We drilled a 210 m-thick succession of Quaternary sediments and extended it 30 m upsection with information that we collected from an adjacent outcrop. In the 240 m-thick succession we identified 12 different lithofacies, grouped them into five facies assemblages, and distinguished two major sedimentary sequences. A sharp contact at 103 m depth cuts off cross-beds in sequence A and separates them from the overlying horizontal beds in sequence B. Although the lowermost facies assemblage of each sequence includes a till deposited during a period of ice cover, the two tills differ from each other. In particular, the till at the base of sequence A is dominated by large clasts derived from the underlying Molasse bedrock, whereas the till at the base of sequence B has no such Molasse components. Furthermore, the till in sequence A bears evidence of glaciotectonic deformation. Both tills are overlain by thick assemblages of subaqueous, most likely glaciolacustrine and lacustrine facies elements. The cross-bedded and steeply inclined sand, gravel, and diamictic beds of sequence A are interpreted as deposits of density currents in a subaqueous ice-contact fan system within a proglacial lake. In contrast, the lacustrine sediments in sequence B are considered to record a less energetic environment where the material was most likely deposited in a prodelta setting that gradually developed into a delta plain. Towards the top, sequence B evolves into a fluvial system recorded in sequence C, when large sediment fluxes of a possibly advancing glacier resulted in a widespread cover of the region by a thick gravel unit. Feldspar luminescence dating on two samples from a sand layer at the top of sequence B provided uncorrected ages of 250.3
Periods of global cooling have caused fluctuations in ice volume throughout the Quaternary, which is, for example, recorded in marine sediments
Among the various glacial archives, surficial deposits of the Last Glacial Maximum
Here, we conducted such a drilling in the confluence area of the Rhône and Aare glaciers (Fig.
The study area is located in the Middle Aare Valley (MAV) between the border of the Alps and the city of Bern (Fig.
In the MAV, the bedrock consists of early Miocene fluvial sandstones and overbank fines, which are referred to as the Lower Freshwater Molasse
A first 160 m-deep drilling on the western flank of the Aare branch (Fig.
Left: modified log of the Thalgut drill core
A second drilling conducted north of Bümpliz into the Meikirch trough (Fig.
The main goal of this study was to recover the record of at least one glacial–interglacial cycle, which should, preferably, include the Eemian (i.e., MIS 5e) or any older interglacial. We considered locations with inferred sediment thicknesses over 100 m to increase the probability of recovering one full glacial–interglacial cycle. Furthermore, the newly drilled section should potentially allow a correlation with the previously drilled MAV suites. Thus, we collected information from drilling reports and geological maps to assess the stratigraphic architecture of the MAV sedimentary fill. We targeted outcrops of Pleistocene lacustrine deposits during this reconnaissance, because they had the greatest potential to contain fine-grained sediments that could be enriched in pollen and were most likely formed within the desired age range. Using these constraints, we located the drilling above the Bümpliz trough (Fig.
Drilling was conducted by Fretus AG between February and May 2019, using a Nordmeyer DSB 1/6 mobile rig to a total depth of 211.5 m. A total of 207 core sections, each 1 m long, and 9 sections, 0.5 m in length, were retrieved in plastic liners. A down-the-hole percussion coring tube was used up to a depth of 37 m, and it was replaced by a CSK-146 wireline diamond coring system for the remaining 171.5 m of unconsolidated sediment and 3 m of bedrock. The first 30 m were cased with 324 mm casing, followed by 6 m with a 280 mm casing, while the lower diamond coring section remained uncased. When the rod was removed to exchange the drill bit, the hole was stabilized by a drilling mud consisting of water with additions of biodegradable polymer and bentonite. Water inflow was recorded at depths of 3–4 and 8–9 m. Major drilling fluid losses occurred at a depth of 168–169 m. A core recovery rate of 92.3 % was achieved, 6.6 % of the core had lost all structures due to disturbance during drilling, and only 1.1 % of the core was physically lost.
After retrieval, the liners were capped, provisionally marked, and kept in wooden boxes with direction markers, name tags, and the respective driller depth to allow proper storage. The cores were weighed in as a first estimate of core recovery. Parallel to the drilling operations, cores were regularly transported to the Institute of Geological Sciences, University of Bern, for their investigation in the off-site core laboratory. The drill hole was cemented after the completion of the drilling operations, following official cantonal regulations. The core is stored in the Swisstopo core repository and is available for further investigation.
We measured
After splitting, individual cores were logged (Sect.
The sediment was logged at the scale of individual beds in the direction of the drilling, i.e., downsection. The description followed a modified template of
Two samples (RE-01 and RE-03) from a well-sorted, 10 m-thick sand unit within the succession at the Rehhag outcrop were collected for feldspar luminescence dating (Fig.
Luminescence measurements were analyzed with the R.Luminescence package
A total of 10 samples (2 cm
Throughout the core, 16 clayey silt layers and an additional 49 silt and sand layers were sampled for the determination of their carbon content. The TC, TIC, and TOC concentrations were measured with a Thermo Scientific Flash 2000 Smart (Thermo Fischer, Waltham, MA, USA).
The CaCO
The Rehhag drilling recovered a total of 211.5 m of core material (Fig.
Drill log of the presented Rehhag drilling and the sequentially overlying outcrop in the Rehhag clay pit. Blank spaces in the log show considerable core loss. The depth from the top of the drilling is shown on the left-hand side and absolute elevation on the right-hand side of the figure. The figure also displays the logging results of the magnetic susceptibility,
The overlying 208.5 m-thick Quaternary suite (Fig.
This is a summary of the 12 lithofacies elements identified in the drill core, their main properties, and interpretation. Detailed
descriptions can be found in Appendix
Photos of eight illustrative core sections. For lithofacies descriptions, see Appendix
Contacts with a dip angle
A staircase structure cuts across the contact between a medium-grained sand and a silty fine-grained sand in the center (46 cm section depth) of the core section 124–125 m. This staircase consists of three steps that have relative vertical displacements of 4 and 1.5 cm from the lower left-hand to upper right-hand sides of the core, respectively. Upsection of the staircase, beds appear to be displaced along two subvertical, undulating lines that separate the three steps. Downsection of the staircase, the displacement becomes less visible, and it dissipates into small-scale ruptures and irregularities from 55 cm and below. We consider these steps to result from a displacement of the material along a set of two subvertical shear planes (i.e., faults). Because the fault traces dissipate downsection of the staircase, we interpret this structure as resulting from strike-slip rather than normal faulting.
A third feature was found in the lowermost 3.5 m (205–208.5 m) above the bedrock contact within diamictic beds of Dmm.1 (Fig.
The mean density within the bedrock section of the core is as high as 2.55 g cm
This is a summary of the identified lithofacies assemblages, their extent from the top of the drilling, their extent in absolute elevation (surface 571 m a.s.l.), and the different lithofacies they comprise. The table also shows how the facies assemblages are grouped into the major sequences. Along with the assemblages, the mean values of the measured
After the rejection of outliers, the magnetic susceptibility (MS) data show two distinct patterns along the core (Fig.
Generally, MS peaks are caused by fine-grained sediment particles or gravel clasts with an iron component. Highest MS peaks can be caused by metallic particles chipped off the drill head and rod. Furthermore, data peaks are encountered in core sections that were disturbed by the drilling (e.g., meters 171–173).
Following the first step of the R routine, which was applied to the IRSL measurements of samples RE-01 and RE-03, the
Details of the IR50 and pIRIR225 luminescence analyses of samples RE-01 and RE-03, including the details of the first step analysis of the apparent (i.e., uncorrected)
Despite the additional hydrofluoric acid treatment, the concentrations of fine-grained lithogenic components in the pollen samples were still remarkably high. Pollen concentrations were extremely low in all samples. In fact, pollen was found in only 4 of the 10 samples, and never more than three grains were encountered. Four grains could be identified as either
The mean TOC throughout the core is 0.19
The sorting, stratification, and fabric encountered in lithofacies 1 to 10 point towards a deposition mechanism with support by water (Sect.
We interpret Dmm as a subglacial till because the lowest CaCO
If the interpretation of the Dmm as subglacial till is correct, then the oversteepened beds, strike-slip faulting, and subhorizontal shear planes described in Sect.
The oversteepening described as the first structure in Sect.
Subvertical faults and displaced bedding in the preglacial substratum could be the result of loading and the anisotropic compaction of the subglacial stratum. Alternatively, differences in the longitudinal motion of the glacier and differences in the motion transfer to an anisotropic substratum could result in the occurrence of strike-slip faulting. However, we are not aware that similar structures have previously been described and explained by glaciotectonics. Thus, the ambiguity in determining the formation process of the second feature in Sect.
Subhorizontal shear planes with slickensides as encountered at the base of the Quaternary suite were also observed beneath Breiðamerkurjökull
In summary, the subglacial tills represented by the lithofacies Dmm.1 and Dmm.2 as well as their subjacent beds bear deformation features that most likely resulted from glaciotectonic processes
In this section, we group the 12 lithofacies elements into 5 FAs (Fig.
FA 1 comprises beds of Dmm with interbedded Sdf, Shb, and Scb. This assemblage is encountered at depths between 84 and 103 m as well as 194 and 208.5 m and is dominated by the presence of Dmm beds, which overlie their substratum with sharp contacts.
The structures preserved by the Dmm beds have been interpreted as recording the occurrence of glaciotectonic processes, which we have linked with the shear mechanisms that occur below a moving glacier.
In such an environment, the interbeds could be interpreted as subglacial deposits that were detached from the substratum and thrust between layers of till as floes of intact sediment or reworked sediment patches
FA 2 comprises an ensemble of the seven lfs Ghb, Gcb, Sm, Shb, Scb, Slm, and Dms and occurs at depths between 141 and 194 m upon a gradual contact with FA 1 sediments. Planar cross-bedding (10–30
FA 3 assembles beds of the five lfs Dms, Sm, Shb, Scb, and Sdf. This assemblage is found at depths between 103 and 141 m and overlies FA 2 upon a gradual contact. In contrast to the FA 2 sediments, FA 3 comprises up to 50 % of Dms beds recording the dominance of hyperconcentrated density flows. Because such currents could result from the efflux of subglacial conduits, we consider that the Dms beds record the occurrence of an ice-contact fan
FA 4 assembles beds of Sm, Shb, Sdf, Fm, and Fdf. This assemblage is found in the outcrop
FA 5 comprises beds of Gsc only. This assemblage is found only in the outcrop within the Rehhag clay pit at heights of
The pollen content of the encountered sedimentary suite was too low to allow a relative chronological positioning of the sedimentary succession and a paleoenvironmental reconstruction. We explain the low pollen concentration by the energetic depositional environment, which was dominated by the sediment supply through different density and turbidity flows. For pollen to accumulate, a persistently stagnant water column would be required, as was probably the case at the Meikirch and Thalgut sites. In fact, we have not encountered laminated and organic- and pollen-rich silt and clay layers in the drilled sedimentary suite.
The luminescence dating on a well-sorted sand of the mouth bar delta in FA 4 was more successful than the pollen analysis. Yet the luminescence signal of the tested sediment lies in the high dose range of the feldspar luminescence system (Fig.
In the following, we use the succession of distinct FAs encountered in the Rehhag core and clay pit to reconstruct a scenario of how the landscape evolved throughout the deposition of the analyzed sedimentary succession (Fig.
Schematic figures showing the evolution of the Bümpliz trough (Sect.
Subsequently, flow tills (i.e., hyperconcentrated density flows) dominated the sedimentary processes in a proximal ice-contact fan environment (FA 3), thus documenting a second advance of a glacier close to the depositional site (Fig.
Overall, the sediment that was retrieved from the Bümpliz trough documents a shallowing-up sequence. Following an erosional glacial advance, the formation of the sequence started in an underfilled trough or a basin. After one break of the sequence, the lake basin was filled to a paleo-base level and evolved into overfill conditions. This overfill probably occurred in parallel with the third advance of a glacier close to the Bümpliz trough. Therefore, and as already suggested in Sect.
The segmentation of sequences A and B is additionally supported by the results of the
We analyzed 208.5 m of unconsolidated Quaternary sediment and 3 m of Molasse bedrock in a drill core recovered from the Bümpliz trough, a branch of the overdeepening in the MAV. It is the first scientific drilling in the MAV that reached bedrock. The retrieved sediments revealed two major sequences, A (lower) and B (upper), with each of them starting with a till deposited when the area was glaciated. Each glacial period was accompanied by erosion of Molasse bedrock as documented by the relative depletion of the CaCO
This lf is composed of an amalgamation of moderately to well-sorted, clast-supported, fine-grained to cobble-sized gravel beds and is only encountered in the Rehhag outcrop (Fig.
This lf is composed of moderately sorted, clast- and matrix-supported, fine- to medium-grained gravel beds. The matrix can vary from fine- to coarse-grained sand, which has a brownish-gray to gray color (Fig.
This lf comprises moderately sorted, clast- and matrix-supported, fine- to medium-grained gravel beds. The matrix is composed of fine- to coarse-grained sand, which is brownish gray to gray. Individual beds are between 5 and 70 cm thick and show inclinations where the dip angles range from 15 to 30
This lf is composed of well-sorted, massive, silty fine- to medium-grained sand beds. The color of the material varies between brownish gray and gray brown. Individual beds are between 30 and 100 cm thick and commonly have a sharp base. Individual beds and layers do not show any evidence for inclination, and thus we consider them to be horizontally bedded. Isolated clasts of coarse-grained sand to medium-sized gravel are embedded within the silty sand and are considered outsized clasts. Such clusters of outsized clasts are encountered throughout entire beds or in limited sections within individual beds.
This lf comprises well-sorted, silty fine- to coarse-grained sand. In the core, the color varies between brownish gray and gray brown (Fig.
This lf comprises well-sorted, silty fine- to medium-grained sand. Bed thicknesses range from 5 to 40 cm. The color varies between brownish gray and gray brown. Individual beds have a sharp base, and they are inclined with dip angles between 10 and 35
This lf is composed of well-sorted, silty fine- to medium-grained sand. The color varies between brownish gray and gray brown. Individual beds are between 20 and 50 cm thick and have a sharp basal contact. The layers are either horizontally bedded or inclined with dip angles between 10 and 35
This lf comprises well-sorted, silty fine- to medium-grained sand. The color varies between brownish gray and gray brown. Individual beds are between 5 and 50 cm thick and have either sharp or gradual basal contacts. The beds can be horizontally bedded or are occasionally inclined. Internally, this lf shows a slightly chaotic arrangement of layers as well as wavy and folded structures. These waves and folds are visible due to slight changes in sediment composition.
This lf comprises layers that are composed of clayey silt and beds with silty to fine-sandy medium-grained sand. The color varies between olive gray and gray brown (Fig.
This lf comprises massive, very well-sorted clayey silt layers (Fig.
This lf is poorly sorted and comprises subangular to rounded outsized clasts that range in size from coarse sand to cobble and that are embedded in a silty fine-grained sand matrix (Fig.
This lf is very poorly sorted, matrix-supported, and comprises coarse-grained sand to cobble-sized clasts that are embedded in a silt to medium-grained sand matrix with a brown to ocher color (Fig.
The beds of Dmm.1 at the base of the cored sediment contain sandstone clasts of the LFM bedrock. These clasts range from coarse-grained sand to boulder size with the largest components close to the bedrock, and these are angular to subangular. Alpine clasts are absent within the lowermost meter of the cored sediment. Above this lowermost meter the proportion of Alpine to Molasse constituents increases upsection. Furthermore, patches and bands in the Dmm.1 matrix have a similar or even the same color as the red-, yellow-, and green-colored Molasse bedrock. The beds of Dmm.2 only contain Alpine clasts, and blocks derived from the LFM are missing.
Two samples (RE-01 and -03) were recovered from a ca. 10 m-thick set of Shb within FA 4 for the measurement of infrared stimulated luminescence at 50
Modified post-IR IRSL 225
IRSL measurements were analyzed using the R.Luminescence package
Analysis of luminescence measurements for sample RE-03, using R.Luminescence
All data obtained during the core processing and luminescence measurements as well as the respective analyses can be found in a separate data publication (
MAS analyzed the data and wrote the paper, with support by FS. MAS designed the drilling campaign and FS organized the funding. PS conducted the pollen analysis, and DB was responsible for the gravity survey. NG supported the OSL dating, and GAD provided input during fieldwork and core analysis. All the authors approved the text and the figures.
The authors declare that they have no conflict of interest.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We acknowledge the professional drilling by Fretus AG that provides a core with excellent quality. Special thanks also go to Toggenburger AG for dislocating their equipment and for giving access to the drill site and to Rehhag AG (Roland Schütz) for offering the ground for drilling. Peter Hayoz is kindly thanked for transporting the drilled material to the Swisstopo core repository, where the material is currently being stored. The carbon content analyses were run by Julia Krbanjevic, Institute of Geological Sciences, University of Bern. Gamma spectrometry was carried out by Sönke Szidat, Department of Chemistry and Biochemistry, University of Bern. We thank the research group for paleoecology of Willy Tinner at the Institute of Plant Sciences, University of Bern, for providing their facilities for the palynological investigation. We thank Urs Marti at Swisstopo for providing the gravimeter for the conduction of the gravimetry survey. We are grateful to the Direktion für Tiefbau, Verkehr und Stadtgrün of Bern for helping with the carving of trenches in the Rehhag clay pit.
This research has been supported by the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (grant no. 175555), Swisstopo, the Stiftung Landschaft und Kies, the Gebäudeversicherung GVB, the Aaretal Kies AG (KAGA), and the city of Bern.
This paper was edited by Ulrich Harms and reviewed by Cesare Ravazzi, Volker Wennrich, and one anonymous referee.