Survey and sampling of tephra samples was conducted at three floodplain transects (including active floodplain zone and lower terrace) in the south of Niederweimar gravel quarry where the floodplain shows one of its greatest transverse expansions (Fig. 1). Transects cover a maximum width of 973.9 m and are at 170.9 m a.s.l. (above sea level) with a height difference of ±1.4 m. Sampling was carried out with the help of a hand auger (Pürckhauer, Ø 2 cm, 2 m depth) and pile core probing (Ø 6–8 cm, 3 m depth). Soil properties and stratigraphy were documented in the field according to the national standard soil classification scheme (KA5, Ad-hoc AG Boden, 2005). Tephra-containing layers were identified visually (visible tephra grains) or by smeary consistence (owing to higher contents of allophane) (Jahn et al., 2006), proportion documented (percentage of tephra grains estimated by area according to KA5), and extracted for further analysis.

Figure 1Study area located within the Lahn River catchment in the south of Marburg and transects with interpolated upper and lower tephra boundaries. (a) Transects with sampling points (drill cores) and location of reference profile within the quarry. (b) Interpolated upper and lower tephra boundaries under terrain surface. Deviation in the interpolation (lower limit) results from large distances between the boreholes (no borehole due to missing permission). Data basis: Hessian State Agency for Soil Management and Geoinformation (HVBG, 2019).

Samples for method validation and as comparison material (reference samples) were taken from a recently excavated profile (32U 480710 5621590) in the Niederweimar gravel quarry, which was excavated during archaeological work. The profile consists of Holocene floodplain loams (silt to sandy loam) above four LST layers (sandy loam, partwise alternating with black sands and ripple) and flood loam (late Pleistocene) at the base (Fig. 2). Three samples were taken from the upper (LST bands mixed with sandy loam), middle (thick LST bands, between black sands) and lower (LST bands interrupted by black sands with ripple shapes) parts of the LST layer (Fig. 2).

Figure 2Lithological description of reference profile (with exemplary separated tephra grains) and exemplary drill cores (104: north transect; 304a: south transect).

Download

Reference samples and samples from the transects were dried at 100 ^∘C (drying chamber) under weight loss control, subsequently carefully ground, and sieved to <2 mm (stainless steel sieve, Retsch, Haan, Germany). Subsample material (20g per sample) was then mixed with 150 mL saturated NaCl solution (density adjusted, ρ=1.2 g cm⁻³) within glass beakers, stirred (1 min, magnetic stirrer) and allowed to sediment for 20 min. This allows the tephra grains to be separated from other mineral components (sand to clay grains), despite the heavy minerals contained (Lomax et al., 2018), because the grains float in the dry state due to their many cavities (volcanic origin). The floating tephra grains were then sieved to >50 µm (Atechnik, Leinburg, Germany) to separate clay and/or silt particles and filtered by vacuum filtration (cellulose filter, LLG-Labware, Meckenheim, Germany) to rinse out remaining salts. Dried filters were visually examined using a stereomicroscope (Motic SMZ-161 TL, Motic, Hong Kong). From the reference samples, 30 randomly selected tephra grains were extracted and measured (Moticam, Motic, Hong Kong). Samples from the transects were inspected according to the presented method, and the presence of tephra grains or their fragments was considered a positive finding. In addition, 19 samples containing proven LST were selected, and the particle size distribution was determined according to DIN ISO 11277 (2002) and the integral suspension pressure method (Durner et al., 2017).

In total 56 tephra containing stratigraphically distinguishable layers were identified and sampled. Qualitative tephra analyses within the laboratory show a positive rate of 69.6 % in which case tephra could be detected under the microscope for the 56 samples. In 30.4 % of the layers, no tephra grains or larger fragments could be found even if the Greasing effect occurred in the field. The applied method is therefore suitable for qualitative detection, is fast and inexpensive, and can be extended by other methods such as mineral analysis and dating. Tephra grains occur usually entire with grey-brown to greenish colours, clear holes and glassy surface structure (Fig. 2). The length to width ratio of the grains from a random sample of 30 grains comprises an average length of 782.7 (±288.4) µm and width of 557.12 (±179.5) µm. Grain surface and average and maximum length (1565.9 µm) correspond clearly to the reference samples (same colour, holes and glassy structure) with an average length of 724.3 (±245.6) µm.

Tephra layers come up at average depths of 68.6 cm below surface and end at average depths of 166.5 cm (±46.3 cm). Tephra layers have thicknesses from 6.0 up to 132.0 cm with an average thickness of 52.0 cm (Figs. 1b and 2). Grain size analyses of tephra-containing layers has a mean distribution of 30.2 % clay, 46.8 % silt and 23.1 % sand. However, the share of sand is ranging between 2.7 % and 57.9 %, resulting in the samples comprising the grain size classes clay, silty clay, silty clay loam to silt loam (outliers within loam and sandy loam), and they are thus very heterogeneous.

Stratigraphic classification of tephra layers corresponds to the findings of our reference profile and other profiles within the Niederweimar quarry (e.g. Lomax et al., 2018): The tephra layers are covered by Holocene floodplain sediments (floodplain loams, silty loams) and rarely with single clay layers or isolated gravel deposits. The lowermost tephra layers, partly with the same pattern of banding (LST bands and black sands) (Fig. 2), however, are only poorly visible within the drill sample in contrast to quarry profiles. Below the tephra, a thin band (approx. 5 cm) of sand with gravel occurs, followed by a thick layer of clay (dark and rich in organic matter), before the gravel deposits of the lower terrace begin.

Regarding the vertical and lateral spatial distribution of tephra layers it can be stated that it is present nearly all over the floodplain area of the middle Lahn valley. The interpolation of the upper and lower tephra boundaries (Fig. 1b) shows that it is independent of the terrain surface today. This indicates that the tephra follows the morphology of Pleistocene gravel and flood loam deposits, as observed within the quarry (reference profile). The tephra is missing at the edge of the floodplain (drill point 101a: transition to the lower slope, colluvial formation), in the area of the active floodplain (drill points 108a and 2010a: river erosion) and in parts with inactive channel situations (drill point 304b) where Pleistocene gravel structures are found directly below the terrain surface. The interpolation also indicates, despite spatial inaccuracies, that the position of the lower tephra border allows for a partwise reconstruction of the terrain surface that existed at the time of deposition, with structures of various flow channels preformed in the Pleistocene.

Regarding the depositional conditions for LST within the Lahn River valley the heterogeneous grain size distribution indicates slow to medium flow velocities (clay–silt deposition) during LST deposition. From the findings presented in this report an area-wide deposition of LST in the middle Lahn valley can be assumed, which encompasses the entire width of the floodplain. LST deposits seem to follow the structure of preformed Pleistocene flow channels throughout the floodplain, as already stated for the quarry profiles (Bos and Urz, 2003; Lomax et al., 2018). The heterogeneous grain size distribution, banding of the LST with sandy loam intermediate layers and the absence of LST in the floodplain margin and active floodplain area (erosion) indicate fluvial deposition. Therefore, it can be concluded that the LST deposits originate from the entire upper catchment area and have been reworked within the entire floodplain.

The digital elevation model is used with the permission of the Hessian State Agency for Soil Management and Geoinformation. All further data generated during this study are included in this article or are available from the corresponding author upon request.

CJW conceived the project. CJW and VMHD collected reference samples and performed the profile description in the field. VMHD performed the spatial survey and further sample collection. CJW and VMHD have developed, tested and applied the density separation method for tephra grain extraction. VMHD performed the laboratory work. All authors contributed to the interpretation of the data and have contributed to the manuscript's initial and review processes.

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.

This paper was edited by Elisabeth Dietze and reviewed by one anonymous referee.