Frictional properties of the rupture surface of a carbonate rock avalanche
Introduction
Frictional resistance along the rupture surface is one of the main factors controlling the velocity at which a rock mass can slide on the rupture surface and therefore it affects the mobility of rock avalanches.1, 2, 3 Most scholars have agreed that reduction or even loss of strength along the rupture surface leads to the rapid initiation and sliding of a displaced rock mass1,4, 5, 6, 7, 8 before it moves out of the toe of the rupture surface.9 However, the mechanisms causing the strength decrease are very complicated and much debated.
Due to the shortcomings of traditional rock mechanic tests, frictional resistance along the rupture surface in rock avalanches remains largely unexplored, although analyses and observations suggest a low mobilized friction coefficient during their runout.3,6,7 Several hypotheses have been proposed, such as frictional heating,10,11 frictional-heat-triggered gaseous pore pressure,12, 13, 14 and frictional-heat-triggered mineral decomposition and production of CO2,15,16 but none has been experimentally proven or ruled out.
In recent years, high-velocity-friction tests originally designed to examine the frictional properties of seismic fault planes,17, 18, 19 have been used to test frictional resistance variation during the sliding of detached rock masses on the rupture surface of rock avalanches. Dong et al.20 conducted high-velocity-friction tests on gouge and shale powder to get the curves of friction coefficient with shear displacement, and finally obtained a curve of velocity-displacement-dependent friction law. Yang et al.21 conducted high-velocity-friction tests on gouge from faults parallel to the bedding at shear rates from 0.0023 to 1.3 m/s under a normal stress corresponding to the overburden pressure of the landslide mass. The results revealed three types of frictional behavior: (1) slip-strengthening behavior to the peak friction followed by slight slip-weakening or nearly constant friction; (2) transitional behavior with slip-strengthening to the peak friction followed by slip-weakening; (3) marked slip-weakening behavior. They found that slip-weakening was essential in initiating the landslide and a low friction coefficient (0.08–0.1) made the high speed of the landslide possible. But the conclusions of both of Dong et al.20 and Yang et al.21 were only applicable to rockslides which failed along the gouge with stable thickness. Hu et al.22 used high-velocity-friction tests to replicate the temperatures and mineral changes on the rupture surface during initiation. They confirmed heating above 800 °C on the rupture surface, and thermal decomposition of dolomite to magnesium and calcium oxides in the shallowest samples. Because the aim was to determine the change and depth of mineral distribution, they only conducted tests at two shear rates of 1 and 2 m/s and did not investigate the variation of frictional properties with shear rate and the mechanisms leading to a low friction coefficient.
However, no studies have paid attention on the micro-mechanisms by which the weakening is achieved on the landslide rupture surface, and whether the mechanisms proposed for the shear weakening of faults are applicable to that of the landslide rupture surface as well.
The catastrophic Jiweishan rock avalanche had a 720-m-long rupture surface, which developed along an upper limestone and lower interbedded shale. It was very likely that the friction on the rupture surface played an important role on the acceleration of the sliding rock mass during initiation and its subsequent rapid and long-distance transport. We conducted high-velocity-friction tests with limestone and shale samples from the Jiweishan rock avalanche at different shear rates to examine the frictional properties along the rupture surface. The tests indicated there is a critical value of shear rate between 0.05 and 0.2 m/s as the demarcation point of shear weakening and shear strengthening. And we propose two micro-mechanisms which was used to explain sharp drop of frictional resistance on seismic faults as the possible reasons leading to the extremely low friction coefficient on the rupture surface of carbonate rock avalanches.
Section snippets
Case study
The Jiweishan rock avalanche occurred on 5 June 2009, in Tiekuang Town, Wulong County, Chongqing, China.23 The rock avalanche initially started as a rockslide and then turned into a catastrophic rock avalanche, travelling more than 1500 m and burying 74 people.24,25
Friction coefficient and temperature curves
Fig. 3 presents temperature and friction coefficient curves for the tests at shear rates of 0.05, 0.2, 0.5, 1.0 and 2.1 m/s, respectively. The friction coefficient at five different rates all increased rapidly to a peak value of 0.55 to 0.8 immediately after shearing began, then fluctuated sharply due to the rough undulating shear surfaces of the limestone and shale samples. The length of the fluctuation differed because of the roughness of the shear surfaces. For example, the curve for a shear
Compositional and textural changes
The shear surfaces of the samples obviously had gone through mineral decomposition and recrystallization during shearing that caused mineral and texture changes. Many studies have shown that calcite decomposes to CaO and lime CO2 gas at 600–850 °C.31, 32, 33 The temperature in our tests at the shear surface increased to higher than 600 °C at shear rates of 0.5, 1.0 and 2.1 m/s. Because the thermal probes did not touch the shear surface, and the temperature obtained by them was actually a little
Conclusions
High velocity friction tests were used to investigate the frictional properties along the rupture surface of the Jiweishan rock avalanche. The results indicated that shear-strengthening and rate-strengthening of friction occurred on the rupture surface when the sliding rock mass slid at a low velocity less than a critical value, probably due to the thermal expansion of abrasion-induced gouge. Shear-weakening and rate-weakening of friction occurred when the sliding rock mass slid at a high
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.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (41772334 and 41920104007). The high-velocity-friction tests were conducted at the Institute of Geology, China Earthquake Administration. We thank Dr. Eileen McSaveney for editing our paper to make it more readable.
References (46)
- et al.
Dynamics, mobility-controlling factors and transport mechanisms of rapid long-runout rock avalanches in China
Eng Geol
(2013) - et al.
Numerical modelling of the 2008 Wenchuan earthquake-triggered Daguangbao landslide using a velocity and displacement dependent friction law
Eng Geol
(2016) - et al.
Dynamics of the 2008 earthquake-triggered Wenjiagou Creek rock avalanche, Qingping, Sichuan, China
Eng Geol
(2016) - et al.
The Vaiont Slide- A geotechnical analysis based on new geologic observations of the failure surface
Eng Geol
(1987) - et al.
Initiation, movement, and run-out of the giant Tsaoling landslide- what can we learn from a simple rigid block model and a velocity-displacement dependent friction law?
Eng Geol
(2014) - et al.
The 2009 Jiweishan rock avalanche, Wulong, China: Precursor conditions and factors leading to failure
Eng Geol
(2018) - et al.
Is air pollution causing landslides in China?
Earth Planet Sci Lett
(2018) - et al.
Dynamics of the Wulong landslide revealed by broadband seismic records
Earth Planets Space
(2017) Mechanism and kinetics of thermal decomposition of carbonates
Thermochim Acta
(2002)- et al.
Impact-generated carbonate melts: evidence from the Haughton structure, Canada
Earth Planet Sci Lett
(2001)