Search

Rainfall-induced landslides on steep slopes are potential hazards to life and property. The state of Kerala in India is witnessing unprecedented death and damage since last 3 years due to landslides triggered by heavy rainfall. In this paper, we present a cause-impact analysis of landslides that occurred due to the monsoon rainfall in the years 2018, 2019 and 2020. We have created an event-based landslide inventory post-rainfall season for the entire state and then selected six major landslides for detailed analysis. Event-based landslide inventory maps are prepared from Earth observation satellite data such as Resourcesat-2, 2A, Cartosat-2S, WorldView-2, Pleiades-1A and 1B, SPOT-6 and 7 and Sentinel-2 acquired before and after the rainfall events. Then, rainfall data were analysed to find out the role of daily and antecedent rainfalls on landslide trigger. Finally, we have carried out numerical debris flow modeling to characterise landslides and assess the impact. Our results show that landslides are triggered by antecedent rainfall of variable durations and are mostly debris flow type that has followed the channel path. The runout length and lateral spreads are controlled by variable frictional parameters of the slope material. Furthermore, reactivation at the crown region is also noticed in few cases that need to be monitored. [ABSTRACT FROM AUTHOR]

Kerala is the third most densely populated state in India. It is a narrow strip of land, of which 47% is occupied by the most prominent orographic feature of peninsular India, The Western Ghats mountain chain. The highlands of Kerala experience several types of landslides, of which debris flows are the most common. They are called "Urul Pottal" in the local vernacular. The west-facing Western Ghats scarps that runs the entire extent of the mountain system is the most prone physiographic unit for landslides. The highlands of the region experience an annual average rainfall as high as 500 cm through the South-West, North-East and Pre-Monsoon showers. A survey of ancient documents and early news papers indicates a reduced rate of slope instability in the past. The processes leading to landslides were accelerated by anthropogenic disturbances such as deforestation since the early 18th century, terracing and obstruction of ephemeral streams and cultivation of crops lacking capability to add root cohesion in steep slopes. The events have become more destructive given the increasing vulnerability of population and property. Majority of mass movements have occurred in hill slopes >20° along the Western Ghats scarps, the only exception being the coastal cliffs. Studies conducted in the state indicates that prolonged and intense rainfall or more particularly a combination of the two and the resultant pore pressure variations are the most important trigger of landslides. The initiation zone of most of the landslides was typical hollows generally having degraded natural vegetation. A survey of post-landslide investigation and news paper reports enabled the identification of 29 major landslide events in the state. All except one of the 14 districts in the state are prone to landslides. Wayanad and Kozhikode districts are prone to deep seated landslides, while Idukki and Kottayam are prone to shallow landslides. [ABSTRACT FROM AUTHOR]

The Taliye landslide, a hill slope debris flow, is one of the recent large-scale landslides in India in terms of mortality and socio-economic predicaments. In general, the Taliye area exhibits a moderately dissected plateau relief, and the landslide was initiated from a north-westerly slope of a west trending major ridge and flowed through a paddy field to bury houses located just below it. The landslide had a total runout of 563 m and maximum width of 230 m at the toe. The heavy rainfall and presence of a narrow stream may be the prime triggering factors, but the affected area had also been exposed to human interference for farming as well as for habitation. Given its socio-economic relevance and landslide proneness of this area, a post-event engineering geological investigation and numerical modelling of this landslide have been carried out to have a closer look at its type, causes of failure, back calculation of run-out characteristics, as well as to calibrate frictional parameters for this area. The numerical modelling package, rapid mass movements (RAMMS) was used to undertake back analysis of this debris flow by referring the original flow shape. For the precise preference of Voellmy frictional parameters, dry-Coulomb friction (µ) and viscous-turbulent friction (ξ) coefficients, this study applies the popular receiver operative characteristics (ROC) technique as the model validation tool. The landslide may have initiated as a talus or translational failure of shallow soil on the hillock, but later might have transformed into a debris flow by entraining materials from the paddy field. Amongst the different combinations of frictional parameters, the model with µ of 0.06 m/s2 and ξ of 1450 m/s2 has emerged as the best with an area under the curve (AUC) value of 0.883 for the ROC assessment. With these calibrated frictional parameters, the maximum flow velocity of this debris flow was simulated to be in the order of 5.26 m/s taking place at the middle reaches, and the maximum flow velocity and pressure were derived as 21 m/s and 0.92 kPa, respectively, concentrated in the hillock zone. Although it was a rain-induced failure, the investigation suggests that improper slope management practices might have amplified its magnitude and got manifested as devastating landslide. Therefore, human activities in these hilly regions shall be supported with systematic landslide hazard evaluation, and here in this case, the calibrated frictional parameters may be useful for debris flow modelling and landslide risk reduction in the area. [ABSTRACT FROM AUTHOR]