Bedrock Fragments Research Articles

Coarse-Clast Storm Deposit and Solitary Boulders on the Island of Mana (NP Kornati, Central Adriatic, Croatia)

There is growing evidence that many large coastal boulder deposits found on the exposed rocky ocean shores were deposited by extreme storm waves rather than by catastrophic tsunamis, as previously thought. In addition, before the first discovery in the northern Adriatic a few years ago, such deposits were not expected in relatively shallow semi-enclosed inland basins. Here we report on a large coastal coarse-clast deposit on the central Adriatic island of Mana, which also contains numerous large storm boulders that weigh up to several tons. Large solitary boulders are also located outside of the deposit, closer to the partly submerged sea cliff and the wave impact. The erosion of the cliff top and displacement of the carbonate bedrock fragments began when the extreme waves inundated the lowermost part of the cliff edge, probably during the late Holocene sea-level rise. The UAS photogrammetry-based fragmentation analysis of the storm deposit and the calculated fractal dimension value indicate that the material was fragmented by multiple high-energy events. A comparison of the available photographs indicates that displacements of the most exposed solitary boulders probably occurred during Vaia, the last extreme storm that hit the Adriatic on the 29th of October 2018. However, the modeled maximum wave height south of Mana during the peak of the storm would be insufficient to move these boulders. Yet local geomorphology probably further influenced the increase in wave height that, in combination with specific geological features, caused displacements of the boulders. There is a shorter fetch affecting Mana Island with respect to the northern Adriatic boulder field in southern Istria. Thus, such an active local erosion of the generally stable eastern Adriatic karstic coast depends on the extreme storms that have a weaker impact in the central than in the northern Adriatic.

Abrasion regimes in fluvial bedrock incision

AbstractRiver incision into bedrock drives landscape evolution and couples surface changes to climate and tectonics in uplands. Mechanistic bedrock erosion modeling has focused on plucking—the hydraulic removal of large loosened rock fragments—and on abrasion—the slower fracturing-driven removal of rock due to impacts of transported sediment—which produces sand- or silt-sized fragments at the mineral grain scale (i.e., wear). An abrasion subregime (macro-abrasion) has been hypothesized to exist under high impact energies typical of cobble or boulder transport in mountain rivers, in which larger bedrock fragments can be generated. We conducted dry impact abrasion experiments across a wide range of impact energies and found that gravel-sized fragments were generated when the impact energy divided by squared impactor diameter exceeded 1 kJ/m2. However, the total abraded volume followed the same kinetic-energy scaling regardless of fragment size, holding over 13 orders of magnitude in impact energy and supporting a general abrasion law. Application to natural bedrock rivers shows that many of them likely can generate large fragments, especially in steep mountain streams and during large floods, transporting boulders in excess of 0.6 m diameter. In this regime, even single impacts can cause changes in riverbed topography that may drive morphodynamic feedbacks.

Пространственная неоднородность свойств почв в зоне распространения островной мерзлоты (Приполярный Урал)

Пространственная неоднородность свойств почв в зоне распространения островной мерзлоты (Приполярный Урал)

Earthquake environmental effects produced by the Mw 6.1, 20th May 2016 Petermann earthquake, Australia

Earthquake Environmental Effects (EEEs) identified in the source region of the 20th May 2016 intraplate moment magnitude (Mw) 6.1 Petermann earthquake in Central Australia are described and classified using the Environmental Seismic Intensity (ESI-07) scale. EEEs include surface rupture, ground fissures and cracks, vegetation damage, rockfalls, and displaced (jumped) bedrock fragments. The maximum ESI intensity derived from EEEs is X, consistent with previous observations from some moderate Mw crustal earthquakes. Maximum ESI isoseismals correlate with the location of the surface rupture rather than epicentre area due to the dipping geometry of the reverse source fault. ESI isoseismals encompass a larger area of the hanging-wall than the footwall, indicating stronger ground motions on the hanging-wall due to increased proximity to the rupture source and ground motion amplification effects. The maximum areal extent of secondary (seismic shaking-induced) EEEs (300 km2) is significantly smaller than expected using the published ESI-07 scale (approx. 5000 km2). This relates to the low topographic relief and relatively homogeneous bedrock geology of the study region, which (i) reduced the potential for site response amplification of strong ground motions, and (ii) reduced the susceptibility of the landscape to EEE such as landsliding and liquefaction. Erosional degradation of the observed EEE features and decreasing confidence with which they can be uniquely attributed to a seismic origin with increasing time since the earthquake highlight challenges in using many of the natural features observed herein to characterise the locations and attributes of paleo-earthquakes.

Micrometer silicon isotope diagnostics of soils by UV femtosecond laser ablation

This study presents the first Si isotope data of the principle Si pools in soils determined by a UV femtosecond laser ablation system coupled to a multicollector inductively coupled plasma mass spectrometer (MC-ICP-MS). This method reveals accurate and precise Si isotope data on bulk materials, and at high spatial resolution, on the mineral scale. The following Si pools have been investigated: a) the Si source to soils on all major silicate minerals on thin sections from bedrock fragments in the soil profiles; b) bulk soils (particle size < 2 mm) after fusion to glass beads with an iridium-strip heater or pressed into powder pellets; c) separated clay fractions as pressed powder pellets and e) separated phytoliths as pressed powder pellets. Multiple analyses of three rock standards, BHVO-2, AGV-1 and RGM-1 as fused glass beads and as pressed powder pellets, reveal δ 30Si values within the expected range of igneous rocks. The MPI-DING reference glass KL2-G exhibits the same Si isotope composition after remelting by an iridium-strip heater showing that this technique does not alter the isotope composition of the glass. We used this approach to investigated two immature Cambisols developed on sandstone and paragneiss in the Black Forest (Germany), respectively. Bulk soils show a largely uniform Si isotope signature for different horizons and locations, which is close to those of primary quartz and feldspar with δ 30Si values around − 0.4‰. Soil clay formation is associated with limited Si mobility, which preserves initial Si isotope signatures of parental minerals. An exception is the organic horizon of the paragneiss catchment where intense weathering leads to a high mobility of Si and significant negative isotope signatures as low as to − 1.00‰ in bulk soils. Biogenic opal in the form of phytoliths, exhibits negative Si isotope signatures of about − 0.4‰. These results demonstrate that UV femtosecond laser ablation MC-ICP-MS provides a tool to characterize the Si isotope signature of the principle Si pools left behind after weathering and Si transport have altered soils. These results can now serve as a fingerprint of the residual solids that can be used to explain the isotope composition of dissolved Si in soil solutions and river water, which is mostly enriched in the heavy isotopes.

High cadmium concentrations in Jurassic limestone as the cause for elevated cadmium levels in deriving soils: a case study in Lower Burgundy, France

Cadmium (Cd) is a highly toxic element and its presence in the environment needs to be closely monitored. Recent systematic surveys in French soils have revealed the existence of areas in eastern and central France, which show systematically high cadmium concentrations. It has been suggested that at least part of these anomalous levels are of natural origin. For the Lower Burgundy area in particular, a direct heritage from the Jurassic limestone bedrock is highly suspected. This potential relationship has been studied in several localities around Avallon and this study reports new evidence for a direct link between anomalously elevated cadmium contents of Bajocian and Oxfordian limestone and high cadmium concentrations in deriving soils. Soils in this area show cadmium concentrations generally above the average national population values, with contents frequently higher than the ‘upper whisker’ value of 0.8 μg g−1 determined by statistical evaluation. In parallel, limestone rocks studied in the same area exhibit cadmium concentrations frequently exceeding the mean value of 0.030–0.065 μg g−1 previously given for similar rocks by one order of magnitude, with a maximum of 2.6 μg g−1. Mean ratios between the cadmium concentrations of limestone bedrock and deriving soils (Cdsoil/Cdrock), calculated for different areas, range from 4.6 to 5.7. Calculations based on the analyses of both soils from a restricted area and fragments of bedrock sampled in the immediate vicinity of high-concentration soils are around 5.5–5.7. Cdsoil/Cdrock is useful in determining the potential of soils in Lower Burgundy to reflect and exacerbate the high concentrations of cadmium present in parent bedrocks.

Ferromanganese deposits from the Knipovich Ridge (North Atlantic)

Geological‐geophysical investigations were carried out on the Knipovich Ridge (North Atlantic) during cruises 24 and 25 of the R/V Akademik Nikolai Strakhov in 2006 and 2007, respectively. Dredging of the ridge yielded not only bedrocks, but also numerous samples of ferromanganese deposits (FMD) as thick boulders or crusts enveloping the bedrock fragments. Samples were recovered from the slopes of basins and rises of the Knipovich spreading ridge between latitudes 77 ° and 79 ° N (Cruise 24), as well as from rises at the southern end of the Knipovich Ridge at its junction with the Mona spreading ridge at ~74 ° N (Cruise 25) (see figure). The FMD samples mentioned above are likely to be the first ones taken from such high latitudes. They are characterized by the frequent occurrence of a bouldertype shape with separate fragments up to tens of centimeters in size. We encountered for the first time wet and relatively soft ore material that could easily be cut with a knife. Thin slices of the ore material were dried rapidly and disintegrated into powder. Signs of bedding were rare. The largest concretions lacked any substrate and likely represented chips of a similar ore basement. The distribution range, thickness, and relatively soft structure of FMDs can be of practical interest. Table 1 presents general information concerning the FMD dredging stations. Table 2 presents the results of the atomic absorption analysis of the chemical composition and brief characteristics of samples. One can see that the samples represent Fe-rich sediments with the average Mn/Fe ratio varying from 0.4 in deposits from the northern stations of Cruise 24 to 0.55 in deposits from stations of Cruise 25. Increase in the Mn content in sediments from stations of Cruise 25 is accompanied by increase in the average contents of Cu (from 0.032 to 0.057%) and Ni (from 0.12 to 0.21%), while the content of Co remains approximately stable (0.22%). Unfortunately, FMD samples from the intermediate sector between the sites established on the Knipovich Ridge are lacking. Nevertheless, we can assume that the rocks exposed in this sector can also be characterized by a high degree of ferromanganese mineralization. It should be noted that the thickness of ore sediments and their composition are atypical of their counterparts found near rift zones of spreading ridges [1]. In general, such zones contain young coatings or fine ferromanganese crusts with low contents of trace elements related to the endogenic activity in pull-apart walls of rifts. Thick ore sediments in the zones mentioned above testify to either the weak manifestation or the absence

Frost-Heaved Bedrock Features: A Valuable Permafrost Indicator

Frost-heaved bedrock features are periglacial forms produced by the vertical displacement of bedrock fragments. Blocks, frost-wedged from bare bedrock along joints, are raised above the general surface by heave. Although mentioned in the literature of the late 19th century, they have been described and studied in detail only recently. They are widespread in the northern hemisphere, notably in Canada, Greenland and Spitsbergen, where they develop in lithologies with well-developed systems of joints. Commonly, heaved blocks exhibit weathered and lichen-covered surfaces except at their base where freshly exposed rock indicates recent heaving. They result from frost processes, particularly from wedging and heaving due to pressure of the freezing of free-water in joints. Active and most non-active features are located in permafrost regions. The southermost occurrence in the northern hemisphere is the Groulx Mountains, in Québec (51°45'N, alt. 1000 m). The mean annual air temperature for the area of best-developed features ranges from -4° to -100C, with the number of days of frost ranging from 178 to 300. A thin snow cover is common to most sites. Based on climatic data and on the geographical distribution of frost-heaved bedrock features, it is suggested that permafrost occurrence is obvious. Relict features found outside the present-day permafrost zones should indicate former permafrost conditions.

The role of glacitectonic rafting and comminution in the production of subglacial tills: examples from southwest Ireland and Antarctica

Sedimentological and structural geological data from two sites in southwest Ireland and Antarctica provide evidence for the formation of subglacial till by the brecciation and crushing of bedrock rafts. Up‐sequence transitions, from undisturbed bedrock, to deformed bedrock, to crushed and brecciated bedrock, to massive matrix‐supported till with far‐travelled erratics, represent a process‐form continuum of till production. Initially, bedrock fragments and rafts up to several metres in length are liberated from the substrate by glacitectonic thrusting and plucking. These rafts are then crushed to produce the matrix of a till. Such products are commonly referred to as comminution tills, although the original definition focused primarily on the second phase of the process (crushing of bedrock rafts and fragments) as well as abrasion of bedrock. Data from Ireland and Antarctica indicate that rafting of bedrock is an essential part of the process of till formation. This process is facilitated by weak sedimentary bedrock, which can be displaced along joints and bedding planes to form rafts that are then incorporated into the ‘proto‐till’ prior to being crushed subglacially. Our field data suggest that bedrock failure and displacement of such rafts can occur to depths of 3 m. The occurrence of erratics in the uppermost part of the till demonstrates that the glacier effectively mixes far‐travelled and local materials.

First Characterization and Dating of East Antarctic Bedrock Inclusions from Subglacial Lake Vostok Accreted Ice

Environmental Context.Lake Vostok is a large subglacial lake trapped below the East Antarctic ice sheet. The meteoric ice from deep Vostok ice cores has been used to document the climatic history of the Earth over hundreds of millennia, while the deeper part of the core preserves some basal rock fragments. These rock fragments represent unique geological samples of the inhospitable, ice-covered East Antarctic Plateau. Abstract.The Vostok (East Antarctica, 78°S, 106°E) ice core preserves, below the meteoric ice keeping the climatic memory of the last 420 000 years, ice formed by freezing of subglacial Lake Vostok water. This latter contains some bedrock fragments representing unique samples for the geological investigation of the East Antarctic Plateau, covered by ~2–4 km of ice. The first geochemical (87Sr/86Sr versus 143Nd/144Nd) and mineralogical characterization of these inclusions as well as the dating of one of them (Nd model age on whole-rock sample) has given evidence for a Mid-Proterozoic age of the basement lying below the ice sheet, consistent with recent geophysical data. The geochemical characteristics of bedrock inclusions within the accreted ice zone are markedly different from those of the mineral dust of aeolian origin archived in the uppermost part of the Vostok ice core and originating from deflation of the Southern Hemisphere continents, and easily discriminates between the two contributions.

Measuring the Extent of Chemical Weathering in Subarctic Alpine Environments: Implications for Future Research

Chemical transformations associated with the weathering of bedrock to regolith, and ultimately to soils are poorly known despite the importance of these transformations in the generation of surface and groundwater solutes, the preparation of bedrock for erosion, the production of plant nutrients, and ultimately the formation of soil itself. This lack of knowledge is particularly apparent in polar and subpolar lands where traditionally chemical processes have been regarded as subordinate to mechanical weathering processes. This paper examines the nature of chemical changes associated with the transformation of bedrock to soil in three subarctic alpine locations by comparing weathering indices calculated from whole-rock chemical analyses of surface soil, regolith, rock fragments, and bedrock. The results of this largely post hoc integration of our previous reconnaissance studies suggests that as bedrock weathers to soil there is a loss of silicon, alkali earths, and alkalis relative to resistant chemical constituents, as might be expected from first principles. However, the loss is not systematic, as bedrock fragments derived from crustal rock represent the most resistant elements of a weathering/soil profile, and surface soils commonly experience eolian addition of chemical constituents. The results of this study are viewed as being “enticing” and suggest a potentially productive line of chemical weathering investigation in polar and subpolar environments that will throw light on patterns of release of chemical constituents in the initial stages of the production of residual soils.

Gold distribution in glacial sediments and soils at Boston Property, Nunavut, Canada

Geochemical exploration for gold in the Slave Structural Province, Nunavut, Canada is hampered by a complex cover of Quaternary sediments and a lack of information on the effects of glacial dispersion and periglacial conditions on gold distribution. This study investigates distribution of gold in surficial materials on the Boston Property in the southeastern portion of the Hope Bay Greenstone Belt, 650 km northeast of Yellowknife, N.W.T. The principal surficial materials are till with a discontinuous veneer of post-glacial marine sediments. In areas of shallow overburden, frost boils mix weathered bedrock with drift and bring bedrock fragments to the surface. Samples were collected from till and from soil profiles on lines across the mineralized zone. The −212 and −53 μm fractions were separated by wet sieving and heavy mineral concentrates (HMCs) prepared. Gold was determined by fire assay–atomic absorption spectroscopy (FA–AAS) or by a more sensitive aqua regia-column preconcentration–ICP procedure. There are two types of gold anomalies: (1) anomalies in the regional till, and (2) anomalies developed by ongoing frost boil action and colluvial processes. Anomalies are absent in marine sediments. Gold values in the regional till range from 15 to 1030 ppb in −106+53 μm HMCs and from 10 to 65 ppb in the −53 μm fraction, which contains ~60% of the total gold content. Anomalous values extend at least 2 km down-ice from the gold mineralization. In profile, the most recent gold anomaly is mushroom-shaped, up to ∼100 m wide at the surface, and drapes over a ridge of mineralized bedrock. The anomaly in near-surface horizons is 10–50 m broader than the anomaly in the underlying horizons. Maximum gold values (>5000 ppb) are found in areas of weathered bedrock and where frost boils include abundant fragments of mineralized bedrock. For reconnaissance-scale exploration, frost boils in the regional till should be sampled on a 500×40 m grid with lines perpendicular to ice-flow direction. The −53 μm fraction should be analyzed for gold with a detection limit of 1 ppb or lower. At the property scale, with anticipated higher gold values, it is feasible to analyze the coarser −212 μm fraction from near-surface samples collected at a 10–20 m spacing along grid lines 100 m apart.

Lithological evidence for the evolution of weathered mantles in Uganda by tectonically controlled cycles of deep weathering and stripping

Landscape evolution in terrains that have been unaffected by glacial or aeolian erosion, occurs by cycles of deep weathering and stripping. Several factors have been proposed to control these cycles including sea level, climate and tectonics. In this study, a tectonic model of landscape evolution recently developed for Uganda, was tested by detailed study of the weathered mantle. The study involved an analysis of weathered-mantle stratigraphy, texture, mineralogy and elemental geochemistry, and was conducted in two catchments of central and southwestern Uganda which feature contrasting geomorphic processes (deep weathering and stripping) and different (post mid-Miocene) evolutionary histories. Application of graphical (sedimentological) logs to weathered-mantle stratigraphy is introduced in this work and represents an improvement upon more subjective classification schemes employed previously. The stratigraphy, texture, mineralogy and elemental geochemistry of the weathered mantle support cycles of deep weathering and stripping predicted by the tectonic model of landscape evolution. In central Uganda, deep weathering of the landsurface as a result of tectonic quiescence since the mid-Miocene is indicated by thick (30 m) weathered profiles that result from prolonged chemical denudation. This is evident from the progressive alteration of crystalline rock from relatively unweathered bedrock fragments at the base of the profile to a ferricrete crust at the top. Lithological evidence indicates that the weathered mantle has developed in situ and that the water table separates environments of different weathering intensity. In southwestern Uganda, stripping of the landsurface following mid-Pleistocene uplift is reflected by thinner, truncated weathered profiles that are not capped by ferricrete and are composed primarily of coarser, less weathered material.

Kelp-plucking: coastal erosion facilitated by bull-kelp Durvillaea antarctica at subantarctic Macquarie Island

Erosion of bedrock from lower intertidal reefs by waves acting on attached plants of bull-kelp (Durvillaea antarctica) was investigated over one year at Macquarie Island, Southern Ocean. At a site on the more sheltered east coast, such erosion occurred during four separate storms, each casting up 834–1078 large kelp plants km−1 of coast, of which 30–45% were still attached to jagged, freshly quarried bedrock fragments over 2.5 cm long. The largest fragment weighed was 74.6 kg; rounded cobbles and boulders attached to kelp plants and weighing up to 102.2 kg (and probably more than 160 kg) were also cast up on beaches. 19–21% of the standing crop of large kelp plants was removed by storms during the year of observation. Break points for 10 kelp stipes were found to be at least 90–161 kg. Total annual erosion by kelp-plucking is at least 1.56 tonnes of rock km−1 of coast. However, in terms of erosion this computes to only 0.1 mm yr−1, far below the rate of uplift of the island.

The Role of Debris-Flow Deposits in the Development of Dryland Salinity in the Yass River Catchment, New South Wales, Australia

Aeolian deposits known as parna are remobilised as debris flows that occur widely across the Yass River catchment in the state of New South Wales, Australia. The deposits consist of smectite-rich clays that contain varying salt loads and have a grain size that ranges from very fine to fine silt. During the Pleistocene Epoch, debris-flow mechanisms mixed the aeolian material with locally derived windblown sand and fragments of bedrock. Several episodes of debris-flow emplacement are identified, and these may correlate with climatic variations during the Pleistocene Epoch. Multiple episodes of debris-flow deposition and removal probably occurred during this time. Conditions in the catchment prior to clearing represented a meta-stable phase in the erosion of the more recent debris-flow deposits. Destabilisation could have resulted from the onset of a drier climate or by the clearing of deep-rooted vegetation.

GROUNDWATER CONDITIONS AROUND A COAL-COMBUSTION RESIDUE SITE IN SOUTHERN ILLINOIS

Coal-combustion residues, predominantly composed of scrubber sludge, served as backfill to an abandoned surface-mine pit. A low-relief valley now trends across the area that was once the old pit, replacing the former hummocky terrain left by the mining operations. Twelve wells, installed within and adjacent to the fill, provide data on hydraulic head at the base of the spoil, in the coal, and in the coal-combustion residues. The bases of the screens ranged in depth from 5.6 to 15.9 m. Three of the wells were sampled for groundwater chemistry. Hydraulic conductivity of the spoil from nine falling-head (slug) tests ranged from 1.9X10{sup -7} to 4.1X10{sup -3} cm/s, with a geometric mean of 1.8X10{sup -4}, somewhat greater than the 9.8X10{sup -6} cm/s hydraulic conductivity of the coal-combustion residues. The mine spoil is heterogeneous, composed of disturbed surficial sediment, including loess and diamicton, blended with fragments of bedrock. Prior to reclamation, groundwater flowed away from the pit lake to topographic lows to the north, east, and south. Reclamation has not altered groundwater flow significantly. A two-dimensional numeric random-walk contaminant-transport model, based on (1) properties of the coal-combustion residue as determined from column studies, (2) data from a finite difference model of groundwatermore » flow at the site, and (3) conservative estimates of dispersivity and porosity, suggests that leachate generated by the fill poses little threat to ambient groundwater quality.« less

Dynamic Glacigenic Deformation of Near-Surface Metamorphic Bedrock: Heath Steele Mines, New Brunswick

Near-surface metamorphic bedrock at Heath Steele Mines, New Brunswick, was fractured to depths of at least 35 m and possibly as deep as 120 m, during two stages of glacier overriding. Fracturing was accompanied by crack-filling of sediment-laden melt waters, increased pore pressures, freezing, and joint expansion. The upper few meters of the bedrock surface were extensively brecciated and folded. Brecciated bedrock fragments were reoriented as passive strain markers of interstitial ice deformed by subglacial shear. Dynamic glacigenic deformation is distinguished from older structures by: (1) structural associations with direction of glacial movement; (2) fracture infillings of distally derived and striated material; (3) iron-staining along open fractures; (4) near-surface strain partitioning of deformation, and (5) absence of axial plane foliations in fold structures.

Origin and Distribution of Sulfate in a Fractured Till in Southern Alberta, Canada

Glacial till in a 22‐km2 area of the Interior Plains Region of southern Alberta, Canada consists of an upper brown gypsiferous weathered zone (up to 20 m thick) and a lower discontinuous grey nonweathered zone (up to 25 m thick). Although the total sulfur content of the two till zones is the same (0.34%), the mean total sulfate content of the weathered till (4.0 mg g−1) exceeds that of the nonweathered till (0.8 mg g−1). Similarities suggest that all SO42− in the nonweathered till and 20% of the sulfate in the weathered till was derived from sulfate‐rich bedrock fragments incorporated in the till during deposition. Laboratory experiments and δ18O and δ34S analyses of total SO42− and the sulfur forms in the two zones show that the majority of the sulfate in the weathered till was derived from the oxidation of organic S at some time after the till was deposited. Mass balance and flux calculations indicate that this oxidation occurred under partially saturated conditions. Groundwater flow simulations suggest that the water table was much lower and that a thick partially saturated zone was possible during the drier Altithermal period (11,000 to 3000 years B.P.). The following chemical processes account for the major ions that are found in the groundwater but were generated when the partially saturated zone existed. The oxidation of reduced S produced SO42− and H+. The H+ reacted with carbonate minerals under elevated PCO2, with Ca2+ loss by exchange for Na+ and gypsum precipitation. Volume shrinkage in the weathered till resulting from these geochemical processes may have enhanced the development of fractures and caused the associated increased hydraulic conductivity.

Micromorphology of Pech-de-l'Azé II sediments

A micromorphological study of the sediments from the cave of Pech-de-l'Azé II using the terminology of Brewer (1964) was undertaken in order to describe them, to compare the results and their palaeoclimatic interpretations (e.g., Laville, 1973) with those of classical sedimentological analyses and to evaluate the role of micromorphology in the analysis of cave and rock-shelter deposits. The soil skeleton is composed of quartz sand and silt resulting principally from the decay of bedrock with local reworking by solifluction and ruissellement. Lithorelicts consist of bedrock fragments of cryoclastic origin and detrital phosphate grains which can be identified as being large carnivore (bear or hyaena?) coprolites. The plasma is composed of mostly iron rich clay and although it was most abundant in reddish layers of previously recognized soils, no micromorphological evidence for clay illuvation was found. The latter might have been caused by post-depositional destruction by freeze and thaw which produced the generally well developed banded fabrics in many of the beds. Free grain argillans occur sporadically but are particularly well developed on phosphate grains. Secondary calcite, occurring either in dispersed zones or as void calcitans and neocalcitans occurs throughout but is more marked in the upper layers and bed 6. Micromorphologically based palaeoclimatic interpretations agreed on the whole with those obtained by classical sedimentology although several questions were raised which should make us more cautious in interpreting “classical” data: such as the origin of the clay, especially in “soil” horizons. The advantages of micromorphology in delineating sequence of events were indicated.

Fluidisation and bedrock fragments in ejecta

Rounded and indented bedrock fragments amongst pyroclastics are evidence of fluidisation during maar eruption. Some fragments also show shrinkage cracks caused by volcanic heat. Compaction of bedding may form striated cone-like features after deposition.