranges, 3044 lakes were analyzed basinwide in the Hindu Kush-Karakoram-Himalaya (HKH) ranges of Pakistan using multisensor remote sensing data of the 2001–2013 period. An overall increase in glacial lakes was observed at various altitudinal ranges between 2500 and 5500, m out of which noticeable change by number was within the 4000–4500 m range. The analysis carried out by glacial-fed lakes and nonglacial-fed lakes in different river basins indicated variable patterns depending on the geographic location in the HKH region. The correlation analysis of parameters like lake area, expansion rate, and elevation was performed with 617 glacial lakes distributed in various river basins of the three HKH ranges. Lake area (2013) and elevation showed a negative relationship for all basins except Hunza, Shigar, and Shyok. The correlation between the expansion rate of lakes and elevation was on the positive side for Swat, Gilgit, Shigar, and Shingo basins—a situation that may be attributed to the variable altitudinal pattern of temperature and precipitation. In order to explore such diverse patterns of lake behavior and relationship with influential factors in the HKH, detailed studies based on using high resolution image data coupled with in situ information are a prerequisite. Although an increase in lake area observed below 3500 m would be favorable for water resource management, but could be alarming in context of glacial flood hazards that need to be monitored critically on a long-term basis.

We present the study of a subglacial drainage system that developed in the base of the Scandinavian Ice Sheet at the border of the Silesian Lowland and Silesian Upland (southern Poland) during the Drenthian Glaciation as a result of large outburst flood events. This drainage system was characterised by a spatially complex structure, i.e. it was composed of an individual an N-channel (tunnel channel) incised in Mesozoic rocks, i.e. Cretaceous and Triassic marls and limestones, and an R-channel, deposits of which form a set of large esker ridges which also occur on bedrock in the SE prolongation of the tunnel channel. Unlike most known cases of the co-occurrence of such forms, the studied tunnel channel and eskers are directly related, having been formed during the same flow events. The studied forms are unique within the entire southern sector of the Scandinavian Ice Sheet, which, in Central Europe, advanced mainly over soft unconsolidated Quaternary and Neogene sediments. The tunnel channel was formed as a result of enormous erosion of the ice sheet bed. The material eroded from the channel was partially deposited in a subsequent part of the drainage system. The location and orientation of the drainage system did not result directly from water pressure gradients, but was strongly determined by the internal structure of the ice sheet, i.e. it presumably developed along the marginal part of the ice stream supplying the Upper Odra ice lobe, which represented distinct weaknesses within the ice sheet. The development of the drainage system was the effect of the drainage of meltwater stored in the ice sheet system, most probably as a supraglacial lake. Gravelly-sandy rhythms recorded within the esker succession indicate that the process of lake drainage took place in several stages, or that the lake developed and drained several times.

In June 2013, the “Himalayan Tsunami” occured in the Indian Himalayas. Heavy rain triggered a lake outburst flood, landslides and flash floods. An estimated several thousand people were killed, mainly around the Hindu temple in Kedarnath, and the tourist infrastructure between four Hindu temples was massively destroyed along with the source of income it had provided for the local population. This article addresses the nexus of tourism, vulnerabilities and disasters and analyzes its manifestations during and after the events of 2013. What role did tourism play in the disaster as a driver of social vulnerability during and especially after the events of 2013? To answer this research question, the paper describes the historical development of tourism in the region, tourism-related vulnerabilities and their actualization by the “Himalayan Tsunami” and in the process of reconstruction. The research shows that the disaster continues years after its supposed end due to tourism-related vulnerabilities, and the significance of secondary effects of the disaster threatens livelihoods more than the immediate events.

Near the Pleistocene Termination, a glacier-dammed lake in central British Columbia suddenly drained to the south along the Fraser River valley. Floodwater travelled 330 km down the valley to Hope, British Columbia, and from there to the west into the Salish Sea near Vancouver. The flood was caused by the failure of an ice dam formed by the terminus of glaciers flowing from the central Coast Mountains across the British Columbia Interior Plateau. The ice dam impounded several hundred cubic kilometres of water to a maximum elevation of about 810 m asl (above sea level); at its maximum, the lake at the ice dam was over 250 m deep. Geomorphic and sedimentary evidence for the flood includes streamlined boulder-strewn bars, gravel dune fields, and terraces sloping up Fraser and lowermost Thompson valleys, opposite the present direction of river flow. The gravel bars and flood terraces are underlain by sheets of massive to poorly sorted gravel containing large boulders and rip-up clasts of silt and till. Shortly after the flood, a landslide near the northern margin of the former glacier dam impounded water to an elevation of about 550 m asl. This lake emptied due to overflow and incision of the landslide dam. The outburst flood from glacial Lake Fraser and the subsequent draining of the landslide-dammed lake deeply incised the older sediment fill in Fraser Valley and transported much of this sediment into the proto-Salish Sea west of Vancouver, British Columbia and Bellingham, Washington. TCN ages on flood-transported boulders at three localities along the flood path agree with radiocarbon ages on inferred flood layers in ODP cores collected from Saanich Inlet, a fiord on southern Vancouver Island, 80 km south-southwest of Vancouver.

In this paper we present a suite of erosional remnants, mega deposits and subtle bar morphology that we relate to the outburst flood from the glacial lake Nedre Glomsjo at the end of the last Ice Age. By using large datasets of airborne LiDAR data implemented in a geographic information system (GIS), we have mapped flood related features along the Glomma and Vrangselva rivers in southeastern Norway. The unprecedented overview of the valley reaches obtained by the vegetation-free LiDAR-derived digital elevation models (DEM) has revealed a set of hitherto undocumented landforms. Persisting erosive lines – indicators of the uppermost flooded level – are carved into surficial deposits in the hillsides and are found as high as 80–90 m above the modern valley floor. By using the indicators as an upper flood boundary, we have computed cross-sectional profiles showing that the flood in some reaches inundated more than 120 000 m2 of the valley. Large, streamlined bed forms, which we interpret as flood bars, drape sections of the valley floor, some several kilometers long. The most morphologically striking – pendant bars – are developed behind flood flow projections, such as bedrock knolls or in lee of a valley bend. Flood bars occur in the entire study area, but are more widespread in the north and generally decrease in size moving in a downstream direction. Kettle holes and ice-block obstacle marks from icebergs arrested during the flood are common. These features support the theory of a catastrophic drainage event, but also indicate a pattern of differential erosion and deposition that allowed us to interpret palaeoflow on individual bars. Vast aeolian dune fields in the region are interpreted as a secondary product of the flood, as deposits related to the event were mobilised by northerly winds momentarily after the flood waned. The dune fields cover an excess of 50 km2 and reveal that the region was a highly active periglacial desert after the flood. Our mapping highlights the outburst flood's role as a landscape-defining event. Morphological evidence determine the southern fringe of the Scandinavian ice sheet to c. 15 km north of Elverum at the time of the outburst flood. From calibrating existing 14C dates we postulate an age of c. 10–10.4 cal ka BP for the Nedre Glomsjo drainage event and the contemporaneous margin of the ice sheet.

A set of sedimentary cores, high resolution swath bathymetry and subbottom profiler data provides new insights on sedimentary processes in meromictic maar Lake Pavin, France. Three sedimentary environments (i.e., littoral, plateau and basin) have been identified in the lake from sediment composition using bulk organic geochemistry and the analysis of hydroacoustic images. Various forms of rapidly deposited layers (RDLs) have been identified and radiocarbon dated. An up to date stratigraphy of sedimentary events matching coeval RDLs across the lake is presented and illustrates a wide range of natural hazards linked to Lake Pavin during the last 2000 years. In AD 600, a sudden lake outburst triggered a slump deposit along with a 9 m lake-level drop that drove shifts in sedimentary organic matter composition. Outside the lake, outburst flood deposits have been described downstream and provide sedimentary evidence for this event. The lake-level drop also favored the generation of gravity reworking processes, as shown by (1) a regional earthquake-triggered large slope failure on the plateau connected to a masswasting deposit in the basin dated to AD 1300, and (2) a succession of turbidites in AD 1825 and AD 1860 contemporaneous to two historic earthquakes, suggesting that this lake is sensitive to earthquakes with a minimum epicentral intensity of V. Finally, past observations of lake water color changes in AD 1783 and AD 1936, similar to reports in other meromictic lakes, match iron-rich deposits identified in maar lake sediments and suggest that Lake Pavin could have undergone limnic eruptions.

Even though Patagonia is ideally located to study climate of the southern mid-latitudes, many questions on the late Quaternary climate evolution remain unresolved. The timing of maximum glacier extent is still uncertain in vast areas, and the postglacial evolution of the Southern Westerly Wind Belt (SWWB) remains highly debated. Here, we study the sedimentary infill of a glacigenic lake (Lago Castor; 45.6°S, 71.8°W) located at the leeside of the Andes in Chilean Patagonia to i) reconstruct the deglacial evolution of the eastern flank of the Patagonian Ice Sheet (PIS), and ii) discuss postglacial changes in wind strength at a critical location where westerly wind records are critically lacking. A dense grid of high-resolution reflection-seismic data was used to reconstruct the large-scale infill history of the lake, and a radiocarbon dated sediment core penetrating all lacustrine seismic units, was retrieved. Results indicate that the deglaciation of the lake basin and its catchment occurred no later than ∼28 cal kyr BP (i.e. an early LGM), but possibly even already after MIS 4. Afterwards, the Lago Castor area was covered by a large proglacial lake that drained – possibly through an outburst flood – when the PIS outlet glaciers retreated to a critical location. Subsequently, very dry conditions caused the lake to desiccate, as evidenced by an unconformity visible on the seismic profiles and in the sediment core. This dry period likely resulted from the increased orographic effect of the PIS-covered Andes, accompanied by weaker westerlies. From ∼20 kyr BP onwards, the combination of a shrinking PIS and a southward shift of the SWWB resulted in increased precipitation, which caused the lake level to rise. After ∼17 cal kyr BP, lake sedimentation was more directly influenced by the southern westerlies, with the formation of sediment drifts resulting from strong bottom current during periods of intense westerly winds. Our results suggest a progressive increase in wind strength at 46°S from 11.2 to 4.5 cal kyr BP, which supports the hypothesis that the SWWB broadened during the early and middle Holocene.

During the last glaciation (marine oxygen isotope stage 2), outburst floods from glacial Lake Missoula deposited diagnostic sediments within glacial Lake Columbia. Two dominant outburst flood lithofacies are present within glacial Lake Columbia deposits: a flood expansion bar facies and a finer-grained hyperpycnite facies. We conclude that the flood sediments have a glacial Lake Missoula source because: (1) current indicators indicate westward flow through the lake, and upvalley flow followed by downvalley flow in tributary valleys; (2) no flood sediments are found north of a certain point; (3) there is a dominance of Belt–Purcell Supergroup clasts in a flood expansion bar; and (4) some of the finer-grained beds have a pink colour, reflective of glacial Lake Missoula lake-bottom sediments. A new radiocarbon age of 13,400 ± 100 14C BP on plant detritus found below 37 flood beds helps constrain the timing of outburst flooding from glacial Lake Missoula.

Unlike glaciers in other parts of the world, Karakoram glaciers seem to be stable or gaining in mass in response to global climate change, a phenomenon known as ‘the Karakoram anomaly’. Many of the glaciers experience irregular, frequent, and sudden advances (surges) that pose an increasing threat of ice dam lake formation and subsequent outburst flooding throughout the region. In this study, we document 179 glacial lake outburst floods (GLOFs) that occurred from 1533 to 2020 in five major valleys. Sixty-four of the events took place after 1970, and 37 of these had remote sensing imagery that covered the GLOF formation to breaching sequence. Thirty-six glaciers were associated with GLOFS due to ice-front advance building ice barriers in rivers. The Kayger and Khurdopin glaciers are the most hazardous examples, being responsible for 31.8% of major GLOFs in the entire Karakoram. Using a cross-correlation feature-tracking technique on remote sensing imagery, we analyzed ten surge glaciers and documented six surge events from 1990 to 2019. Results show periodic surge cycles for the Khurdopin, Kyager, Shishper, and Chilinji glaciers of c. 15–20 years, with a surge velocity in the mid-2010s higher than that of the late 1990s for all studied glaciers. The higher velocity of a glacier increases the risk of flooding downstream of the terminus because the transfer of a huge ice mass towards the terminus during the surge is a key factor for formation and reformation of series of ice-dammed lakes, thus determining the magnitude and frequency of outburst flood events. The response of Karakorum glaciers to global warming and climate forcing, comprising a continuum of glacier mass gain, ice thinning and ice advance, has resulted in lake formation and ice dam failures. We predict the frequency of GLOFs will increase in the future. These findings support the increasing anomalous behavior of glaciers in the Karakoram region. To synthesize the detailed observations, a conceptual model is presented of ice-dammed lake formation and GLOF initiation in response to glacier surging.

Glacier lake outburst floods (GLOFs), especially those in the Arctic, can deliver exceptionally high volumes of sediment and solutes to fjords and shallow-marine settings, in comparison to typical seasonal river flows. These sediments and solutes strongly affect coastal geomorphology and aquatic ecosystems, yet are rarely observed. In this study, we have quantified the short-term geomorphological response of the most distal part of the Zackenberg River (northeast Greenland), where it enters Young Sund, to a glacier lake outburst flood that occurred on August 6th 2017. The main aims were to: (1) quantify riverbank and floodplain geomorphology changes that occurred as a consequence of the flood; (2) analyse the spatial patterns of those geomorphological changes and suggest the key controls on them. We used a time-series of very high-resolution UAV-generated images taken on the 5th, 6th, and 8th of August 2017, which enabled us to compare pre- and post-flood fluvial geomorphology. The GLOF induced intense and widespread geomorphological changes, which was surprising because several floods of a similar magnitude have occurred along this river. Approximately 30% of the area of interest experienced changes that were larger than the minimum level of detection (0.15 m). Lateral erosion reached almost 10 m in some places. The total volume loss from bank erosion was at least 26,561 m3 (+/− 14%), whereas the deposition was at least 7745 m3 (+/− 39%). Such an intensive geomorphological response resulted from a combination of factors; namely: (1) bank geometry; (2) composition of bank material; (3) time of occurrence of the event; (4) presence of permafrost; (6) channel geometry; and (7) multitude and diversity of geomorphological processes. We speculate the severity of the geomorphological impact relative to that from previous floods could have been due to warming air temperatures that provided sediment from thawed permafrost, and to an aggrading delta that raised the river base level. Overall, we contend that climate warming will not only make outburst floods from glaciers more likely but that those floods will achieve more geomorphological work with mechanical erosion of permafrost. Erosion and gravitational failures during future flood events will perhaps become even more widespread and intense.

Glacial lake outburst floods (GLOF) are one of the most destructive natural disasters. Understanding GLOF evolution, and their impacts, plays a fundamental role in GLOF hazard assessment and risk management. Reconstructing historical GLOFs is an important exercise because detailed case studies of such glacial hazards are rare, which hinders the capacity of glacial hazard practitioners to learn from these events. In this study, we reconstruct a historical GLOF from moraine-dammed Chongbaxia Tsho (89.745°E, 28.211°N) in the Eastern Himalaya, which is a unique case study because the outburst flood cascaded into two further lakes downstream. We employ a combination of i) multi-source and multi-temporal satellite imagery, ii) field investigation (including an unmanned aerial vehicle survey), iii) numerical dam breach and hydrodynamic modelling and, iv) qualitative and quantitative cryospheric and meteorological analysis, to investigate the evolution of the GLOF hazard, simulate moraine dam failure and GLOF propagation, and explore the role that long- and short-term climate trends played in providing the conditioning factors for the outburst. Chongbaxia Tsho expanded rapidly between 1987 until 2001 in response to glacier recession most likely caused by a regional warming trend of +0.37 °C per decade. Based on satellite image analysis we refine the outburst date to be 6 August 2001, instead of 6 August 2000, as previously reported, and attribute an ice avalanche into the glacial lake originating from the receding parent glacier as the most likely trigger for moraine dam failure. Through DEM differencing and lake level decrease, we estimate that a total water volume of 27.1 ± 1.6 × 106 m3 was released from the lake during the event, and using dam breach modelling we estimate that the peak discharge at the breach was >6600 m3 s−1. The GLOF flowed through downstream Chongbamang Tsho and Chongbayong Tsho, both of which served to attenuate the GLOF and reduce downstream losses; the latter stored an estimated 96% of the flood volume. Precipitation totals in the weeks preceding the GLOF exceeded the historical mean by up to 40%, and may have contributed to instability of the parent glacier, and generation of an ice avalanche with enough impact energy to cause lake water to overtop the moraine dam and initiate breach development. A future GLOF from Chongbaxia Tsho cannot be ruled out, but more field data, including detailed lake bathymetry, and information pertaining to the sedimentological and geotechnical characteristics of the moraine dam, are required for a more robust parameterization of a predictive GLOF model and quantification of the hazard posed by a future GLOF.

Sedimentary characteristics of landslide dam outburst flood deposits are distinct from those of rainfall flood deposits due to their very high hydraulic energy and sediment concentration. While they have been studied well in geological record, investigations of modern outburst flood deposits are relatively rare. Well-preserved flood deposits and landforms caused by the June 2000 landslide dam outburst flood in the Yigong River, eastern Tibet provide an opportunity to evaluate such sedimentological response to a high energy modern event. The sedimentary characteristics of boulder bars and channel evolution over c. 17 km downstream of the breach were investigated by field survey and remote sensing interpretation. The results show that the post-flood channel migrated downstream and increased in curvature. The bar shape is very little changed toward downstream, but the bar area shows a fluctuating decrease. Poorly sorted accumulations composed of massive granular gravels and boulders near the breach indicate sediment-rich bedload depositing processes occurred during rapid rising stage flows. In contrast, there are eddy bars and deposition of fine gravel with horizontally bedded laminations 17 km downstream of the dam, suggesting suspended sediment dominated deposition with a rapid loss of transport capacity. Comparison with previous hydrodynamic simulations indicates that the rates of the rising stage and the falling stage strongly affect the sedimentary characteristics of boulder bars and slackwater deposits.

Results from geomorphological, sedimentological and geochronological analyses, together with micropaleontological and mineralogical characteristics of lacustrine deposits in five locations within the Kuray intermountain depression, southeast Altai, mountains of south Siberia, support the thesis of repeated formations of ice-dammed lakes during MIS-2 and their draining by high energy floods. Our data suggest that the timing of one of the last cataclysmic draining events in the area can be estimated by an Optically-Stimulated Luminescence (OSL) age of 19.0 ± 1.1 ka for a sandy layer at the top of the diluvial (i.e. large flood) deposit, revealed in a sedimentary sequence of the 1570 m a.s.l. strandline – one of the lowest preserved strandlines in the western part of the basin. New OSL and radiocarbon ages, augmenting previously published dates, indicate that the last lake to occupy the Kuray depression occurred around 19–16 ka with a depth of at least 170 m in the central part of the basin and to a depth of no less than 220 m near the glacier dam. Lacustrine deposits are represented by two horizons of sandy clays separated by interlayers of mixed-size sands. The mineralogical data, supported by analysis of sedimentological and micropaleontological records, indicate accumulation of a lower lacustrine horizon in a deeper reservoir. Finding of Leucocythere sp.1, Leucocythere sp.2, and Leucocythere dorsotuberosa ostracod species in lacustrine deposits characterizes these reservoirs as periglacial freshwater cold and deep lakes. The presence of well-crystallized mica and chlorite in lacustrine silts and clays from the lower lacustrine horizon indicates cold, dry conditions at the time of their formation, as well as a predominance of physical weathering of rocks within the denudation area. After an abrupt dropping of the lake level around 16 ka, determined from OSL dating, the lake never recovered its former depth. The available radiocarbon ages for organic material in subaerial deposits within the study area and the new OSL ages suggest that the last ice-dammed lake in the Kuray basin was drained between ~16.7 and 9.9 ka. The presence of this lake might explain the absence of late Paleolithic surface finds within the basin that remained generally unsuitable for human habitation until its final drying. The last outburst flood passed along the Chuya and Katun river valleys, which had been already carved by older cataclysmic floods, but did not significantly affect the topography downstream of the Kuray-Chuya intermountain depressions. We numerically simulated the draining of a palaeolake in the Kuray basin with the water level 1650 m a.s.l. (maximal depth about 220 m near the dam) with different scenarios of breaching the ice dam. In contrast to a relatively gradual breach of the ice dam due to thermal erosion, an instantaneous dam break due to structural failure can cause an outburst flood with a peak discharge of around 2 × 106 m3 s−1. The high speeds of the water flow, 1.9–5.6 m s−1, with the maximum Froude numbers of 0.06–0.22, and peak Shields values of 0.03–0.25 indicate competence to mobilize gravel. Generally, the simulated flow remained subcritical, suggesting that bedforms developed under supercritical flows, such as antidunes, could not have developed, although the development of dunes cannot be precluded. Our data also contribute to the issue of correlating the low lake strandlines in the Kuray basin with the landforms associated with cataclysmic outburst floods.

Holocene and Upper Pleistocene sediments in the eastern Himalayan syntaxis represent an extraordinary record of glacial outburst megafloods in one of the most tectonically active landscapes on Earth. Glacial damming and outburst floods in the syntaxis may have focused erosion in the steep Tsangpo Gorge and inhibited river incision into the margin of the Tibetan Plateau. However, few flood slackwater deposits have been studied, and it is unknown which of the hundreds of known glacial impoundments in Tibet may have sourced the floods. Here we report n=1438 new detrital zircon U-Pb data from individual ancient megaflood and historical outburst flood slackwater deposits to examine the provenance and erosive potential of these events. Detrital zircon U-Pb geochronology of megaflood deposits show that megaflood provenance is more consistent with impoundment of the Yarlung River drainage at or west of the Namche Barwa massif, than impoundment of eastern drainages along the Yigong and Parlung Rivers. Compared to active bedload in the Siang River, and historical floods within the same drainage area, megaflood samples overall contain a disproportionately large amount of zircons eroded from the Tsangpo Gorge. We interpret this preferential derivation to reflect preferential erosion during large discharge events—supporting the hypothesis that Quaternary megafloods are a primary contributor to rapid exhumation of the eastern Himalayan syntaxis.

Flood events can induce considerable sediment transport which in turn influences flow dynamics. This study investigates the multiple effects of sediment transport in floods through modelling a series of hydraulic scenarios, including small-scale experimental cases and a full-scale glacial outburst flood. A non-uniform, layer-based morphodynamic model is presented which is composed of a combination of three modules: a hydrodynamic model governed by the two-dimensional shallow water equations involving sediment effects; a sediment transport model controlling the mass conservation of sediment; and a bed deformation model for updating the bed elevation. The model is solved by a second-order Godunov-type numerical scheme. Through the modelling of the selected sediment-laden flow events, the interactions of flow and sediment transport and geomorphic processes within flood events are elucidated. It is found that the inclusion of sediment transport increases peak flow discharge, water level and water depth in dam-break flows over a flat bed. For a partial dam breach, sediment material has a blockage effect on the flood dynamics. In comparison with the ‘sudden collapse’ of a dam, a gradual dam breach significantly delays the arrival time of peak flow, and the flow hydrograph is changed similarly. Considerable bed erosion and deposition occur within the rapid outburst flood, which scours the river channel severely. It is noted that the flood propagation is accelerated after the incorporation of sediment transport, and the water level in most areas of the channel is reduced.

The glaciolacustrine history of the Huron-Erie lowland is reinterpreted using a closed basin model (CBM), based on ground surveys, LiDAR imagery, and digital elevation mapping (DEM). According to the CBM, paleolakes Utica, St. Clair, and Rouge were formed about 13,813 cal yr BP in closed depressions as the level of glacial Lake Elkton dropped below that of bounding morainal swells. The CBM explains why no physical connections could be established previously between paleolakes Algonquin and St. Clair, or between paleolakes St. Clair and Rouge. The CBM obviates the need for a spillway at Port Huron during the time of early Lake Algonquin. The Port Huron spillway is reinterpreted as having first formed about 5,728 cal yr BP simply as a consequence of early Holocene glacial rebound, southward tilting of the Lake Huron basin, and rising water level during the Nipissing transgression. LiDAR and DEM maps suggest that spillway development in the Port Huron moraine caused catastrophic flooding during initial formation of the St. Clair River channel. Channeled scablands-like topography in the form of braided scour channels, streamlined erosional residuals, and boulder lag deposits downriver from Detroit suggest that the Detroit River may have formed by outburst flooding as the Detroit moraine was breached by the rising water level of paleolake St. Clair.

Understanding of complex flood-riverbed interaction processes in large-scale (field) outburst floods remains incomplete, not least due to a lack of well-constrained field data on hydraulics and sediment transport, but also because consensus on an appropriate model framework has yet to be agreed. This study presents a novel full 2D hydro-morphodynamic model containing both bedload and suspended load capability. Firstly, the model design is presented with an emphasis on its design to simulate rapidly-varied sediment-laden outburst floods and also the associated geomorphological impacts. Secondly, the model is applied to a large-scale (field) glacier outburst flood or ‘jokulhlaup’ at Solheimajokull, Iceland. For this real-world event, model scenarios with only water and with inclusion of sediment with different parameter setups were performed. Results indicated that grain size specifications affected resultant geomorphological changes, but that the sensitivity of the simulated riverbed changes to the empirical bedload transport formulae were insignificant. Notably, a positive feedback occurred whereby the jokulhlaup led to significant net erosion of the riverbed, producing an increase in flow conveyance capacity of the river channel. Furthermore, bulking effects of sediment entrainment raised the peak discharge progressively downstream, as well as the flood volume. Effects of geomorphological changes on flood water level and flow velocity were significant. Overall, despite the increased computational effort required with inclusion of sediment transport processes, this study shows that river morphological changes cannot be ignored for events with significant in-channel erosion and deposition, such as during outburst floods.

In April 2000 a large-scale rock avalanche dammed the Yigong Zangpo River, forming an extensive rockslide-dammed lake. The impoundment lasted for 62 days before a catastrophic breaching caused a massive outburst flood in the Yarlung Zangpo (Tibet) and the Dihang rivers (India) that travelled downstream to the main floodplain of the Brahmaputra in northeastern India. In response to discrepancies in the published literature on the event, we present a review and re-evaluation of the characteristics of the rock avalanche and associated landslide-dammed lake. We use digital topographical data (SRTM-3) and dynamic landslide modelling (DAN-W and DAN3D) to determine the salient characteristics of the damming landslide and to characterise its behaviour. Our analysis indicates that the volume of the damming rockslide was ca. 115 Mm3, including 91 Mm3 from the initial rockslope failure (bulked during disaggregation to 109 Mm3) and 6 Mm3 from entrainment during its 10.1 km travel down Zhamulong gully. The debris travelled with an average velocity of 15–18 m/s and resulted in a landslide dam on the Yigong River with a minimum height of about 55 m. Using LANDSAT-7 imagery (obtained before, during, and after impoundment) in conjunction with an SRTM-3 DEM, we reproduced the filling of the lake. We determine that the landslide dam formed an extensive reservoir with an impounded volume of 2.015 Gm3 and a maximum possible lake level of 2264 m asl (rounded to 2265 m asl). Our figures differ from those previously published but are believed to be well-constrained verifiable estimates of the volumes of the 2000 Yigong events. The outburst occurred after an attempt by army personnel to manually dig a spillway over the landslide debris and resulted in the entire volume of the lake draining in about 12 h. The outburst flood travelled over 500 km south into India, with a recorded rise in river level of 5.5 m at the Pasighat gauging station, 462 km downstream. In terms of historical outburst volumes from rockslide-dammed lakes, the volume of the 2000 Yigong event is only exceeded by that of the 1841 outburst flood from the Indus River rockslide-dammed lake, northern Pakistan.

article i nfo No glacial lake census exists for the Third Pole region, which includes the Pamir-Hindu Kush-Karakoram- Himalayas and the Tibetan Plateau. Therefore, comprehensive information is lacking about the distribution of and changes in glacial lakes caused by current global warming conditions. In this study, the first glacial lake in- ventories for the Third Pole were conducted for ~1990, 2000, and 2010 using Landsat TM/ETM+ data. Glacial lake spatial distributions, corresponding areas and temporal changes were examined. The significant results are as follows. (1) There were 4602, 4981, and 5701 glacial lakes (N0.003 km 2 ) covering areas of 553.9 ± 90, 581.2 ± 97, and 682.4 ± 110 km 2 in ~1990, 2000, and 2010, respectively; these lakes are primarily located in the Brahmaputra (39%),Indus (28%), and AmuDarya (10%) basins. (2) Small lakes (b0.2 km 2 ) are more sensitive to climate changes. (3) Lakes closer to glaciers and at higher altitudes, particularly thoseconnected to glacier ter- mini, have undergone larger area changes. (4) Glacier-fed lakes are dominant in both quantity and area (N70%) and exhibit faster expansion trends overall compared to non-glacier-fed lakes. We conclude that glacier meltwa- ter may play a dominant role in the areal expansion of most glacial lakes in the Third Pole. In addition, the pat- terns of the glacier-fed lakes correspond well with warming temperature trends and negative glacier mass balance patterns. This paper presents an important database of glacial lakes and provides a basis for long-term monitoring and evaluation of outburst flood disasters primarily caused by glacial lakes in the Third Pole.

Pleistocene floods, from ice-dammed lakes in the Russian Altai Mountains, are among the largest freshwater megafloods in Earth history. Improved dating techniques have led to new dates constraining the timing of flood erosional and depositional events. In addition to a variety of interpretations and explanations of event histories, a proliferation of dates may increase confusion with respect to the characteristics and chronology of the flood events. In a succinct review, the main ice-dammed lakes features, the outburst floods and their repeated formation are presented. Among other indicators, interbedded units of lacustrine deposits and outburst flood sediments in Inya River valley demonstrate conclusively that there was a series of outburst floods with significant evidence for at least three high magnitude flood events. Prior study focussed on the interpretation of: lake sediments, shorelines, giant bars, gravel dunes; dated by radiocarbon, luminescence and cosmogenic nuclide assay. The literature presents sometimes contradictory findings with respect to flood chronology. These different interpretations, evidence and arguments, are reviewed for plausibility and consistency. This consideration includes methodical and technical dating problems: limited bleaching for luminescence techniques; low purity for cosmogenic nuclide dating; contamination of samples dated by radiocarbon assay. At the current stage of knowledge, the main period of repeated high magnitude outburst floods occurred between 28 ka and 15 ka BP. Indicators of earlier events are less distinct. Possible younger smaller floods can be identified after 15 ka BP, but it is difficult to relate these to catastrophic lake-draining. Rather, they may relate to non-catastrophic draining of residual small lakes within the larger basins. The termination of the “lake period” took place before 9.9 ± 0.3 ka cal BP in the Kuray Basin, and in the Chuya Basin before 8.2 ± 0.2 ka cal BP. Indicators are lacking for any drainage events after 7.7 ± 0.6 ka as indicated by OSL ages of aeolian deposits at low elevations in the Katun valley. Relating the outburst floods to the temporal and spatial dynamics of the impounding glacier is challenging as, with regard to Pleistocene glaciation, several conceptual models have been established supported by field evidence that, in part, can be contradictory. During the period of the high magnitude outburst floods the extension/thickness of the impounding valley glaciers were sufficient to form deep glacial lakes around the time of the last glacial maximum (LGM); yet the timing of the LGM remains controversial. Transferring glacial chronologies from mountain areas adjacent to the Russian Altai Mountains to the environment of the megafloods does not provide a clearer picture as there is evidence for maximum stages at MIS 4 and MIS 3 - in addition to recent postulations of a LGM in MIS 2 for the Russian Altai Mountains. The Russian Altai Mountains are located in a climatic transitional zone across which different LGM stages could be developed due to different moisture availability and temperature chronologies. The effects of spatially-variable rain-shadows influencing the supply of moisture from the regional prevailing winds and local moisture sources from expanded ice-dammed lakes complicates the picture and require further investigations.

We present geomorphological evidence of large, previously undocumented, early Holocene ice-dammed lakes in the Scandinavian Mountains of northwestern Sweden. The lakes extents indicate that the last remnants of the Scandinavian Ice Sheet were located east of the mountain range. Some early pioneering works have presented similar reconstructions, whereas more recently published reconstructions place the last ice remnants in the high mountains of Sarek. Using high-resolution airborne LiDAR data we have mapped a large number of hitherto undocumented shorelines in some of the main valleys within the northern Scandinavian mountain range. Our results indicate that a larger system of ice-dammed lakes existed in this region than previously thought. The lakes were dammed between the main water divide to the west and the retreating ice sheet margin to the east. The shorelines dip towards the northwest with gradients ranging from 0.5 to 0.4 m/km, from the oldest to the youngest. Further, we have compiled Lateglacial and Holocene shoreline data along the Norwegian coast and from within the Baltic Sea basin and reconstructed the isostatic uplift along a 1400 km long northwest-southeast transect from the Norwegian Sea to Lake Ladoga. By comparing the measured ice-dammed lake shoreline gradients to the dated marine shorelines, we infer that the lakes may have existed for several centuries following 10.2 cal ka BP. We also describe large deposits and extensive erosive features, which demonstrate that a catastrophic glacial lake outburst flood (GLOF) took place eastward along the Pite River Valley. Based on cross-cutting relations to raised shorelines developed in the early Holocene Ancylus Lake (Baltic Sea basin) we conclude that the flood and thus the final phase of deglaciation took place within the time interval 10.3–9.9 cal ka BP. publishedVersion

This paper discusses Howard et al. (2012) who reconstruct the peak discharge of a glacial outburst flood, or ‘jökulhlaup’, for part of the Jökulsá á Fjöllum in north-central Iceland. They propose that this flood was the largest on Earth.Weconsider that the magnitude of the jökulhlaup proposed by Howard et al. (2012) warrants much more robust field evidence and demands more carefully parameterised hydraulic modelling. For these reasons we firstly (i) present their study in the context of previous research (ii) highlight issues with attributing landforms and sediments to jökulhlaups, and (iii) consider uncertainty regarding the timing and magnitude of jökulhlaups along the Jökulsá á Fjöllum. We argue herein that whilst a range of landforms and sediments that are attributable to jökulhlaups can be observed along the Jökulsá á Fjöllum, these are not necessarily diagnostic of jökulhlaups. Secondly, we critically discuss (iv) the major underlying assumptions of their study, and (v) their calculations and subsequent interpretations. These assessments lead us to consider that the proposal by Howard et al. (2012) of the largest flood on Earth is highly unrealistic, especially when due consideration is given to a possible source area and a trigger mechanism.

Previous studies show that the area of a moraine-dammed lake can provide a good indicator of the significance of its outburst. For a glacier-dammed lake however, because its area and depth fluctuates with the melting of its ice dam, it is difficult to predict the outburst of the glacier-dammed lake by using its area alone. A characteristic of the surface of Lake Merzbacher is a large amount of floating ice therefore, a method is proposed in this article to extract the area of floating ice on the lake and the area of ice free water in the lake by using Environment and Disaster Monitoring Small Satellite images respectively. Furthermore, based on the area of floating ice extracted through the image information of Lake Merzbacher in 2009 and 2010, we determined the relationship between the ice area and the outburst of the lake, then formulated the Index for hazard of Glacier Lake Outburst Flood (IGLOF) of Lake Merzbacher, which cannot only predict the flood outburst, but also determine the specific outburst period after the lake drainage had occurred. This can be shown in a recalculation of the lake drainages in the years 2009 and 2010. Research results indicate that when IGLOF is less than 0.5 and the lake area is larger than 3 km(2), the outburst process is in early-warning period and GLOF will occur in the next 5-8 days. Also, the successful outburst prediction of Lake Merzbacher in 2011 showed that the index described in this paper provides a quick methodology for forecasting and warning against Lake Merzbacher outburst floods. However as our research was based on a short observation period (2009-2011) and also cannot be supplemented by other images, it will still be needed to be checked and validated by continuous observation and improvement in future. (C) 2012 Elsevier B.V. All rights reserved.

The Towada Caldera Volcano is located between the prefectures of Aomori and Akita — Northeast Honshu Island, Japan. The caldera, today filled by a lake, has produced 15,000 years ago a complex eruption emplacing an ignimbrite topped by the lake outburst flood deposit, through which the present Oirase River erodes. This flood deposit has shaped the geomorphologic feature named Sanbongi Fan, on which Towada City extends. This flood event hypothesis is mainly based on sedimentological and geomorphological evidences of floods mainly from outcrops retrieved from the Sanbongi Fan area. Because of the lack of extended outcrops – typical of the Japanese environment – the present paper has therefore put the theory to the test using GPR (Ground Penetrating Radar) radargram extending along a 640 m length. The GPR used for the survey was a Pulse-ekko-Pro with 50 MHz antennas, and the software Reflex was used to process the data. The radargrams have displayed a sole unit, which the GPR could not penetrate. It can be interpreted as the ignimbrite. On top of this deposit a series of subhorizontal layers, with the alternation between a backset and a foreset extends between 5 m and 3 m depth. Above 3 m, the units are regular and subhorizontal. The deposit is also characterized by the extensive presence of boulders, which are located along three bands: (1) on top of the ignimbrite; (2) in the units deposited by the outburst flood, between 3 and 5 m depths; (3) and in the units close to the surface, although part of these punctual elements are most certainly anthropogenic. Compared with the outcrops, the present research confirms that the material located above the ignimbrite material have been deposited by the outburst flood, creating large-sheet patterns, which have transported boulders. These sheets display backset and foreset patterns, depending on the position of the deposit, indicating flow pulsation or surges under a ‘high-energy-flow’ condition.

Stratigraphy and age control of late Pleistocene loess, associated glacial outburst flood deposits and flood-cut unconformities in the Channeled Scabland, Washington State, United States, indicate a significant Cordilleran ice sheet advance during marine Oxygen Isotope Stage 4. Glacial outburst flood deposits from stage 2 (classic Bretz flood deposits, ca 21 to 13 ka) and related features in the Channeled Scabland overlie a widespread layer of loess that contains buried soils and the Mount St. Helens set C tephra (ca 46 ka). This loess in turn overlies deposits of the penultimate episode of giant outburst floods and an unconformity cut by those floods. Regional trends in the thickness, texture, and overall composition of the older loess are strikingly similar to those from the youngest loess, known to be derived from stage 2 flood deposits. We conclude that the older loess also is derived from fine-grained flood deposits. Luminescence ages, tephrochronology, and soil development rates indicate that the bulk of deposition of the older loess occurred during stage 3, following glacial outburst flooding marked by a regional flood-cut unconformity. The apparent cyclical pattern of cold-climate buried soils, flood deposits, and thick loess accumulations demonstrate that sediment supply renewed by flood episodes is a major control on accumulation of loess on glacial timescales.

During the late Saalian Drenthe glaciation ice-damming of the Upper Weser Valley led to the formation of glacial Lake Weser. The lake drained catastrophically into the Munsterland Embayment as the western ice dam failed, releasing up to 110 km3 of water with a calculated peak discharge of 2.5 × 105 m3/s to 1.3 × 106 m3/s. Geographic information systems (GIS) and high-resolution digital elevation models (DEM) were used to map streamlined landforms and channel systems in front of lake overspills. Geological maps, 2450 boreholes and the DEM were integrated into the 3D modeling program GOCAD to reconstruct the distribution of flood-related deposits, palaeotopographic surfaces and the internal facies architecture of streamlined hills. The drainage pathways are characterized by the occurrence of deep plunge pools, channels, streamlined hills and 4 km long and 12 m deep V-shaped megaflutes. Plunge pools are deeply incised into Mesozoic basement rocks and occur in front of three major overspill channels. The plunge pools are up to 780 m long, 400 m wide and 35 m deep. Approximately 1–10.5 km downslope of the overspill channels fan shaped arrays of streamlined hills are developed, each covering an area of 60–130 km2, indicating rapid flow expansion. The hills commonly have quadrilateral to elongated shapes and formed under submerged to partly submerged flow conditions, when the outburst flood entered a shallow lake in the Munsterland Embayment. Hills are up to 4300 m long, 1200 m wide, 11 m high and have characteristic average aspect ratios of 1:3.3. They are separated by shallow, anabranching channels in the outer zones and up to 30 m deep channels in the central zones. Hills partly display V-shaped chevron-like bedforms that have apices facing upslope, are 1.6–2.5 km long, 3–10 m high, 0.8–1.2 m from limb to limb, with limb separation angels of 20–35°. These bedforms are interpreted as mixed erosional depositional features. It is hypothesized that the post-Saalian landscape evolution of the Munsterland Embayment has considerably been influenced by catastrophic floods of glacial Lake Weser, creating large and deep valleys, which subsequently became the new site of river systems. The outburst floods probably followed the east-west-trending Saalian Rhine-Meuse river system eventually flowing into the North Sea, the Strait of Dover and the Bay of Biscay. It is speculated that the Hondsrug ice stream may have been enhanced or even triggered by the formation and outburst of glacial lakes in the study area.

Summary Hydraulic behavior and unsteady sediment transport within dam break outburst floods are inherently inter-dependent, yet poorly constrained. This study experimentally determined how the mobility and clast size of channel bed sediments affected the peak flow depth and flow velocities, wave propagation and attenuation, sediment entrainment and duration of transport, flood power, energy transfer and geomorphological impact of outburst floods. Due to our choice of clast size and of a unimodal size distribution the predominant sediment transport mechanism was bedload. We found that frontal wave propagation was slower and attenuated faster for flows over a mobile bed versus those over a fixed bed. Peak flow depth was greater and was attained earlier for flows over a mobile bed and with increased clast size. Geomorphological power declined logarithmically with time after an initial period where gravitational forces exceeded frictional forces. Both peak and time-to-peak geomorphological power varied inversely with clast size. Peak deposition rates were greatest with coarser bed sediment but time to peak deposition rate was inversely proportional to clast size. Volume changes down channel were described by an exponential curve, which we suggest indicates that sediment transport proceeded in waves via both translation and dispersion mechanisms. Peak suspended load occurred coincident with peak flow velocity, peak bedload occurred coincident with peak flow depth, and peak sediment discharge lagged behind peak water discharge. These descriptions of kinematic transformation, multi-phase flow and of geomorphological work due to outburst floods help to provide data for numerical modelling, for mitigating outburst flood hazards and for informing on the immediate and longer-term geological legacy of outburst floods.

Glacial outburst floods have been observed or inferred in most glaciated regions of the world. Outburst flooding events along past ice-sheet margins have been discussed as possible causes for changes in ocean circulation while subglacial lake drainage under ice sheets may influence ice-sheet stability. The recognition of these geologically instantaneous events in the marine sediment or rock record is critical in order to characterize the magnitude and frequency of past outburst events as well as to predict the likelihood of future occurrences. In this paper, we characterize outburst lithofacies and infer depositional processes from two outburst floods within Disenchantment Bay, Alaska, and compare these deposits to other recognized outburst deposits within marine depositional settings. Here we show that the outburst deposits are lithologically similar and are composed of stratified mud, sand, and diamicton. We interpret these facies to have been formed by sediment gravity flows and hypopycnal plume sedimentation. These results show that the volume of sediment, lithofacies, and inferred depositional processes cannot be easily distinguished from normal sediment accumulation within the proximal temperate glacimarine depositional environment. The paucity of fluvial subglacial discharge in polar glacimarine environments suggests the potential for recognition of such deposits in the Antarctic.

^3He concentrations were measured in zircon, apatite and pyroxene separates from rhyolite surfaces along the Snake River near Twin Falls, Idaho. These fluted and potholed surfaces were presumably scoured by the Bonneville outburst flood ~17.5 ka. Several of the samples contain inherited ^3He from prior exposure, reflecting the complex incision history of the Snake River canyon and suggesting earlier flood events. Each individual mineral from the remaining surfaces yielded ^3He concentrations that are within error of each other. ^3He produced by neutron capture on ^6Li was established from shielded samples and a simple neutron production model. The remaining ^3He is spallogenic and was used to calibrate the production rate of cosmogenic ^3He in zircon, apatite and pyroxene by assuming minimal erosion since the Bonneville flood. In all three phases the resulting production rates are ~7–12% lower than previous ^3He production rate estimates obtained by reference to various production rates for ^(10)Be in quartz. This disagreement is partially reconciled here by recalculating previous results with a ^(10)Be production rate of 4.51 at g^(−1) a^(−1) in quartz. Adopting this revised rate brings three independent ^3He production rate estimates for zircon and apatite into agreement within 5%, with grand means of 103 ± 3 at g^(−1) a^(−1) in zircon and 133 ± 6 at g^(−1) a^(−1) in apatite. The major source of uncertainty in cosmogenic ^3He dating of these phases is now correction for ^3He produced by neutron capture on ^6Li. Calculations are presented for assessing the amount of uncertainty introduced by this correction as a function of Li concentration, cosmic-ray exposure age, and He closure age.

Catastrophic partial drainage of Pangong Tso, one of the largest lakes in Tibet, is supported by the geomorphology of the Tangtse Valley, Ladakh, northern India and cosmogenic 10Be nuclide ages of roche moutonnees, strath terraces, and a flood deposit downstream from the former spillway. The former spillway for Pangong Tso is ~ 20-m-high and likely allowed ~ 18 km3 of water to drain catastrophically down the Tangtse Valley over a period of about 2 days sometime during the latest Pleistocene to early Holocene. The largest flood deposit, composed of imbricated granitic boulders up to 4.5 m in length, is present ~ 33 km downvalley of the spillway. These boulders have a cosmogenic 10Be exposure age of 11.1 ± 1.0 ka, the age of the outburst flood. The minimum calculated discharge was ~ 110,000 m3 s− 1. One set of strath terraces, upvalley of the flood deposit along the flood's drainage path, shows that the rate of fluvial incision 0.3 ± 0.1 mm y− 1 during 122–10.5 ka increased to 1.5 ± 0.5 mm y− 1 during 10.5 ka to the present. The temporal overlap of this increase in the rate of fluvial incision with the main flood deposit suggests that the flood was important in defining the incision along the Tangste valley. A second set of strath terraces shows little change in incision, from ~ 0.6–0.9 to ~ 0.9–1.4 mm y− 1, sometime between 18 and 27 ka. Roche moutonnees, upvalley from strath terraces, yield a cosmogenic 10Be age of 35.8 ± 3.0 ka, defining the time when glaciers last occupied the Tangtse valley. However, the lack of glacial sediment along the Tangste valley suggests that the flood eroded glacial depositional landforms and sediments resulting in high sediment loads in the floodwater, which in turn increased fluvial incision to form strath terraces. Much of the eroded glacial sediment was subsequently redeposited as the main flood deposit. The catastrophic drainage of Pangong Tso may be the result of breaching of the Pangong–Tangste spillway during very high lake levels in a period of intensified monsoon (10.7–9.6 ka) and/or possibly the consequence of seismic activity along the Karakoram Fault that is associated with the initial formation of Pangong Tso.

This study reports on geomorphic and sedimentary evidence for a gigantic breakout flood from Towada caldera, Honshu Island, northeast Japan, as observed along the Oirase River downstream from the outlet of the caldera lake. A number of features of the Oirase River are inconsistent with its present discharge, including the occurrence of 1) hanging valleys and a horseshoe-shaped waterfall in the upstream gorge area, 2) boulder bars and scattered flood boulders, 3) a dry valley at a high elevation in the gorge, and 4) a large fan (the Sanbongi fan) in the downstream area of the river. The Sanbongi fan is composed of thick, lithic-rich hyperconcentrated flow deposits that include pumice clasts derived from the 15 ka Towada–Hachinohe ignimbrite and well-rounded meter-sized (and outsized) boulders derived from bedrock of welded ignimbrite. The deposits are entirely aggradational, with no major channels, indicating the absence of a major hiatus during sedimentation. The depositional facies also indicate that a single sheet-like flood event deposited the sediment within the Sanbongi fan area. Based on the age of the Sanbongi fan, the flood occurred between 15 and 12 ka, after eruption of the Towada–Hachinohe ignimbrite. The most probable water source for the flood is Towada caldera lake, as suggested by landforms and sediments along the lake outlet. A paleohydrological analysis indicates that at least 6 km 3 of water was released from the caldera at a peak discharge of > 2 × 10 4 to 3 × 10 5 m 3 s − 1 during the breakout flood. Although the Sanbongi fan was previously considered to be a “normal” alluvial fan that formed during a stage of low-stand sea level, the present results show that the formation of the fan was closely related to a catastrophic flood from Towada caldera rather than perennial fluvial activity, climate change, or a change in relative sea level.

Landslides may obstruct river flow and result in landslide dams; they occur in many regions of the world. The formation and disappearance of natural lakes involve a complex earth–surface process. According to the lessons learned from many historical cases, landslide dams usually break down rapidly soon after the formation of the lake. Regarding hazard mitigation, prompt evaluation of the stability of the landslide dam is crucial. Based on a Japanese dataset, this study utilized the logistic regression method and the jack-knife technique to identify the important geomorphic variables, including peak flow (or catchment area), dam height, width and length in sequence, affecting the stability of landslide dams. The resulting high overall prediction power demonstrates the robustness of the proposed logistic regression models. Accordingly, the failure probability of a landslide dam can also be evaluated based on this approach. Ten landslide dams (formed after the 1999 Chi-Chi Earthquake, the 2008 Wenchuan Earthquake and 2009 Typhoon Morakot) with complete dam geometry records were adopted as examples of evaluating the failure probability. The stable Tsao-Ling landslide dam, which was induced by the Chi-Chi earthquake, has a failure probability of 27.68% using a model incorporating the catchment area and dam geometry. On the contrary, the Tangjiashan landslide dam, which was artificially breached soon after its formation during the Wenchuan earthquake, has a failure probability as high as 99.54%. Typhoon Morakot induced the Siaolin landslide dam, which was breached within one hour after its formation and has a failure probability of 71.09%. Notably, the failure probability of the earthquake induced cases is reduced if the catchment area in the prediction model is replaced by the peak flow of the dammed stream for these cases. In contrast, the predicted failure probability of the heavy rainfall-induced case increases if the high flow rate of the dammed stream is incorporated into the prediction model. Consequently, it is suggested that the prediction model using the peak flow as causative factor should be used to evaluate the stability of a landslide dam if the peak flow is available. Together with an estimation of the impact of an outburst flood from a landslide-dammed lake, the failure probability of the landslide dam predicted by the proposed logistic regression model could be useful for evaluating the related risk.

Fed by a persistent conduit system, three outburst floods released pressurized water from subglacial conduits midway through the 1993–95 surge of Bering Glacier, Alaska. The foreland substrate was mobilized and streamlined by overriding ice, but the effects of hydraulic erosion and deposition were far more effective in altering overridden terrain. Two sub-kilometer basins were scoured decameters into foreland terrain immediately up-glacier from outburst flood sites. These emerged fully formed from beneath the retreating glacier. Flood deposits in scoured subglacial channels consist of meter-size lag boulders beneath meters of well-sorted, graded gravel couplets. Eskers consisting of flood deposits formed within subglacial conduits during outburst flood events were exposed during retreat. Knowledge of pre-surge topography combined with foreland stratigraphy provides a basis for the recognition of subglacial, paleo-flood landforms and deposits.

summary Lahars, and other sudden onset floods, are highly dynamic with temporally and spatially evolving sediment loads. Numerical models of sudden onset flood behaviour, particularly those applied to real world events, have tended to neglect sediment transport processes. This is a serious though perhaps understandable shortcoming given a lack of field measurements on flow-bed interactions, the extra computational complexity required, and the problem of iteratively updating ‘plastic’ channel boundaries. This paper applies a fluid dynamics model with suspended sediment and bedload transport to calculate flow conditions and rapid landscape change due to an initially dilute lahar from Mt. Ruapehu, New Zealand. We take advantage of an unprecedented dataset of net topographic changes determined from pre- and post-lahar airborne LiDAR surveys and time-series of hydrological measurements to run and validate the model. A sensitivity analysis of key controls on water surface elevation and final sediment thickness illustrates how the flow was supply-limited in the upper reaches. It was capable of transporting a greater volume and calibre of material than that available. Modelled patterns of net erosion and deposition can replicate those observed (r 2 = 0.79) and are dominantly controlled by channel gradient and topographic

In this study postglacial continuous high-resolution data from five marine sediment cores from Hudson Bay and Hudson Strait are presented. Up to now studies in this area focused on the distinction between large scale climatic variations like the glacial–postglacial transition or most recent processes. However, as demonstrated in this study, using sedimentological, geochemical and geophysical parameters, postglacial sediments provide provenance signals distinguishing between Proterozoic Dubawnt Group sediments, Paleozoic limestone sources underlying parts of Hudson Bay and Strait and Precambrian granitoids from the Canadian Shield. For example it is shown that redistribution of Dubawnt Group sediments is of minor importance after the outburst flood of Lake Agassiz. Paleoproductivity indicators also give insights into paleotemperature trends. These are in phase with terrestrial reconstructions from that area. Colder conditions in Hudson Strait until ca. 4500 cal BP followed by a signal of warmth until ca. 4000 cal BP are suggested. Subsequently, a cool interval lasting until approximately 2000 cal BP occurred. Thereafter, a gradual decrease in temperature was observed reaching lowest values during the Little Ice Age period and increasing again afterwards.

During glacial outburst floods (jokulhlaups) large ice-blocks are often released from the glacier margin. Ice-blocks ground at points of flow deceleration, generating distinctive patterns of scour and deposition scaled to the obstacle size. The large-scale ice-block obstacle marks produced within the proglacial zone are diagnostic of jokulhlaups. We use ground-penetrating radar (GPR) to investigate the sedimentary architecture of ice-block obstacle marks generated by the November 1996 jokulhlaup at Skeiðararjokull, Iceland. The GPR data reveal a varied sedimentary architecture, interpreted to suggest that ice-block obstacle mark sedimentology is controlled by flow conditions, sediment supply, and whether ice blocks are deposited in isolation or as a cluster. Following ice-block grounding, deflection of flow down and away from the obstacle develops horseshoe vortices that maintain a non-depositional hollow around the obstacle, whilst continued outwash deposition away from the obstacle raises the level of the surrounding sandur. Where ice-blocks ground early during the jokulhlaup, hollows are infilled with foreset beds at the up-flow end and backset beds (antidune) distally. However, where ice-blocks ground on the waning stage when sediment flux is reduced, antidunes are not preserved and the hollow is infilled with low-angle foreset beds. In both cases the lee-side shadow ridge is composed of pseudo-anticlinal beds proximal to the obstacle (where shed vortices are strong) and plane beds distally (where flows converge and the shed vortices weaken). Where flow parallel ice-block clusters develop, the grounding of smaller ice-blocks in the lee of the initial obstacle results in only limited shadow ridge aggradation. These data provide improved insight into the controls on ice-block obstacle mark sedimentary architecture and allow identification of high-magnitude jokulhlaups in the sedimentary record.

A series of piston cores from Flathead Lake, Montana, USA and a new radiocarbon date from the sedimentary record provide the basis for describing sedimentary processes related to deglaciation of the Flathead Lobe of the Cordilleran Ice Sheet and for interpreting the retreat history of the lobe. The oldest part of Flathead Lake sediment core records is Late Pleistocene in age and consists of cm-scale rhythmites of silt and clay, interpreted here as annual varves. Each varve contains a light-colored coarser-grained portion, inferred to represent deposition during peak annual runoff, and a darker-colored finer-grained portion interpreted to represent sediment accumulation during seasonal low-flow conditions. The coarser-grained portions, especially in the stratigraphically older sections of each core, contain sedimentary structures that reflect traction transport. Based on these sedimentary structures, their facies characteristics, and their spatial distribution within the lake, we interpret the thicker, light-colored portion of each varve to be deposited by hyperpycnal flows caused by seasonal melt events rather than more classic turbidity currents. Immediately overlying Late Pleistocene rhythmites in all Flathead Lake cores is a unique, significantly coarser-grained dm-scale silt bed with a median grain size up to 50 µm. This silt bed has a sharp, locally erosional base and fines upward but does not contain any other sedimentary structures. In contrast to the rhythmites, we interpret this silt bed to represent a single, short-lived catastrophic sedimentation event generated by a large glacial outburst flood. Overlying this distinct bed are several other cm-scale beds of mainly silt that exhibit a basal upward-coarsening followed by an upward-fining median grain size profile. We interpret these beds and their grain size trends as reflecting the rising and falling hydrograph limbs of outburst floods derived from more distal sources located in the upstream parts of the upper Flathead watershed. The sediment record from Flathead Lake, together with results from geologic and geomorphologic 1:24,000 scale mapping around the lake margins, provide a series of constraints regarding the paleogeographic evolution of the area during deglaciation. Overall upward-thinning and upward-fining of the varved portion of the sediment core records reflects northward retreat of the southern Flathead Lobe ice margin starting at latest 14,475 ± 150 cal yr BP, the depositional age of the oldest varve in any of our cores. The depositional age of silt beds overlying the varved records is constrained as between 14,150 ± 150 cal yr BP and 13,180 ± 120 cal yr BP. Within the available chronostratigraphic constraints, the outburst floods interpreted to have delivered this silt to the Flathead Lake basin also downcut a bedrock nick point below the Flathead Lake outlet, oriented a series of large boulders downstream, and deposited a series of large flood bars on the lower Flathead River floodplain.