Elsevier

Engineering Geology

Volume 292, October 2021, 106281
Engineering Geology

Relationship between the spatial distribution of landslides and rock mass strength, and implications for the driving mechanism of landslides in tectonically active mountain ranges

https://doi.org/10.1016/j.enggeo.2021.106281Get rights and content

Highlights

  • We back calculate rock mass strengths ranging from 60 to 770 kPa in the study area.

  • More landslides occurred on the hillslopes with greater rock mass strength.

  • The dominant drivers of slope failure are rock uplift, river incision and earthquakes.

Abstract

The relationship between landslides and rock mass strength is fundamental for assessing landslide hazards. Some researchers have proposed that there is an inverse relationship between the number of landslides and rock mass strength. However, in some tectonically active mountain ranges, higher rates of landsliding appear to be associated with greater rock mass strength. We investigated the relation between landslides and rock mass strength in the Langxian (LX), Lulang (LL), and Tongmai (TM) regions in southeastern Tibet by identifying and mapping 294 large bedrock landslides using 10-m resolution lidar bare-earth imagery. An inverse relationship between topographic relief and the slope angle of historical landslides demonstrates that rock mass strength is an important factor controlling relief in the study area. Applying Culmann's method, we back-calculated rock mass strengths ranging from 60 to 770 kPa at the landscape scale. Our data show that, at the landscape scale, more landslides have occurred on the hillslopes with greater rock mass strength than on those with lower rock mass strength. We conclude that the stability of slopes in our study areas is controlled by rock mass strength, but the dominant drivers of failure are rock uplift and river incision, rather than a reduction in rock strength as has been proposed in some tectonically passive regions.

Introduction

Rock mass strength is controlled by intact rock strength, the presence of structural discontinuities (joint sets, faults), and rock bridges (Hoek, 1983, Hoek, 1998). At the hillslope scale, rock mass strength plays a fundamental role in landscape evolution (Tucker and Slingerland, 1994; Korup, 2008; Moore et al., 2009; Clarke and Burbank, 2010) and must be considered in studies of slope stability targeted at hazard analysis (Wang et al., 2014). Rock mass strength can be understood as the laboratory strength of a complete rock block of the slope, and also as the strength of the whole rock slope. At the outcrop scale, rock mass strength can be determined by testing representative rock blocks in the laboratory. The measurements from such tests provide inputs into numerical models that are used to estimate, albeit with large uncertainties, the stability of specific slopes (Hoek, 1983; Clarke and Burbank, 2010; Saroglou et al., 2018). The strength of the whole rock slope is very difficult to define in a quantitative way (Hoek, 1983; Clarke and Burbank, 2010; Saroglou et al., 2018). In this study, we focus on the rock mass strength of the entire slope, including the complete rock block and its joints and fissures.

Many researchers have argued that, all other factors being equal, relatively high rock mass strength, even in areas of high relief and steep slopes, results in lower frequencies of landslides (Sklar and Dietrich, 2001; Stock et al., 2005; Molnar et al., 2007; Corominas and Moya, 2008; Moore et al., 2009; Qi et al., 2010; Wang et al., 2014; Wang et al., 2020). In rapidly uplifting mountain ranges, however, high hillslope relief and steep slopes are the hallmarks of high rates of denudation (Montgomery and Brandon, 2002; Clague and Evans, 2003; Korup et al., 2007; DiBiase et al., 2010), which suggests that rock mass strength and landslide rates in tectonically active mountain ranges are positively related, at least on long timescales.

This apparent contradiction may arise from scales of study. Laboratory experiments are based on ‘hand’ samples, typically without fractures and in an environment with a non-varying “climate” (Hoek, 1983; Sklar and Dietrich, 2001; Saroglou et al., 2018). It is true that non-fractured hard rocks are more difficult to abrade or fail than fractured soft rocks, and it is also true that hard rock will host high-relief landscapes. All other factors being equal, high-relief landscapes in hard rock should erode slowly. It therefore follows that high-relief landscapes are prone to erosion due to other factors, for example fracture density, weathering, rainstorms, or rapid uplift (Korup, 2008; Moore et al., 2009; Clarke and Burbank, 2010).

This discussion highlights the need to further examine the relationship between rock mass strength and landslides. This relationship has implications, not only for rock mass strength, but also for external factors such as rock uplift, river incision, earthquakes, and rainfall, which clearly drive landslides in tectonically active mountain ranges. Many researchers have assumed or argued that relatively high rock mass strength, even in areas of high relief and steep slopes, leads to lower landslide frequency (Corominas and Moya, 2008; Qi et al., 2010; Wang et al., 2014). However, if there is a positive relationship between rock mass strength and denudation driven by landslides, the hazard posed by slope failure might be higher than assumed based on strength alone. In this study, we focus on the relationship between rock mass strength at the hillslope scale and landslides in the tectonically active mountains of southeastern Tibet. We document the spatial distributions of 294 large bedrock landslides in three study areas and infer their relations to inferred rock mass strength, relief, and possible triggering factors. Using lidar bare-earth imagery with 10 m resolution and relief and slope values for historical landslides (Figs. 1b and c), we back-calculate rock mass strength at the landscape scale using Culmann's two-dimensional slope stability model (Culmann, 1875; Schmidt and Montgomery, 1995).

Section snippets

Study area

To explore the relationship between rock mass strength at the hillslope scale and landslides, we selected for our study the Langxian (LX), Tongmai (TM), and Lulang (LL) areas, located in the Eastern Himalayan syntaxis in southeastern Tibet (Fig. 1). The LX study area (640 km2) is dominated by phyllite, granite, and conglomerate (Chen et al., 2004; Xu et al., 2015a; Liu et al., 2020) (Fig. 2c); most of the landslides in this area have occurred within phyllite and conglomerate.

Topographic measurements and landslide size distribution

We identified 294 large bedrock slumps and rotational rockslides (minimum area 6 × 105 m2) and 1289 slope units (Fig. 1) using hillshade and slope maps derived from a 2014 10 m-resolution DEM, satellite imagery, and helicopter-based remote sensing imagery. Differences in slope, slope aspect, and roughness allowed us to accurately delineate landslides and to distinguish source areas from deposits (Fig. 2a, b; McKean and Roering, 2004). In plotting the boundaries of landslide and slope units, we

Topographic characteristics of landslides

The constant γ in the relationship between landslide area and volume (V = α×Aγ) is related to the properties of rock or soil (Larsen et al., 2010). We used a value of γ = 1.526 (σ = 0.07) for our dataset (Fig. 3), which is near the upper end of the range of values (1.3–1.6) in a global dataset of bedrock landslides (Larsen et al., 2010). The range of values of volume in the volume-area scaling relationship (V = 0.009 × A1.526, R2 = 0.60) reflects uncertainty in defining the area of landslide

Relationship between rock mass strength and landslide occurrence

Landslide volumes in sub-areas E and F of LX are higher than those in the other four sub-areas (Fig. 8). In addition, rock mass strengths of hillslopes in areas E and F are higher than those of hillslopes in the other sub-areas (Fig. 7). This result is contrary to the previously argued inverse relationship between landslides and rock mass strength (Sklar and Dietrich, 2001; Stock et al., 2005; Molnar et al., 2007; Moore et al., 2009; Wang et al., 2020). Hence, we infer that, at the landscape

Conclusions

Our estimates of rock mass strength in an area of ~1320 km2 in southeast Tibet range from 60 to 770 kPa. We show that more landslides have occurred on the hillslopes with greater rock mass strength at the landscape scale than on hillslopes with lower rock mass strength. We conclude that Culmann's (1875) slope stability model explains this relationship. At the landscape scale, rock mass strength strongly affects the stability of slopes in mountain regions with high rates of uplift. Large

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research was supported by the National Key R&D Program of China (No. 2019YFC1509703), the Second Tibetan Plateau Scientific Expedition and Research Program (STEP) (Grant No. 2019QZKK0904), the Foundation of China Scholarship Council.

References (48)

  • J. McKean et al.

    Objective landslide detection and surface morphology mapping using high-resolution airborne laser altimetry

    Geomorphology

    (2004)
  • D.R. Montgomery et al.

    Topographic controls on erosion rates in tectonically active mountain ranges

    Earth Planet. Sci. Lett.

    (2002)
  • S.W. Qi et al.

    Spatial distribution analysis of landslides triggered by 2008.5.12 Wenchuan earthquake, China

    Eng. Geol.

    (2010)
  • D. Stead et al.

    A critical review of rock slope failure mechanisms: the importance of structural geology

    J. Struct. Geol.

    (2015)
  • A. Valagussa et al.

    Seismic and geological controls on earthquake-induced landslide size

    Earth Planet. Sci. Lett.

    (2019)
  • X.L. Wang et al.

    Dynamic rockfall risk analysis

    Eng. Geol.

    (2020)
  • X.L. Wang et al.

    Fault controls on spatial variation of fracture density and rock mass strength within the Yarlung Tsangpo Fault damage zone (southeastern Tibet)

    Eng. Geol.

    (2021)
  • D.W. Burbank et al.

    Bedrock incision, rock uplift and threshold hillslopes in the northwestern Himalayas

    Nature

    (1996)
  • Q. Chen et al.

    Spatially variable extension in southern Tibet based on GPS measurements

    J. Geophys. Res.

    (2004)
  • J.J. Clague et al.

    Geologic framework of large historic landslides in Thompson River Valley, British Columbia

    Environ. Eng. Geosci.

    (2003)
  • P. Coudurier-Curveur et al.

    A composite rupture model for the great 1950 Assam earthquake across the cusp of the East Himalayan Syntaxis

    Earth Planet. Sci. Lett.

    (2020)
  • G.B. Crosta et al.

    Global scale analysis of Martian landslide mobility and paleoenvironmental clues

    Journal of Geophysical Research: Planets

    (2018)
  • C. Culmann

    Die Graphische Statik. Meyer and Zeller, Zurich

    (1875)
  • N.J. Finnegan et al.

    Coupling of rock uplift and river incision in the Namche Barwa–Gyala Peri massif, Tibet

    Geol. Soc. Am. Bull.

    (2008)
  • Cited by (18)

    • Insights into some large-scale landslides in southeastern margin of Qinghai-Tibet Plateau

      2023, Journal of Rock Mechanics and Geotechnical Engineering
      Citation Excerpt :

      Some studies have revealed that large landslides on the eastern margin of the QTP are mainly distributed along major (mainstream) rivers because these rivers usually have steep deeply incised valleys that follow active faults and fractured zones; these features make slopes along large rivers more sensitive to earthquakes, and rainfall. ( Zhao et al., 2019a, b, 2021b; Zhao, 2020; Wang et al., 2021a). Unfortunately, most infrastructures, such as roads, railways, and bridges, on the southeastern margin of the QTP are mainly constructed along the rivers, especially in the middle and lower sections of the valley slopes, due to the restrictions of construction space and cost.

    • A study on the differences in radon exhalation of different lithologies at various depths and the factors influencing its distribution in northern Shaanxi, China

      2022, Science of the Total Environment
      Citation Excerpt :

      Moreover, the bending and fracture phenomenon occurred during the rock formation, which further stimulated the secondary distribution of rock formation and related structural reorganization. However, the basic properties of the mineral components do not undergo a major transformation, leading to the enhancement in the compactness during rock formation (Liao et al., 2020; Tu et al., 2020; Wang et al., 2021). Fig. 6 shows the relationship between the radon exhalation rate of each drilled core and the borehole depth.

    View all citing articles on Scopus
    View full text