1 Introduction

A landslide has been a natural hazard frequently occurring in mountainous regions and causes serious property damage, communication passage and sometimes human lives (Ferlisi et al. 2012; Singh et al. 2014; Chen et al. 2015). The frequency of landslide is maximum during the monsoon period from July to September and after snowfall from January to March. Sometimes, landslide occurred by endogenic forces like strong earthquakes and volcanic activities (Singh et al. 2013). Moreover, study shows that the landslide mortality rate exceeds one per 100 km2 per years in developing countries like India, China, Nepal, Peru, Venezuela, Philippines and Tajikistan (Petley et al. 2005; Nadim and Kjekstad 2009). A later study confirmed that about 80 % of landslide mortality tolerated by developing countries and India accounts about 8 % (Kirschbaum et al. 2010) of it. Moreover, 15 % (about 0.49 million km2) landmasses in India prone to landslide hazard that includes Himalayan terrain, Ranchi plateau, Eastern and Western Ghats (Wadhawan et al. 2013). About 20 states of India are affected by different intensities of landslide hazard (GSI 2015). On the basis of landslide hazard map prepared by Geological Survey of India (GSI), states of Sikkim, Mizoram and some districts of Jammu and Kashmir fall under very high hazard zone and most of the districts of Jammu and Kashmir, Himachal Pradesh, Arunachal Pradesh, Nagaland and Manipur come under high hazard class (GSI 2015). Moreover, the hilly tracts of states like Maharashtra, Goa, Karnataka, Andhra Pradesh, Tamil Nadu, Madhya Pradesh and Kerala comes under moderate risk zone (GSI 2015).

Malin landslide was one of the immense landslides occurred on July 30, 2014, in the Malin village of Pune district in Maharashtra, India. The landslide was so extensive that it wiped out whole village except a primary school and few houses remained safe from the disaster. Approximately 55 houses have been buried, and more than 153 peoples were trapped in this massive landslide that hit early in the morning when almost all resident were either asleep or inside the houses due to continuous heavy rainfall from last couple of days. The event was first noticed by a bus driver, who saw that the whole village had disappeared under debris due to the landslide.

This paper mainly describes the field investigations to find out the primary causes of Malin landslide along with numerical analyses to determine the stability of the hill slope and the critical failure surface. In the present scenario, n-number of conventional and/or numerical methods existing to study the landslide and slope stability (Sarkar et al. 2008; Singh et al. 2008; Palma et al. 2012; Ramakrishnan et al. 2013). For the present study, the stability analyses were performed with the help of numerical program Slide v.6 and Phase 2, based on limit equilibrium method (LEM) and finite element method (FEM), respectively (Rocscience 2010).

2 Study area

Malin village is the northern part of Western Ghat of Deccan Trap and is located on the one end of the reservoir of Dimbhe dam, situated on Ghod river, a tributary of Bhima river in the Ambegaon Tehsil of Pune district in Maharashtra, India. The coordinates of the village are N19°09′41.4″ and E73°41′16.8″, and the area comes under Topo sheet number 47E/12. It is located about 40 km from the famous Bhimashankar temple and about 150 km from Pune. Location and accessibility map of Malin village are shown in Fig. 1, and the topography consists of highly dissected terrain with flat summits and entrenched valleys. The landslide-affected area is totally covered by Deccan basalt with top portion (~10 m) and is highly weathered and converted into soil. Two types of flows, i.e., massive and vesicular, are observed. Vesicular flows produce gentler slopes and have infilling of secondary minerals like varieties of silica, calcite and zeolite. The vesicles are of different shapes and sizes and comparatively more abundant in the upper surface. The spheroidal weathering is prominent in the area, which leads to the formation of soil and ultimately reduces the strength of geomaterials. Few joints/cracks were also observed in the slide-affected area that may play a vital role in the percolation of the rain water inside the slope face increasing the pore water pressure. The heavy rainfall has been occurring for last 3 days, and due to the lack/absence of any proper drainage system, water was accumulated at the top of the hill. The accumulated rainwater percolates through either interface of weathered material and rock and/or major/minors joints and cracks at the surface of the hill. It may also lead to extensive landslide that the whole village except few houses and a primary school were buried under a huge mass of debris (Fig. 2). The failure mainly occurred at the top of the hill at the contact zone of weathered material and rock. The geomaterials flow in a channelize way, and it mainly composed of fine grain particles to big boulders. Moreover, unscientific construction (houses at the base of the slope) activities and unplanned cultivations (cultivation of rice and wheat at the top of the hill) were also observed in the area, and these factors play very crucial role of catalyst for natural hazards.

Fig. 1
figure 1

Location and accessibility map of Malin village

Full size image
Fig. 2
figure 2

a View of landmass movement, b landslide derived boulders (~5 ft), broken and partially broken houses, c channel of the slide along the hill slope carries variable size of particles, d views of the landslide from top to bottom of the hill, e the exposure of the failure plane after slide, f localized drainage at crest of slope to deviate the water

Full size image

3 Methodology

A detailed field investigations were carried out to diagnose the primary causes of landslide and also to collect representative slope-forming geomaterial (soil/rock) samples to measure the input parameters for the numerical simulation. The disturb soil samples have been collected from three locations of hill, namely L1 (bottom of the hill), L2 (middle of the hill) and L3 (top of the hill). The rock samples were collected as massive and vesicular basalt from the bottom, middle and top of the hill as the parent geomaterial of the study area from Deccan basalt. All these collected samples were tested in the laboratory as per ASTM and ISRM standards (Table 1) to determine natural moisture content, specific gravity, particles distribution, unit weight, cohesion, friction angle, tensile strength, Young’s modulus and Poisson’s ratio. For the accuracy purpose, three samples have been tested for each geotechnical parameters as per suggested standards. The hill geometry and variation in the lithology of the slope were identified during the field investigation. The vesicular basalt is sandwiches between massive basalt, which is overlain by weathered soil. The determined soil/rock properties and hill slope geometry were used in numerical simulation. The simulation was performed using Slide v.6 and Phase 2 program, based on LEM and FEM, respectively.

Table 1 Standards for determination of different geotechnical and physical parameters
Full size table

4 Numerical simulation

Numerical simulation is a computer-based program in which the domain of the entire problem is discretized, then solves those separately with methods like limit equilibrium, finite element and finite difference and finally merges the partial results into the solution of the entire problem. It is the only method which transforms physical problem into numerical model in the language of mathematical equation. The present problem was analyzed using limit equilibrium method (LEM) and finite element method (FEM).

For the soil slope stability analysis, the LEM is the most popular and widely used technique. This method is based on the principal that the loose geomaterials above the trial failure surface are divided into several vertical slices. The width of each slice need not be same, and it depends upon the slope geometry and geomaterials profile. The equilibrium of force or moment or combination is satisfied for every individual slice (Fig. 3a, b). In this method, Mohr–Coulomb criterion is used as the failure condition for slope geomaterials. In this method, FoS of the slope is calculated as:

Fig. 3
figure 3

Simplified Bishop’s method, a slices of the soil above failure plane, b effect of the forces on the side of a particular slide (Bishop 1955)

Full size image

The forces that act on the typical slice (assume nth slice) are:

W n  = weight of the slice

N r and T r  = normal and tangential component of reaction R, respectively

P n and P n+1 = normal force act on the side of the slice

T n and T n+1 = shearing force act on the side of the slice

In this method, pore water pressure is assumed to be zero, and the forces P n and T n are equal in magnitude to the resultant of the P n+1 and T n+1.

At equilibrium,Forces acting in the vertical direction

$$N_{r} \cos \alpha_{n} = W_{n} + \left( {T_{n} - T_{n + 1} } \right) - U\cos \alpha_{n} - T_{r} \sin \alpha_{n}$$
$$T_{r} = \frac{{C\Delta L_{n} }}{{F_{s} }} + N_{r} \frac{\tan \phi }{{F_{s} }}$$

On solving these equations,

$$N_{r} = \frac{{W_{n} + \Delta T - U\cos \alpha_{n} + \frac{{C\Delta L_{n} }}{{F_{s} }}\sin \alpha_{n} }}{{\cos \alpha_{n} + \frac{{\tan \phi \sin \alpha_{n} }}{{F_{s} }}}}$$

Here, \(\Delta T = T_{n} - T_{n + 1}\)

By substituting and solving,

$$F_{s} = \frac{{\sum\nolimits_{n = 1}^{n = p} {\left[ {C\Delta L_{n} \cos \alpha_{n} + \left( {\left( {W_{n} - U\cos \alpha_{n} } \right) + \Delta T} \right)\tan \phi } \right]\frac{1}{{m_{\alpha } }}} }}{{\sum\nolimits_{n = 1}^{n = p} {W_{n} \sin \alpha_{n} } }}$$

where

$$m_{\alpha } = \cos \alpha_{n} + \frac{{\tan \phi \sin \alpha_{n} }}{{F_{s} }}$$

F s is the FoS along the trial slip surface. This method is limited to circular type of failure (Bishop 1955). So it is very useful for the analysis of slope failure composed of loose geomaterials like soil/debris.

Limit equilibrium methods have some assumption and limitation. To reduce these, finite element method (FEM) has been used. This technique has emerged as most powerful alternative for the soil slope stability analysis. In this method, soil material is modelled employing Mohr–Coulomb failure criterion (Zienkiewic 1977; Jing 2003; Singh et al. 2013). The soil slope fails because the shear strength of the soil is too low to be able to resist the shear stress developed along the sliding surface. Thus, the FoS along the sliding surface is defined as:

$$F_{s} = \frac{\tau }{{\tau_{f} }}$$

where τ, the shear strength of the slope material, which is calculated though Mohr–Coulomb criteria:

$$\tau = C + \sigma_{n} \tan \phi$$

and τ f , the shear stress on the sliding surface. This can be calculated as:

$$\tau_{f} = C_{f} + \sigma_{n} \tan \phi_{f}$$

where C f and ϕ f are related to shear strength parameters of slope by a factor called as strength reduction factor and defined as:

$$C_{f} = \frac{C}{\text{SRF}}$$
$$\phi_{f} = \tan^{ - 1} \left( {\frac{\tan \phi }{\text{SRF}}} \right)$$

where SRF is strength reduction parameters (Matsui and San 1992; Griffiths and Lane 1999; Dawson et al. 1999; Hammah et al. 2007; Kainthola et al. 2012; Singh et al. 2013). In present study, the soil slope was analyzed using Phase 2 software based on FEM. The analysis used Mohr–Coulomb failure criteria under gravitational force and utilized 6 node triangular mesh.

5 Results and discussion

The field studies were carried out to explore primary causes of landslide and the identification of venerable failure zone of Malin hill slope. The vulnerable failure zone was identified at the top of the hill, i.e., at interface of the soil and rock. The numerical analyses of hill slope required geotechnical parameters like unit weight, cohesion friction angle, tensile strength, Young’s modulus, Poisson’s ratio (Table 2) and the hill slope geometry (Fig. 4), which have been determined by laboratory analyses and field observations, respectively. The general physical properties of the slope-forming geomaterial were also determined as per standards (Table 3).

Fig. 4
figure 4

Two-dimensional geometry of the Malin hill slope (Numerical values are in meter)

Full size image
Table 2 Geotechnical parameters of hill slope-forming geomaterial
Full size table
Table 3 General physical properties of hill slope-forming geomaterial
Full size table

The stability of the hill slope was analyzed using traditional LEM 2D slide program. The analysis was first accomplished by Bishop’s simplified method (Bishop 1955), which gave a FoS of 0.784 (Fig. 5a). For precision, the hill slope was also scrutinized using Janbu’s simplified method (Janbu et al. 1956) (Fig. 5b), Janbu’s corrected method (Janbu 1968) (Fig. 5c) and Spencer method (Spencer 1967) (Fig. 5d) that gave FoS 0.785, 0.797 and 0.782, respectively. The variation in the color of slip surface indicates FoS along the slip surface. From the analysis, it was observed that the critical slip surface occurred at the top of the hill slope and at the contact zone of rock and soil, indicating that the driving forces were relatively higher at the contact zone and caused failure.

Fig. 5
figure 5

Analyzed Malin hill slope by LEM, a Bishop’s simplified method, b Janbu’s simplified method, c Janbu’s corrected method and d Spencer method

Full size image

To determine the maximum displacement, displacement direction and developed maximum shear strain at the hill slope, the FEM analysis was carried out keeping same geometry with the help of Phase 2 program. The maximum total displacement is estimated about 20 meters at the top of the hill slope (Figs. 6a, 7), and the displacement direction (Fig. 6b) shows that the geomaterials moved toward the toe of the hill. In the field, it was also observed that the zone EFG (Fig. 7) exhibits maximum mobility and mass movement toward the toe of the hill. Moreover, the maximum shear strain was developed at the top of the hill slope along the slip surface (Fig. 6c) that was validated by field observations also.

Fig. 6
figure 6

Analyzed Malin hill slope by FEM, a total displacement, b total displacement in vector form, c maximum shear strain

Full size image
Fig. 7
figure 7

Displacement variations with distance in the FEM simulation

Full size image

6 Conclusions

By field observations, it is found that the Malin landslide may have occured due to heavy rainfall that last for 3 days, improper drainage system at the top of hill, unscientific constructions and unplanned cultivations. The heavy rain water during 3 days percolated either through interface of rock and weathered soil or through the major/minor joints/cracks reducing the strength parameters of slope-forming geomaterials. Lastly, the landslide was triggered at the top of the hill slope and covered almost entire houses which led to huge monetary and human lives loss.

The numerical analysis shows that the hill slope was unstable with factor of safety less than 1, and the slip surface developed at the interface of weathered soil and rock. The maximum displacement and shear strain developed along slip surface and failed toward the toe of the hill slope.

To reduce/minimize impact and frequency of landslide in the area, the following activities should be adopted:

  • Scientific construction activity should be adopted like houses should be made of brick and cement as there is minor crack observed in buildings near the school, and there were no cracks in school buildings. The poor construction of house may lead to more casualties.

  • There should be a proper drainage system at top of the hill so that there will be less percolation of water through the joints ultimately reducing the pore water pressure.

  • Detailed numerical analyses should be performed to calculate the stability of hill slope for identification of vulnerable zone/zones.