Finds documents with both search terms in any word order, permitting "n" words as a maximum distance between them. Best choose between 15 and 30 (e.g. NEAR(recruit, professionals, 20)).
Finds documents with the search term in word versions or composites. The asterisk * marks whether you wish them BEFORE, BEHIND, or BEFORE and BEHIND the search term (e.g. lightweight*, *lightweight, *lightweight*).
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence.
powered by
Select sections of text to find additional relevant content using AI-assisted search.
powered by
(Link opens in a new window)
Abstract
This study delves into the effectiveness of toe buttressing as a standalone technique for enhancing slope stability, focusing on a landslide in a German quarry. The research combines real-time monitoring and numerical modeling to assess the impact of toe buttressing without additional stabilization measures. Key findings include the significant role of contrasting lithological units in slope instability and the successful stabilization of a large landslide using a relatively modest toe buttress. The study also explores the broader implications of toe buttressing, comparing it to natural processes like glacier debuttressing. Additionally, the research highlights the need for further studies to establish comprehensive databases and transition to 3D modeling for a more detailed understanding of slope response to toe buttressing.
AI Generated
This summary of the content was generated with the help of AI.
Abstract
Slope instability poses significant challenges in geotechnical engineering, necessitating effective mitigation strategies. Among these strategies, toe buttressing has emerged as a prominent method, involving the strategic placement of materials at the base of unstable slopes. However, the efficacy of toe buttressing remains insufficiently explored in scientific literature, particularly through the integration of real-time monitoring and numerical modeling. This paper explores the effectiveness of toe buttressing as mitigation measure for slope stability without the influence of additional stabilization methods. Through a combination of field monitoring, laboratory testing, and the finite element analysis, we assess the influence of toe buttressing on slope dynamics and stability under varying environmental conditions. The study site, located in a basalt quarry in Nentershausen, Germany, experienced a landslide, with the sliding plane located at the interface between basalt formations and the underlying clay layers. Our monitoring results demonstrate that the placement of a 10 m thick toe buttress effectively halted the movement of a 295,000 m³ landslide. Additionally, we investigated the impact of different buttress configurations and groundwater levels on the factor of safety (FoS) using a 2D finite element model. The results clearly show the important role that toe buttresses play in strengthening slope stability. Our study emphasizes the need for more detailed case studies focusing solely on toe buttressing as a means of stabilization. Such research is essential to develop statistical models that can determine the necessary buttress volume and height to support/stabilize different landslide scenarios. Additionally, we discuss the broader implications of our study and future research for slope management applications and underscore the need for interdisciplinary approaches to address the complexities of slope stability in a changing environment.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
Slope deformation and collapse occur in both, natural slopes and engineered structures such as open pit mines or highways. Mitigating the hazards associated with slope instabilities and landslides is a concern in geotechnical engineering, with numerous measures implemented to prevent injuries and economic losses. Among these measures, toe buttressing is a known technique for enhancing slope stability (Ulusay et al. 1993; Abramson et al. 2001; Wyllie and Mah 2004; Akın 2017). This approach involves strategically placing rock material, often waste rock, at the toe of an unstable slope, thus modifying stress distribution and fortifying resistance against gravitational forces and external factors that may lead to slope failure. In open-pit mines, slope stability is a prevalent concern due to excavation processes, and hence a common challenge in the mining industry (e.g. Cambio et al. 2019; Ulusay et al. 1993). Furthermore, beyond mining applications, this mitigation method is also used in urban or coastal areas (Edil 1992; Işık et al. 2004). It is used when there is sufficient space available at the base of the slope and the waste material exhibits effective drainage characteristics (Wyllie and Mah 2004). This helps prevent the accumulation of water pressure between the buttress and the slope. The effectiveness of a buttress in providing support relies on two main factors: the weight of the buttress and its shear resistance (Wyllie and Mah 2004). The weight of the toe buttress itself plays an important role, as a heavier buttress exerts more downward force, potentially improving stability and its ability to resist sliding or movement along its base. This shear resistance is influenced by factors such as its weight, the roughness of the base surface, and its inclination or slope angle (Wyllie and Mah 2004). A rougher base with a greater inclination may offer more resistance against sliding, contributing to stability.
Although toe buttressing is widely recognized as a mitigation technique for slope stability, there remains a noticeable gap in scientific research demonstrating its effectiveness through a combination of real-time monitoring and numerical modeling. Prior investigations into toe buttressing often integrated supplementary stabilization methods such as anchors, unloading techniques, and drainage systems (Edil 1992; Işık et al. 2004; Corkum and Martin 2004; Macfarlane 2009; Oral et al. 2015; Alexandris and Abarioti 2016; Cambio et al. 2019). For instance, Macfarlane (2009) implemented toe buttresses in several landslide cases, but they also used other mitigation techniques such as pumped drainage, gravity drainage and/or infiltration protection, making it hard to assess the sole impact of toe buttressing. Several studies used limit equilibrium and finite element analyses to assess the impact of toe buttress (Ulusay et al. 1993; Işık et al. 2004; Oral et al. 2015; Moriya et al. 2016). However, their studies lack practical applications of the numerical results. The combination of various mitigation measures in previous studies makes it difficult to accurately assess the specific impact of toe buttressing on slope stability.
Advertisement
Slope stability monitoring is essential to understand slope behavior and for the early detection of potential failures. Monitoring techniques such as geodetic surveys, Terrestrial Laser Scanning (TLS), and InSAR (Interferometric Synthetic Aperture Radar) are common high-resolution, surface deformation methods to assess slope movements (Carnec and Delacourt 2000; Abellán et al. 2006; Osasan and Afeni 2010; Francioni et al. 2015; Chandarana et al. 2016; Cambio et al. 2019). Complementing these approaches, change detection methods, including digital elevation model (DEM) differencing and point cloud comparison, are also widely used to identify and quantify changes in slope morphology over time (Burns et al. 2010; Lague et al. 2013; Dai et al. 2020; Schaefer et al. 2023; Donati et al. 2023).
Here, we focus solely on evaluating the effectiveness of toe buttressing without any other stabilization measures. By combining “real-time” monitoring data and numerical back-analysis, we aim to understand the direct impact of the toe buttress as mitigation measure for slope stabilization. The study site is located in a quarry in the Westerwald region, Germany, where the implementation of toe buttressing was undertaken in response to a landslide during mining operations in February 2021. Data of a complete monitoring of the slope dynamics through geodetic measurements and TLS, covering the periods before, during and after the implementation of the toe buttress is the basis of our study. In addition, we carried out a back analysis using finite element software to assess the influence of water table levels and the presence of the toe buttress. Through a comprehensive analysis using field studies and remote sensing approaches, this research aims to advance our understanding of relationship between toe buttressing and slope stabilization processes, contributing valuable insights for both practical applications and theoretical frameworks in geotechnical engineering. Furthermore, this paper contributes to the broader discussion on slope stabilization, drawing connections between practical applications of toe buttressing and the complex interactions between natural processes, such as glacier debuttressing and slope stability in a changing climate.
Study area and landslide history
Geological context
The study area is situated within the Westerwald, a low mountain range east of the Rhine River in the federal state of Rhineland-Palatinate. The Westerwald is part of the Rheno-Hertynian zone of the Rhenish Slate Mountains. Throughout geological time, the following key eras and events shaped the area’s landscape that hosts today’s mining resources of the region.
The Devonian was characterized by large-scale tectonics of the southwest-northeast-striking Moselle syncline (LGB & LFU 2015). This massive unit consists of Lower Devonian sedimentary rocks, which are mostly an alternating sequence of clay shales and siltstones with intercalated sandstones to quartzites. The rocks are folded as a result of the Variscan mountain formation and in some cases are slightly metamorphosed. The mountain range has numerous faults and fault zones and is subject to intense tectonic stress. During the Mesozoic, intensive weathering of the Variscan Mountains was associated with kaolinitization and softening of the Palaeozoic sedimentary rocks. Thick saprolites were formed as a result.
Advertisement
In the Tertiary, an elongated subsidence zone — the Bitburg-Kassel subsidence field — developed in the strike direction of the Variscan and thus created intramontane basins in the Westerwald (LGB 2005). The weathered and softened parts of the Slate Mountains were washed into these basins in the form of clays and sands forming massive clay deposits during the Eocene to Oligocene periods. These are predominantly kaolinitic and illitic clays (sedimentary origin). Volcanic activity began in the Tertiary parallel to the development of the subsidence zone. The majority of the Tertiary volcanics are basaltic lavas that flowed into existing valleys and depressions. This is also the case at the Nentershausen basalt quarry. In addition to the lava flows, volcanic eruptions deposited tephras that weathered into so-called tuffitic clays (volcanic origin). The volcanic deposits mostly date from the Upper Oligocene to Lower Miocene (LGB 2005).
Lastly, the Quaternary cover consists of weathered units and Pleistocene solifluction soil, loess and Holocene alluvial sediments. Around 12,900 years ago, the Laacher See volcano (45 km W of the quarry) erupted and deposited widespread tephra up to 1 m thick in the study area (Fig. 2 in Baales et al. 2002).
In February 2021, the northern wall of the Nentershausen basalt quarry collapsed. The basalt is around 60–70 m thick at the location of the landslide and is underlain and flanked at the failure scarp by Tertiary sediments. This geometry shows that the basalt once flowed into a pre-existing depression. The sediments consist of clays and gravelly and sandy silts and, according to the geological map, belong to the Bubenheim Formation (Eocene) (LGB & LFU (2015), see Fig. 1).
Fig. 1
Geological map and localization map of the study area in Germany. The landslide (red oval) occurred at the northwest edge of the basalt quarry northwest of Nentershausen. The west-east width of the map corresponds to approximately 3.2 km in reality (Map: LGB & LFU 2015)
Operations at the Nentershausen quarry began in 1953 and primarily focused on the extraction of basalt for paving stone production. Presently, the annual extraction of basalt ranges between 340,000 and 500,000 tonnes, with an estimated remaining extraction time of ten years. Beginning in 2001, excavation progressed into the eastern part of the basalt deposits (Fig. 2a), where the landslide would fail 20 years later. Available aerial and satellite images reveal first signs of slope damage from 2019. The slope was stable up to an extraction height of around 60 m. In February 2021, part of the northern wall of the quarry failed as a rotational landslide with an offset of about 11 m (Fig. 2d). Cracks and fissures in the landslide were first noticed on 8 February 2021. Between 26 October 2021 and 18 February 2022, efforts to stop the slope movement were initiated with the implementation of toe buttressing with rock material from the quarry.
Fig. 2
Evolution of the quarry and the landslide slope through time with photographs (source: Rheinland-Pfalz Landesamt für Vermessung und Geobasisinformationen (WMS RP DOP40)) from (a) 2001, (b) 2007, (c) 2019, and (d) 2021
The landslide mobilized the entire available basalt mass, sliding along the interface with the underlying volcanic clays. It extends ~ 150 m in length and ~ 65 m in height, and its headscarp, marked by the yellow line in Fig. 3a and b, reaches a height of around 10 m. Throughout the upper section of the landslide, secondary scarps and minor cracks are visible (Fig. 3a, b, and c). There is an increasing appearance of scarps towards the frontal part of the landslide mass. In this frontal section, a distinct gap separates two landslide blocks at the edge of the slope (Fig. 3c). Figure 3a illustrates the distinct morphological sections or blocks of the landslide mass; denoted as (1), (2), and (3). The hexagonal basalt columns are oriented at roughly horizontally (Fig. 3d). Since basalt columns form normal to the cooling front, we see this as an additional indicator of the lava having flowed into a pre-existing topographic depression and that the contact to the paleo-valley walls is very close to today’s landslide front.
Fig. 3
Geomorphological overview of the landslide area. (a) geomorphological map illustrating the landslide mass, with AA’ representing the profile in Fig. 8. (b) close-up view of selected landslide scarps. (c) zoom showing a sink hole. (d) close-up view on the slope material composed of basalts
The volume of the landslide mass is around 295,000 m³. Our volume calculations are based on point clouds (PC) from high-resolution laser scans (PC02 for the landslide and PC01 and PC18 for the embankment); details of the acquisition and processing of these point clouds are explained in later sections .
The embankment or toe buttress from filling operations to stabilize the landslide extends for approximately 190 m in length, reaching heights up to 12 m. The toe buttress has an estimated volume of 64,000 m³ and was placed directly at the base of the slope, in direct contact with the basalt formation (Fig. 3a).
Weather data
Daily weather data are available from private stations at Großholbach and Kleinholbach, both located about 2 km away from the basalt quarry. We analyzed the data for 2021, 2022 and 2023 (see Fig. 4). On average, monthly temperatures fluctuate between − 5 °C in January to 23 °C in July. In the month preceding the landslide, significant precipitation was recorded, but no unusual precipitation was noted during the month of the landslide event itself (Fig. 4).
Fig. 4
Daily precipitation (in mm) and average daily air temperature (in °C) near the study area for 2021, 2022, and 2023. PC (Point Clouds from the TLS data) and GCP (Ground Control Points) data collection dates are indicated by the dot lines. The weather data for 2021 and 2022 come from the Großholbach station, while the data for 2023 is obtained from the Kleinholbach station
This research includes four main steps and follows the workflow summarized in Fig. 5. The steps include: geological model, slope monitoring, numerical modelling, and toe buttress interpretation and discussion. Step 1 involves the creation of a geological model using borehole data, which also includes laboratory testing results, and Electrical Resistivity Tomography (ERT). Step 2 focuses on slope monitoring through geodetic measurements and TLS, including data before, during, and after buttressing implementation. Step 3 involves numerical modeling to simulate the effects of the placement of the toe buttress under various conditions, including different buttress configurations and water table levels. Steps 1 and 2 feed directly into Step 3, with the geological model and laboratory testing providing information on the construction of the geotechnical model of slope and strength parameters, and slope monitoring data used to validate the numerical model results. Finally, Step 4 encompasses the interpretation and analysis of toe buttressing within the context of the case study, a literature review of similar cases, and a discussion of the broader implications of toe (de-)buttressing. This step also highlights potential avenues for future research.
In this section, we present the geological model of the landslide for which we combined data from three boreholes (two behind and one at the base of the landslide), laboratory testing, and ERT to deepen our understanding in the geological structure of the landslide. By integrating the ERT profile with borehole data and additional relevant literature information, we created a detailed profile showing the characteristics of the slide geometry in relation to the local stratigraphy.
Borehole data and laboratory testing
Three boreholes were drilled in the quarry in June 2021, i.e. after the landslide event (Fein 2022a). Two in the upper area behind the landslide (BH1 and BH2), reaching depths of 45 m and 36 m, respectively, and one at the toe of the landslide (BH3) reaching a depth of 6 m (Fig. 6). In the two boreholes located on the back of the landslide, tuffitic clays were drilled to a depth of around 44 m. The third borehole located at the base of the landslide includes the sedimentary clays. The volcanogenic clays most likely belong to the Breitscheid Formation (Upper Oligocene) while the sedimentary clays belong to the Bubenheim formation. After completion of the drilling process, the water table was measured in boreholes BH1 and BH2 at depths of 29.38 m and 28.73 m below surface, respectively.
Fig. 6
Spatial distribution of different methods employed in this study showing the location of: boreholes, ERT profile, geodetic measurements, and laser scans
A series of laboratory tests were conducted on the clay material extracted from boreholes by Fein (2022b), by the State Office for Geology and Mining Rhineland-Palatinate (LGB) and within the scope of this study. Shear strength parameters of the clays from the BH3 at 2 m depth (volcanic origin) using a direct shear test machine. To obtain reliable cohesion and internal friction angle data, the test has to be repeated three times on samples from the same core at the same depth, applying loads of 0.5 MPa, 1 MPa, and 1.5 MPa. The tuffitic clays exhibit a peak cohesion of 140.47 kPa, a residual cohesion of 2.26 kPa, a peak friction angle of 17.84°, and a residual friction angle of 13.24°. Soil mechanical laboratory tests determined that the sedimentary clays are slightly plastic clays (soil group TL) and silts while the volcanic clays are highly plastic (soil group TA) (Fein 2022b). The volcanogenic clays are at the level of the basalt and, in contrast to the sedimentary clays, have a significant smectite content and are distinctly plastic (soil group TA) (Fein 2022b). The results of the laboratory investigations conducted by the LGB in Table 1 for both clays from volcanic and sedimentary origin. The tuffitic clays (volcanic clays) exhibit significantly higher plasticity and water absorption capacity. This is due to the high content of swelling clay minerals. At high water contents, the tuffitic clays exhibit only very low shear strength. The sedimentary clays material characteristics have a significantly lower shear strength.
Table 1
Results of geotechnical laboratory tests on sample specimens (LGB)
Drilling
Depth (m)
Soil type
Dry density (g/cm3)
Water content (%)
Flow limit (%)
Roll-out limit
(%)
Shrinkage limit
(%)
Shrinkage dimension
(%)
Water absorption capacity (%)
BH 1
16.0–16.4
Volc. clay
1.12
54.8
81.4
44.1
15.6
58.1
126.0
BH 2
17.2–17.6
Volc. cay
1.22
44.0
83.1
38.1
15.2
60.3
118.0
BH 3
2.1–2.4
Sed. clay
-
17.0
32.1
13.8
13.5
33.9
79.8
BH 3
5.1–5.4
Sed. clay
1.79
16.7
30.7
13.4
15.9
31.2
54.8
* Volc. = volcanic, sed.= sedimentary
Electrical resistivity tomography data
ERT is a non-invasive geophysical technique to investigate subsurface structures by means of electrical properties. This technique images the electrical resistivity, i.e., the inherent ability of materials to impede the flow of electrical current, which depends on lithology, porosity and water saturation. The measurement principle is based on injecting electrical currents into the ground between two electrodes and measuring the resulting voltages between other pairs of electrodes. By means of tomographic inversion of all acquired four-point measurements, a 2D model of electrical resistivity, which describes the observed data within its error bounds, can be created (e.g., Binley and Slater 2020).
The ERT profile spans west to east at the back of the landslide (white line in Fig. 6) with 72 electrodes and a spacing of two meters resulting in a total length of the profile of 142 m. We processed and inverted the data using the open-source library pyGIMLi (Rücker et al. 2017).
The result of the ERT measurement after inversion is a 2D distribution of electrical resistivity, featuring the vertical axis representing depth and the horizontal axis depicting longitudinal extension, both in meters (Fig. 7). The relative root-mean-square error between the observed data and the model response is 2.69% reflecting good data quality. Within this profile, two prominent rock units are distinguishable: R1 and R2 (Fig. 7). The first unit (R1) exhibits a value range of approximately 5–20 Ωm. It can hence be attributed to the clay as its resistivity values range from < 10 Ωm to 20 Ωm; clay-rich sediments are typically characterized by low electrical resistivities due to interfacial conduction by clay minerals. The electrical resistivity tomogram predicts a dipping transition from R1 to an underlying strata (R2). According to the borehole data, the depth of this interface coincides with the transition of volcanic clay to sedimentary clay. Although clay is usually more conductive than the values in the tomogram exceeding 100 Ωm, quantitative interpretation of resistivity in this region is not opportune, due to the loss of cumulative sensitivity as depicted by the transparency in Fig. 7. The surrounding subsoil falls within the range of 60–80 Ωm.
Fig. 7
2D electrical resistivity tomography profile designated EE’, indicating the position of the profile as illustrated in Fig. 6. Note that the color bar has a logarithmic scale since electrical resistivity in the shallow subsurface typically spans several orders of magnitude. The colors are faded out with decreasing cumulative sensitivity to avoid misinterpretation in regions not covered by the data
In summary, the borehole and ERT data reveal two distinct layers of clay: one from volcanic origin and most likely belong to the Breitscheid Formation (Upper Oligocene) and the other from sedimentary origin associated to the Bubenheim Formation as in the geological map (Fig. 1). The landslide mass is situated within the basalt layer, with the sliding occurring along the interface between the soft clay layer and the basalt formation. Considering the rotational nature of the slide, estimating the geometry of the sliding plane entails interpolating both its orientation and angle. These were deduced from structural measurements obtained remotely on the point clouds via CloudCompare software, focusing on visible scarps. Additionally, insights from literature research on rotational landslides (e.g. Rickli et al. 2003; Souisa et al. 2020; Yoshizawa 1992), alongside the analysis of borehole data and ERT profiles, helped in determining the lower angle of the sliding plane (Fig. 8). The material constituting the toe buttress covers the daylighting plane (Fig. 8) and has effectively stabilized the slope up to the present day (cf. next section).
Fig. 8
Geological cross section through the landslide body with extrapolation of the sliding plane based on field observation, literature, ERT, and boreholes (BH). The localization of the profile A-A’ can be seen on Fig. 3
In order to assess the behavior of slope movement we carried out an in-depth analysis of geodetic measurements and TLS. By examining geodetic data collected from points strategically positioned at both the toe and on the top of the landslide, as well as multiple laser scans capturing the evolution of the movement of the slope, we gained insights into the behavior and response of the landslide to toe buttressing.
Geodetic measurements
Geodetic data processing
The geodetic data were collected using total stations. Four ground-control points (GCP) are located at the toe of the landslide (GCP 1 to 4, Fig. 6) and four GCPs were established on the landslide body (GCP 5 to 8, Fig. 6). Measurements of the GCPs located at the toe of the landslide were collected from mid-April 2021 for about every two weeks until mid-November 2021 (Table 2). These measurements were subsequently terminated as the GCPs were obscured by debris from the construction of the toe buttress. The GCPs located on the landslide body were installed, for safety reasons, after the start of the toe buttress. They were measured approximately every two weeks from 16 November 2021 until 5 August 2022 (Table 2), for the exception of GCP 8 that was lost in early April 2022. After this period, the data of the remaining GCP was collected once a month until 3 November 2023.
Table 2
GCP survey dates with the indication of the time frame during the data acquisition: pre-failure, pre-toe buttress, during toe buttress placement and post- toe buttress
GCP 1, 2, 3, 4
GCP 5, 6, 7, 8
Dates
Nb*
Dates
Nb
Dates
Nb
Dates
Nb
Pre-toe buttress
16/04/2021
-
Pre-toe buttress
02/07/2021
7
Toe buttress
Post-toe buttress
25/07/2022
17
23/04/2021
7
09/07/2021
7
26/11/2021
10
05/08/2022
11
30/04/2021
7
16/07/2021
7
10/12/2021
14
02/09/2022
28
07/05/2021
7
23/07/2021
7
22/12/2021
12
07/10/2022
35
14/05/2021
7
30/07/2021
7
06/01/2022
15
04/11/2022
28
21/05/2021
7
06/08/2021
7
21/01/2022
15
02/12/2022
28
28/05/2021
7
12/08/2021
6
04/02/2022
14
06/01/2023
35
04/06/2021
7
19/08/2021
7
18/02/2022
14
03/02/2023
28
10/06/2021
6
27/08/2021
8
Post-toe buttress
04/03/2022
14
03/03/2023
28
18/06/2021
8
02/09/2021
6
18/03/2022
14
06/04/2023
34
25/06/2021
7
17/09/2021
15
01/04/2022
14
03/05/2023
27
29/09/2021
12
19/04/2022
18
02/06/2023
30
15/10/2021
16
29/04/2022
10
07/07/2023
35
29/10/2021
14
13/05/2022
14
03/08/2023
27
16/11/2021
18
27/05/2022
14
01/09/2023
29
13/06/2022
17
06/10/2023
35
24/06/2022
11
03/11/2023
28
08/07/2022
14
* Nb = number of days since the previous measurement
Geodetic data analysis
The GCP located at the toe of the landslide showed consistent movement and exhibited displacements of up to 1.4 m (Fig. 9). These displacements suggest a noticeable uplift of the ground in those areas. On the landslide mass, GCP 5, 6, 7, and 8 show notable differences (Figs. 9 and 10). GCP 8 in particular showed significantly higher displacements than the others, marked by a major movement from the end of December 2021 to mid-January 2022, followed by a brief activation of movement in mid-February 2022, a subsequent stabilization, and ultimately, the loss of the point. The first phase of deformation at GCP 8 was also discernible, although with a lower intensity, in GCP 5, and only slightly in GCP 6 and 7 (Fig. 10). During that time, precipitation exceeded 10 mm per day on December 28, 2021, and on January 4 and 8, 2022 (Fig. 4). Both GCP 5 and 6 also exhibited an increase in movement in mid-February 2022, as observed for GCP 8, while GCP 7 displayed a decrease in velocity (Fig. 10). Between February and March 2022, four days registered precipitation levels above 10 mm (Fig. 4). An increase in movement is observable in November 2021 for GCP 5, 6, and 7, but not in GCP 8. During the November 2021 phase of movement, precipitation records indicate minimal rainfall (Fig. 4). Following March 2022, a pattern of alternating increases and decreases in velocity is observed for GCP 5, 6, and 7. From August 2022 onward, it appears that the points have stabilized somewhat, with displacements ranging between 0.01 and 0.03 m (Fig. 10).
Fig. 9
Summary of the displacements from the GCPs located at the toe of the slope (purple arrows) and on top of the landslide mass (yellowish arrows), the arrows are exaggerated 20 times for visualization
- Graph illustrating the cumulative displacement (m) and the velocity (cm/day) during the investigation period for the Ground Control points situated at the toe of the slope (GCP 1, 2, 3, 4) and on top of the landslide (GCP 5, 6, 7, 8)
Figure 11 presents the evolution of the dip direction and dip angle of the GCPs. All GCPs show that the average component of movement is more robust in the vertical direction than in the N-S or E-W directions, with the exception of GCP 3 and GCP 8. The GCPs at the toe of the landslide all show a SW movement direction. We observe a gradual shift in the dip angle for GCP 1 and 4, transitioning from 40° to 19° and 53° to 38°, respectively (Fig. 11). On top of the landslide, GCP 6 and 8 show a shift in both dip direction and dip angle (Fig. 11). Over the period between November 16, 2021 and March 4, 2022, GCP 6 transitions from a SW direction with an average dip direction of 237° and a dip angle of 48.2° to a SE direction with a dip direction of 135.5° and a dip angle of 33° (Fig. 11). GCP 8 underwent a directional change starting from January 6, 2022. It transitioned from a SW direction with a dip direction of 266.5° and a dip angle of 30.6° to a SE direction with a dip direction of 108.2° with the average dip angle remains similar of about 31° (Fig. 11).
Fig. 11
Graph showing variations in dip and dip direction of the GCPs movement. The first graph illustrates the results for GCP situated at the toe of the landslide, while the second graph presents data for GCP located on the landslide mass
For the data acquisition, we used the Teledyne Optech Polaris TLS and collected laser scan data between 16 February 2021 and 30 May 2023. This period overlaps with scans collected using a Riegl VZ400 TLS (PC1, PC2, PC3, and PC16), resulting in a total of 18 data sets (Table 3), each corresponding to a single day. Within each dataset, we have captured up to five scans from various locations in the quarry to enhance the 3D characteristics of the resulting point cloud (Fig. 6). The data timeline is composed three part: pre-toe-buttress (PC01 and PC03), toe buttress phase (PC04 to PC12), and post-toe-buttress (PC13 to PC18), with the latter three corresponding to the post-failure phase.
Table 3
TLS survey dates with the indication of the time frame during the acquisition of the data: pre-toe buttress, during toe buttress placement and post- toe buttress
Name
Dates
Number of scans
Number of days since the previous scan
Pre-toe buttress
PC01
16.02.2021
5
-
PC02
30.04.2021
5
73
PC03
25.06.2021
5
56
Toe buttress
PC04
24.11.2021
1
152
PC05
01.12.2021
1
7
PC06
09.12.2021
3
8
PC07
14.12.2021
2
5
PC08
21.12.2021
2
7
PC09
13.01.2022
2
23
PC10
20.01.2022
2
7
PC11
27.01.2022
2
7
PC12
10.02.2022
2
14
Post-toe buttress
PC13
03.03.2022
2
21
PC14
29.03.2022
5
26
PC15
13.04.2022
3
15
PC16
24.04.2023
5
376
PC17
03.05.2023
5
9
PC18
30.05.2023
5
27
For the data processing, we used the ATLAScan (Teledyne 2022) and CloudCompare (CloudCompare 2023) software. The scans acquired on the same day from various location were co-registered, aligned, and merged together to create a unified 3D point cloud (PC) of the slope. The registration was done through iterative closest point (ICP) algorithms. Raw point clouds often contain noise, outliers and artefacts, with the vegetation being a predominant contributor among these undesirable points. To remove the vegetation, we used the plugin CANUPO (Brodu and Lague 2012) in CloudCompare. We trained the classifier through the segmentation of point clouds, categorizing points into those associated with vegetation and those corresponding to the rockslope. Although the CANUPO plugin produced very satisfactory results, a number of unwanted points remained, which we meticulously removed manually. To compare two point clouds from different days, it is essential that these two are co-registered. Co-registration begins with segmenting the point clouds by isolating the stable part of the quarry within each dataset. Subsequently, these stable segments are registered using ICP algorithms in CloudCompare, resulting in a precise alignment characterized by a transformation matrix. This transformation matrix is then uniformly applied to the entire/unsegmented point cloud, ensuring coherent and aligned results. The final root mean square (RMS) error varies from 0.030 m to 0.044 m.
Change detection of the slope was then carried out using the Multiscale Model to Model Cloud Comparison (M3C2) plugin (Lague et al. 2013) in CloudCompare. The M3C2 plugin is a powerful tool designed to achieve a detailed and accurate comparisons between two point clouds. The algorithm’s core concept involves the generation of cylinders intersecting the reference and compared clouds along the normal direction. The size and orientation of these cylinders are determined by the normal scale and projection scale, which are either defined by the operator or estimated by the algorithm. Following the suggestions of Lague et al. (2013), we opted for a normal scale of 1.55, a projection diameter of 0.62, and a search depth of 12. We also included the registration error for each comparison. To get a full time series analysis of the slope changes, we compared the first point cloud (PC01) with the second point cloud (PC02), then PC02 with PC03, PC03 with PC04, …, and PC17 with PC18.
The volume calculation of the landslide is based on the scan PC02 while the volume of the embankment was determined based on the scan PC01 and PC18. To determine the volume of the landslide mass, we used the volume between the fitted planes (corresponding to the sliding planes) and the slope mass of the PC02. For this computation, we used the 2.5 volume tool in CloudCompare. The fitting planes include an upper plane estimated at an angle of 71° based on measurements of the apparent upper-scarp, while the lower plane, intended to simulate the rotational geometry, was estimated at approximately 14° from the literature and field-data analysis (as explained in Sect. Geological interpretation). Additionally, for calculating the volume of the toe buttress, we segmented scans PC02 and PC18 to isolate the buttress area and used the 2.5 volume tool once more to calculate the volume between these two point clouds.
TLS data analysis
The change detection results show movement patterns and trends of acceleration and deceleration as observed in the geodetic measurements. Figure 12 illustrates three cross-sections depicting the 2D movement of the slope between 16 February 2021 and 30 May 2023. The figure also shows the evolution of movement for selected points within the point clouds. Additional, detailed results of the TLS data processed with the M3C2 plugin are available in the supplementary material (Figs. S1, S2, and S3). Movements below the threshold of 3–4 cm need to be interpreted with caution, because they fall within the range of potential co-registration errors.
Fig. 12
Slope movement overview from TLS point clouds. (a) General map indicating the locations of profiles AA’, BB’, and CC’, along with reference points 1, 2, 3, 4, 5, and 6. (b), (c), (d) Profiles AA’, BB’, and CC’, respectively, illustrating slope morphology/movement on various dates. (e) Graph showing cumulative displacements of specific points in the landslide mass, derived from M3C2 results
Geometry of the finite element model of the slope with the different geological materials, level of water table (W1 to W6) and the heights of toe buttress (B1 to B6 – 2 m to 12 m)
Graph illustrating the outcomes of the sensitivity analysis from the numerical modeling and showing the relationship between different ground water levels (x-axis) (see levels in Figure 13), Strength Reduction Factor (y-axis), and the different heights of toe buttresses from 0 m to 18 m. Each buttress is labeled from 1 to 8, with corresponding heights: Buttress 1 (2 m), Buttress 2 (4 m), Buttress 3 (6 m), Buttress 4 (8m), Buttress 5 (10 m), Buttress 6 (12 m)
Overview of the landslide failure with (a) results from the 2D numerical model for the case of no buttress and with the water table level 3 (W3 - black line) showing the maximum shear strain and the vector displacements (red vectors). (b) view from west to east in the quarry showing the uplifted ground by approximately 1.5 m (red oval) as a result of the rotational movement (red arrow) (Photo 29.04.2021 A. Wehinger)
The profiles AA’ and BB’ (Fig. 12b and c) show a combination of vertical and horizontal movements, without significant deformation, indicating that the slope is primarily sliding rather than undergoing major internal deformation/collapse. Profile CC’ (Fig. 12d) also displays a combination of vertical and horizontal movements, but the lower portion of the slope appears to be deformed showing the ongoing internal deformation above the buttress height. The three profiles provide evidence supporting the rotational sliding behavior observed in the slope.
The largest displacements are observed between the first two TLS acquisitions PC01 and PC02, revealing movements up to 12 m (Fig. S1a) with a downward slope displacement of around 5 m. The change detection analysis comparing PC02 with PC03 (Fig. S1b) and PC03 with PC04 (Fig. S1c) also show continued large movements across the entire landslide mass. During this period (PC03 to PC04), toe buttress filling began, reaching heights from 3 m to 7.5 m. Between PC04 and PC05, there is a significant decrease in slope movement, with only minor rockfalls, and an additional increase in toe buttress height with heights reaching between 8 and 11 m (Fig. S1). The following comparison, between PC05 and PC06, indicates displacements up to 3 cm in certain parts of the slope along with some rockfalls (Fig. S1e), likely due to internal slope adjustments. Subsequent analyses (Fig. S1f through Fig. S3q) show minimal to no movement. Exceptions include comparisons between PC08 and PC09 (23-day timespan - Fig. S2h) and between PC15 and PC16 (376-day timespan - Fig. S3o), which showed several rockfalls and slope deformations of about 5 cm.
We selected six specific points/areas for monitoring displacements and temporal analysis (Fig. 12): Points 1–4 on the landslide body and Points 5 and 6 on the toe buttress (Fig. 12a and e). We chose points within areas unaffected by rockfalls as based on the M3C2 models from Figs. S1, S2, and S3. The displacements presented in Fig. 12e are derived from the M3C2 analysis results. However, due to a significant change between PC01 and PC02, rendering M3C2 results unrepresentative in some areas, the point between PC01 and PC02 was manually selected, and we calculated the distance between the two point clouds using the distance tool of CloudCompare instead of M3C2 results. Figure 12e shows that Point 1 was the most significantly impacted by the initial movement between PC01 and PC02, with all points on the landslide mass exhibiting a similar movement pattern. Points 5 and 6 show the gradual filling process of the toe buttress, and the slight negative displacements observed for these points are attributed to the settling of the rock material. During the buttress placement, we observe some small variations in movement behavior across the different points located on the landslide mass. However, following the buttress installation, all the points show a consistent trend of significantly reduced movement. These results demonstrate the direct effect of the toe buttress on the slope movement, particularly in minimizing large-scale displacements and controlling rotational sliding with the overburden of the buttress on the uplifted area in front the slope (Fig. 12b, c, d).
Limitations of the change detection analysis
Our study showed certain limitations in the change detection analysis based on the M3C2 plug-in of CloudCompare. While the M3C2 analysis proves effective for detecting small movements and rockfalls, it encounters challenges in capturing larger-scale movements exceeding a few meters and even more when involving complex displacements in multiple directions. This limitation can be seen in the changes between the PC1 and PC2 (Fig. S1a), where the analysis does not fully capture the overall downward shift of the slope (by ~ 5 m), as well as its horizontal rotation. The M3C2 plug-in relies on a cylindrical model with a specified preferred orientation, which makes it more sensitive to movements in certain directions while less responsive to others. Additionally, the software cannot generate a composite representation of the x- y- z- directions of point movements, limiting the detailed understanding of the displacement patterns. However, our study was able to overcome these limitations by complementing the M3C2 analysis with geodetic data. The geodetic data provided a detailed perspective on the 3D direction of the slope movements, allowing us to discern patterns that might have been overlooked by the M3C2 analysis alone. Nevertheless, the M3C2 analysis proved highly valuable in capturing the uniformity or non-uniformity of the overall slope movement. By combining these datasets, we achieved an overall and multi-dimensional analysis, effectively overcoming the limitations of M3C2 and enhancing our understanding on the slope deformation patterns.
Assessment of the effectiveness of toe buttressing through numerical modelling
Through 2D numerical modeling and simplified assumptions, we aimed to quantitatively assess the impact of toe buttressing. In this section, we focus on the back-analysis of slope stability response to the implementation of toe buttressing. The simulations were conducted with different toe buttressing heights and different groundwater levels in order to evaluate the factors affecting the stability of the slope.
Modelling method and input parameters
Numerical modelling based on the finite element method (FEM) using RS2 (Rocscience 2024) was carried out to quantify the impact of toe buttressing on the stability of the slope. The FEM method divides the modelled domain into smaller elements (e.g., triangles or quadrilaterals) to approximate the force and displacement by solving the governed equations. Within RS2, the shear strength reduction (SSR) technique is utilized to calculate the factor of safety (FoS) of the slope. The simplified geological geometry, derived from the results presented in Fig. 8. In the model, we implemented six successive toe-buttress heights, ranging from 2 m (level B1) to 12 m (level B6) (Fig. 13). Additionally, we investigated the influence of precipitation on slope stability by introducing six levels of water table and conducting tests ranging from dry conditions to fully saturated conditions (Fig. 13). The water table W3, corresponds to the approximate height in which the ground water table was measure in the boreholes in June 2021. We assume that with high precipitation the groundwater level increases.
Fig. 16
Overview of basalt column behavior in the unstable slope. (a) Photograph capturing the material of the unstable slope. (b) Close-up view of a higher-located basalt column displaying a compact morphology. (c) Close-up view of a lower-located basalt column exhibiting an uncompacted morphology
Laboratory results combined with literature research and sensitivity analyses were performed to evaluate the influence of key parameters on the observed outcomes. For the three different materials (clay (volcanic and sedimentary origin), basalt (rock mass), and buttress debris), we adopt plastic properties with an isotropic nature following the Mohr-Coulomb failure criterion. The clay parameters were acquired from the laboratory test results and from a geotechnical report (Table 4), which studied clays belonging to the same geological formation, Bubenheim formation, as the sedimentary clay in the quarry. Parameters for basalts come from studies by Schultz (1995) and Di et al. (2011), both of whom investigated the rock mass of columnar basalts (Table 4). The toe buttress is assumed to have a unit weight of 19 kN/m3, based on the use of crushed basalt material placed during the buttress construction.
Table 4
Material properties of the clays (volcanic and sedimentary) and basalt (rock mass) used as input parameters for the numerical modelling
Material
Unit weight (kN/m3)
Poisson’s ratio
Young’s Modulus (MPa)
Tensile strength (kPa)
Friction angle (°)
Cohesion
(kPa)
Clay (volcanic)
11
0.3
12
140.5
17.8
140.5
Clay (sedimentary)
17.5
0.3
12
8
23
8
Basalt (rock mass)
27
0.4
1e + 4
750
44
1145
Model results
The results are presented in the form of the SRF (equivalent to the “safety factor” of the slope) in Fig. 14. These values act as indicators of slope stability, highlighting its sensitivity to changes in water levels and toe buttress heights. Under dry conditions and without any buttress, the slope remains stable (with SRF above 1). With increasing water table levels, we observe a gradual decrease in SRF for the initial condition (no buttress) and each subsequent buttress level. The introduction of toe buttressing noticeably impacts the results, showing a visible trend where the increase of toe buttress heights enhances slope stability. Upon introducing the toe buttress, an immediate increase in SRF is observed. With a toe buttress height of 2 m (level B2), stability is achieved for water table level W2. Similarly, at a toe buttress height of 4 m (level B2), stability is attained for water table level W4. This trend continues, with the slope reaching stability even at higher water table levels as the toe buttress height increases (Fig. 14). Ultimately, a toe buttress height of 12 m (level B6) is required to stabilize the slope under fully saturated conditions (water table level W6).
The results of the numerical modelling show that the sliding plane is located at the interface between the clay and basalts, with a rotational form (Fig. 15a), and thus match with our conclusions from field observations. The maximum shear strain is concentrated at the boundary between these contrasting geological layers, highlighting this interface as a critical zone of weakness which is likely due to significant differences in material properties such as strength, stiffness, and cohesion and making this slope more susceptible to failure. We observed that the approximately 20 m of the sedimentary clay layer beneath the basalt mass is also affected by deformation as it shows shear strain values between 1.3 and 4. We did not observe any internal deformation within the landslide mass as it was treated as a single cohesive unit in the model. Although the basalt columns composing the landslide mass are separated by columnar joints in reality, which could lead to internal deformation not reflected in these models, this simplification is not a significant concern as our aim is to assess the impact of the toe buttress on overall slope movement. The uplift observed at the bottom of the slope, as identified in the field (Fig. 15b) and through geodetic data, is also shown by the displacement vectors featured in the numerical model (Fig. 15a). This geometry of rotational sliding is very similar to the slope movement shown in the cross sections displayed in Fig. 12. The sliding geometry remains consistent across all models, regardless of variations in water table levels and toe buttress heights.
Impact of toe buttressing in the study case and the literature
The initiation of slope movement in the Nentershausen quarry in February 2021 likely resulted from prolonged rainfall in the preceding month. The sliding plane is located at the interface between contrasting lithological units: the basalt formation and the adjacent and underlying clay layer (Fig. 8) of the paleo-valley. While the basalt formation is a massive unit with relatively high rock-mass strength, it rests upon the comparatively weak clay layer. The clay layer experienced reduced shear strength due to prolonged rainfall, facilitating the downward movement of the overlying basalt layer. The distinctive rotational characteristics of the landslide are evident in its morphology, the behavior of the GCPs, and the results of numerical modeling. Both the GCPs and TLS data show comparable movement patterns with the same trends of acceleration and deceleration. These data also indicate that the slope underwent non-uniform movement, characterized by distinct sliding blocks (see Fig. 3a).
Monitoring and numerical modelling show the effectiveness of the toe buttress in the case of this landslide. The additional value of this case study is that no other mitigation measure was used and hence all stabilizing effects are due to the buttress alone. For the most unfavorable scenario with a fully saturated slope, our model shows that the height of the toe buttress needs to be at least 12 m. However, continued slope movement after toe buttressing does exist and we attribute this to a combination of rainfall events, settlement of the slope, and the relative movement between individual basalt columns. The latter is documented in TLS data showing recurring rockfalls (Figs. S1b, c, d and e and S2h, and Fig. S3o), which are concentrated at the front of the slope, there is a significant accumulation of debris composed of partially broken basalt columns (Fig. 16a). Field observations further reveal distinct patterns in different areas of the slope (Fig. 16). The weight of the slope, combined with rotational sliding, exerts pressure on basalt columns. The joints along the hexagonal shape of the columns act as discontinuities or weakened planes, facilitating the movement of individual basalt columns (Fig. 16b and c). Consequently, individual basalt columns are displaced/pushed outward from the landslide body and rockfalls result.
The toe of a slope and the environment and material at its toe plays an essential role in ensuring its stability. Analyzing statistics regarding the amount of toe buttress required for stabilizing an unstable slope of a specific volume is important for gaining insights into this mitigation method. Our literature review identified 56 case studies, summarized in Tables S1, S2, and S3 in supplementary material. Among these, 18 studies provided data on landslide volume, and 7 on the height of the failed slope (Table 5). For the toe buttress, 17 studies reported volume data, while 20 provided information on buttress height (Table 5). From this limited data, we were able to calculate 11 ratios of landslide to buttress volume, ranging from 3 to 300 (Table 5). The ratio of landslide to buttress height could only be calculated for 4 cases, with values ranging from 1.7 to 6.5 (Table 5). However, many of these cases involved additional mitigation measures (e.g., anchors, drainage), which compromise the statistics by not isolating the impact of toe buttressing alone. As a result, we are left with only one reliable ratio for both landslide-to-buttress volume and height (Table 5). The challenge of compiling such statistics is bounded by the limited data available, including volume, height information on both landslide and buttress and details about the failure mechanisms, as well as by the lack of studies that focus exclusively on the impact of the buttress without the influence of other stabilization techniques. In the case of the quarry in Nentershausen, approximately 64,000 m³ of material was placed at the toe of a ~ 295,000 m³ unstable slope. Our finite element analyses and monitoring observations showed that a toe buttress with a height of 10 m is sufficient to stabilize the slope with a height of about 65 m and characterized by a rotational failure mechanism, representing a ratio of 6.5.
Table 5
Summary of available information on landslide and toe buttress volumes and heights from the literature review (see Table S1, Tables S2, and Table S3) and the calculated ratio
Impact of toe (de-)buttressing in a broader geological context
In contrast to the stabilizing effect of toe buttressing, examples where the removal of the toe of a slope caused failure highlight the susceptibility of slopes to destabilization when their toe support is compromised. Prolonged erosion along the base of a slope, compounded by human activities such as excavation or deforestation, gradually eroded the toe of the slope and eventually leading to a landslide (Stark et al. 2005; Runyan and D’Odorico 2014; Cambio et al. 2019; Xu et al. 2022). Similar scenarios can also arise due to natural processes, such as river incision or coastal erosion, which gradually undermine the integrity of the toe of the slope, predisposing slopes to instability (Edil 1992; Işık et al. 2004; Osten et al. 2020; Donati et al. 2022).
The concept of toe buttressing in slope stability analysis parallels the important role of glacier (de-)buttressing in mountainous regions. Just as toe buttressing reinforces the base of unstable slopes, glaciers act as natural buttresses, supporting adjacent slopes and restraining their movement. As glaciers retreat and thin, the consequential impact on slope stability becomes a critical concern (Evans and Clague 1994; Kos et al. 2016; Coe et al. 2018; Higman et al. 2018; Lemaire et al. 2024). The effectiveness of a glacier-buttress at influencing rock slope stability has not been fully clarified yet (e.g. McColl and Davies 2012). Some research suggests that as glaciers retreat, the withdrawal of their supporting influence may lead to increased instability and slope failures in glaciated alpine regions (Cossart et al. 2008; Jaboyedoff et al. 2012; Deline et al. 2015). In contrast, other studies propose that glacial debuttressing alone may have a limited effect on slope stability (McColl et al. 2010; McColl and Davies 2012; Grämiger et al. 2017). McColl et al. (2010) suggest that the characteristics of ice, and by extension temperate glaciers, may not provide the desired stabilizing effect on rock slopes, questioning their efficacy as natural slope buttresses due to their ductile behavior. Instead, they emphasize the greater importance of fluctuating groundwater levels and climatic variations in influencing slope stability. Studying toe buttressing serves as a valuable analog for understanding the broader implications of glacier debuttressing. Here, through our investigation of toe buttressing in the Nentershausen quarry and literature review, we have demonstrated that the presence or absence of certain materials at the toe of a slope can have significant implications for slope stability. For instance, the placement of only rock debris at the toe of a slope has been shown to stabilize the slope (in this studied case), while the erosion of soft material at the toe of a slope can destabilize it (Edil 1992; Işık et al. 2004; Osten et al. 2020; Donati et al. 2022). We showed that placing only debris material at the toe of the slope and on top of the daylighting failure area can effectively halt slope movement (also Osten et al. (2020)). The way in which rock debris simply placed at the toe of a slope and on the sliding plane can stop its movement (as our study shows) can be compared to the way in which glaciers, with their weight and weight, cover sliding planes and can potentially halt the movement of a slope. The interplay between glacier debuttressing and unloading and the implementation of stabilization measures such as toe buttressing presents a complex dynamic that requires interdisciplinary research efforts to enhance our ability to address the evolving challenges posed by changing environmental conditions.
Future research perspectives
Our study demonstrates that placing rock waste at the base of a moving slope effectively halts its movement. However, we also show that there are still gaps in available data related to the effectiveness of toe buttressing as a stabilization technique (see previous sections and Tables S1, S2, and S3). Studies are either lacking available detailed information about the landslide and related toe buttress or/and implemented addition mitigation measures, making it difficult to isolate the impact of toe buttressing alone. Moreover, failure geometry most likely influences the toe-buttress effectiveness. For example, toe buttress configurations may vary depending on the failure mechanism, such as planar sliding, rotational sliding with daylighting on steep slopes, or rotational sliding with upheaval in the flat area in front of the slope. While we have shown that toe buttressing is effective for rotational slides involving toe deformation, there is a need for comparative studies to explore its direct effectiveness across other types of slope failures, both in monitoring cases and potential numerical modelling. By building a more comprehensive database that includes detailed information on factors such as landslide volume, buttress dimensions, and sliding mechanisms, it would be possible to develop accurate statistical models or mathematical relationships between landslide volume and toe buttress volume and height. This quantifying relationship could not only be useful for optimizing slope stabilization designs but also for gaining a better understanding of the effectiveness of buttressing effect in analogous natural processes, such as river erosion, or in opposite phenomena like glacier debuttressing.
Conclusions
Our investigation aimed to assess the efficacy of toe buttressing as a standalone mitigation measure for slope stability. Previous research often combined toe buttressing alongside with other methods (Edil 1992; Işık et al. 2004; Macfarlane 2009; Oral et al. 2015; Alexandris and Abarioti 2016; Cambio et al. 2019), making it difficult to isolate its specific impact. To address this gap, we focused only on toe buttressing by integrating both real-time monitoring on a landslide mitigated only with toe buttressing and numerical modelling. For the studied landslide, we observed that the juxtaposition of contrasting lithological units between basalt formations and underlying/adjacent clay layers, significantly influenced the slope stability. Periods of prolonged rainfall weakened the shear strength of the clay layer, facilitating the downward movement of the overlying basalt layer and initiating slope instability. Our monitoring and modelling results have shown that the implementation of a relatively modest toe buttress of 10 m thickness is capable of halting the movement of a 295,000 m³ landslide. The case study of the Nentershausen quarry stands out as one of the few comprehensive examples with detailed documentation on both the landslide and toe buttressing available in the literature, highlighting the importance of investigating toe buttressing as the sole measure for slope stabilization across various geological, morphological, and sliding mechanism contexts. While our numerical modeling offered detailed insights into the effectiveness of the toe buttress, its limitation to a 2D representation may not capture the full complexity of real-world scenarios. Moving forward, further research is needed to validate and expand upon our findings, especially by establishing a comprehensive database from which statistical analyses can derive optimal ratios between buttress size and the volume or height of instability. Moreover, transitioning from 2D to 3D numerical modeling could provide a more detailed understanding of slope response to toe buttress. This effort of research would not only improve our understanding of slope stabilization in response to toe buttress in a mining context, but could also holds potential to provide valuable information applicable for the understanding of natural buttressing mechanisms, such as glacier debuttressing in mountainous regions.
Acknowledgements
The authors would like to thank Jakob Bach GmbH & Co. KG for providing access to their valuable data as well as their support. We also thank Phillip Voicu and Hannes Beier for their assistance with field data collection, and Lena Oeser for her support in the laboratory testing.
Declarations
Competing interests
The authors declare that they have no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Assessing the effectiveness of toe buttressing on slope stability through monitoring and numerical modelling in a quarry in Germany
Authors
Emilie Lemaire
Lisa Fast
Anja Dufresne
Pooya Hamdi
Ansgar Wehinger
Florian M. Wagner
Frieder Enzmann
Wolfgang Fein
Teemu Hagge-Kubat
Gerd Mathes
Stefan Weber
Florian Amann
Abellán A, Vilaplana JM, Martínez J (2006) Application of a long-range terrestrial laser scanner to a detailed rockfall study at Vall de Núria (Eastern Pyrenees, Spain). Eng Geol 88:136–148. https://doi.org/10.1016/j.enggeo.2006.09.012CrossRef
Abramson LW, Lee TS, Sharma S, Boyce GM (2001) Slope Stability and Stabilization Methods. John Wiley & Sons
Akın M (2017) Reliability of shear strength parameters for a safe slope design in highly jointed rock mass. In: Mikoš M, Arbanas Ž, Yin Y, Sassa K (eds) Advancing culture of living with landslides. Springer International Publishing, Cham, pp 445–453CrossRef
Baales M, Jöris O, Street M et al (2002) Impact of the Late Glacial Eruption of the Laacher See Volcano, Central Rhineland, Germany. Quat Res 58:273–288. https://doi.org/10.1006/qres.2002.2379CrossRef
Binley A, Slater L (2020) Resistivity and Induced Polarization: Theory and Applications to the Near-Surface Earth, 1st edn. Cambridge University PressCrossRef
Brodu N, Lague D (2012) 3D terrestrial lidar data classification of complex natural scenes using a multi-scale dimensionality criterion: applications in geomorphology. ISPRS J Photogramm Remote Sens 68:121–134. https://doi.org/10.1016/j.isprsjprs.2012.01.006CrossRef
Burns WJ, Coe JA, Kaya BS, Ma L (2010) Analysis of elevation changes detected from Multi-Temporal lidar surveys in forested landslide terrain in Western Oregon. Environ Eng Geosci 16:315–341. https://doi.org/10.2113/gseegeosci.16.4.315CrossRef
Cambio D, Hicks D, Moffitt K et al (2019) Back-analysis of the Bingham Canyon South Wall: a quasi-static complex slope movement mechanism. Rock Mech Rock Eng. https://doi.org/10.1007/s00603-019-01958-7CrossRef
Chandarana UP, Momayez M, Taylor KW (2016) Monitoring and predicting slope instability: a review of current practices from a mining perspective. IJRET 05:139–151. https://doi.org/10.15623/ijret.2016.0511026CrossRef
Coe JA, Bessette-Kirton EK, Geertsema M (2018) Increasing rock-avalanche size and mobility in Glacier Bay National Park and Preserve, Alaska detected from 1984 to 2016 Landsat imagery. Landslides 15:393–407. https://doi.org/10.1007/s10346-017-0879-7CrossRef
Corkum AG, Martin CD (2004) Analysis of a rock slide stabilized with a toe-berm: a case study in British Columbia, Canada. Int J Rock Mech Min Sci 41:1109–1121. https://doi.org/10.1016/j.ijrmms.2004.04.008CrossRef
Cossart E, Braucher R, Fort M et al (2008) Slope instability in relation to glacial debuttressing in alpine areas (Upper Durance catchment, southeastern France): evidence from field data and 10Be cosmic ray exposure ages. Geomorphology 95:3–26. https://doi.org/10.1016/j.geomorph.2006.12.022CrossRef
Dai C, Higman B, Lynett PJ et al (2020) Detection and assessment of a large and potentially tsunamigenic periglacial landslide in Barry Arm, Alaska. Geophys Res Lett. https://doi.org/10.1029/2020GL089800CrossRef
Deline P, Gruber S, Delaloye R et al (2015) Ice loss and slope stability in High-Mountain regions. Snow and Ice-Related Hazards, Risks, and disasters. Elsevier, pp 521–561
Donati D, Borgatti L, Stead D et al (2022) The characterization of slope damage at the Civita Di Bagnoregio plateau using a remote sensing approach. Geotechnical engineering for the preservation of monuments and historic sites III. CRC
Donati D, Stead D, Rabus B et al (2023) Characterization of the Fels Landslide (Alaska) using combined terrestrial, aerial, and satellite remote sensing data. Remote Sens 16:117. https://doi.org/10.3390/rs16010117CrossRef
Edil T (1992) Landslide cases in the great lakes:issues and approaches. Transportation Research Record
Fein W (2022a) Kernbohrungen BK 1 Bis 3. Wendt Bohrgesellschaft mbH
Fein W (2022b) Geotechnischer untersuchungsbericht Zur Rutschung Im Steinbruch. der Jakob Bach GmbH & Co. KG in Nentershausen
Francioni M, Salvini R, Stead D et al (2015) An integrated remote sensing-GIS approach for the analysis of an open pit in the Carrara marble district, Italy: slope stability assessment through kinematic and numerical methods. Comput Geotech 67:46–63. https://doi.org/10.1016/j.compgeo.2015.02.009CrossRef
Grämiger LM, Moore JR, Gischig VS et al (2017) Beyond debuttressing: mechanics of paraglacial rock slope damage during repeat glacial cycles: PARAGLACIAL ROCK SLOPE MECHANICS. J Geophys Res Earth Surf 122:1004–1036. https://doi.org/10.1002/2016JF003967CrossRef
Işık NS, Doyuran V, Ulusay R (2004) Assessment of a coastal landslide subjected to building loads at Sinop, Black Sea region, Turkey, and stabilization measures. Eng Geol 75:69–88. https://doi.org/10.1016/j.enggeo.2004.05.002CrossRef
Jaboyedoff M, Derron M-H, Jakubowski J et al (2012) The 2006 Eiger rockslide, European Alps. In: Stead D, Clague JJ et al (eds) Landslides: types, mechanisms and modeling. Cambridge University Press, Cambridge, pp 282–296
Kos A, Amann F, Strozzi T et al (2016) Contemporary glacier retreat triggers a rapid landslide response, great Aletsch Glacier, Switzerland. Geophys Res Lett 43. https://doi.org/10.1002/2016GL071708
Lague D, Brodu N, Leroux J (2013) Accurate 3D comparison of complex topography with terrestrial laser scanner: application to the Rangitikei Canyon (N-Z). ISPRS J Photogramm Remote Sens 82:10–26. https://doi.org/10.1016/j.isprsjprs.2013.04.009CrossRef
Lemaire E, Dufresne A, Hamdi P et al (2024) Back-analysis of the paraglacial slope failure at Grewingk glacier and Lake, Alaska. Landslides 21:775–789. https://doi.org/10.1007/s10346-023-02177-6CrossRef
LGB (ed.) (2005) Geologie von Rheinland-Palz.- 400 p., 15 Anl., Landesamt für Geologie und Bergbau Rheinland-Pfalz, E. Schweizerbart’sche Verlagsbuchhandlung, Stuttgart
Macfarlane D (2009) Observations and predictions of the behaviour of large, slow-moving landslides in schist, Clyde dam reservoir, New Zealand. Eng Geol 109:5–15. https://doi.org/10.1016/j.enggeo.2009.02.005CrossRef
McColl ST, Davies TRH (2012) Large ice-contact slope movements: glacial buttressing, deformation and erosion: slope movement; glacier deformation, erosion and entrainment. Earth Surf Process Land 38:1102–1115. https://doi.org/10.1002/esp.3346CrossRef
McColl ST, Davies T, McSaveney M (2010) Glacier retreat and rock-slope stability: debunking debuttressing. Auckland, Aotearoa
Moriya R, Fukuda D, Kodama J, Fujii Y (2016) Effect of buttress on reduction of rock slope sliding along geological boundary. 3rd International Symposium on Mine Safety Science and Engineering 121–126
Moser M, Amann F, Meier J, Weidner S (2017) Tiefgreifende hangdeformationen der alpen. Springer Fachmedien Wiesbaden, WiesbadenCrossRef
Oral D, Akgun H, Kockar M (2015) Characterization and Assessment of Large Landslide Movement Along the Bursa-Inegöl-Bozüyük Highway in Turkey. pp 289–293
Rickli C, Bucher H, Böll A (2003) Oberflächennahe Rutschungen, ausgelöst durch die unwetter vom 15.-.16.7.2002 Im Napfgebiet und vom 31.8.-1.9.2002 Im gebiet Appenzell -. Projektbericht zuhanden des Bundesamtes für Wasser und Geologie BWG
Schaefer LN, Coe JA, Wikstrom Jones K et al (2023) Kinematic evolution of a large paraglacial landslide in the Barry Arm Fjord of Alaska. J Geophys Res Earth Surf 128:e2023JF007119. https://doi.org/10.1029/2023JF007119CrossRef
Schuster RL (2006) Interaction of dams and landslides: case studies and mitigation. U.S. Geological Survey
Souisa M, Sapulete SM, Tuarissa JC (2020) Back calculation of slope stability in Batu Gadjah, Ambon City using the solution of analytical and finite element method. J Phys: Conf Ser 1463:012034. https://doi.org/10.1088/1742-6596/1463/1/012034CrossRef
Stark T, Arellano W, Hillman R et al (2005) Effect of Toe Excavation on a Deep Bedrock Landslide. Journal of Performance of Constructed Facilities - J PERFORM CONSTR FACIL 19:. https://doi.org/10.1061/(ASCE)0887-3828(2005)19:3(244)
Teledyne (2022) Teledyne Geospatial
Ulusay R, Yolerì MF, Doyuran V (1993) Application of a rock buttress in the design of slopes of a strip coal mine adjacent to a highway. Can Geotech J 30:391–408. https://doi.org/10.1139/t93-035CrossRef
Wyllie DC, Mah CW (2004) Rock Slope Engineering: Civil and mining, 4th edn. CRC Press
Xu X, Huang Y, Xing Y, Guo Z (2022) Investigation of rainfall-induced toe-cut slope failure mechanisms in the southeastern coastal area of China. Nat Hazards 110:1761–1782. https://doi.org/10.1007/s11069-021-05011-1CrossRef
Yoshizawa N (1992) Landslide monitoring for presumption of underground slide surface. ISPRS Arch 29:478–485
