Skip to Main Content
*

Abstract

Unpublished reports of investigations carried out after the 1983 Sale Mountain landslide and some of the published papers related to the occurrence, mechanism, and mobility of the landslide are reviewed herein. The landslide occurred on the high, steep south slope of Sale Mountain, which comprises nearly horizontal Pliocene siltstones and mudstones covered by 120 m of Pleistocene eolian loess. The volume of loess involved in the slide was less than one-third of the total volume of the sliding mass, so it is was not a loess landslide, but a loess-covered mudstone landslide. Although the landslide occurred suddenly, before its occurrence there was a long-term preparatory stage, in which gravitational creep and tension fracturing were important processes leading to the final abrupt failure of the slope. Most researchers have suggested that the mechanism of the landslide was progressive failure that began with extremely slow sliding and tension fracturing, and ended with shearing through resisting elements across the bedding of the Pliocene sediments. The sliding velocity of the landslide was extremely rapid. The average velocity was ~20 m/s. The Fahrböschung is 11°, which represents an excessive travel distance of 1120 m.

Introduction

On March 7, 1983, a catastrophic landslide occurred in Dongxiang county, Gansu Province, in the Loess Plateau of China. The geographic coordinates of the landslide are 105°35′10″E, 35°33′40″N (Fig. 1). The peak and the steep south slope of Sale Mountain slipped suddenly: after sliding, the peak had dropped from 2283 m elevation to ~2080 m, a vertical displacement of ~200 m. The toe of the displaced mass pushed forward across the more than 800-m-wide valley of the Baxie River and climbed 10 m up the opposite bank before stopping. Three villages on the second river terrace level just under the foot of the mountain, and near the toe of the rupture surface, were completely destroyed, and 237 people were killed. A farmer on the mountain slope survived by holding the trunk of a nearby tree and traveling with it for ~960 m without injury (Fig. 2).

Figure 1.

Location of Sale Mountain landslide.

Figure 1.

Location of Sale Mountain landslide.

Figure 2.

Tree, back tilted toward main scarp, that carried person safely for ~960 m.

Figure 2.

Tree, back tilted toward main scarp, that carried person safely for ~960 m.

The length of the landslide is 1600 m, the width is 1100 m, and the area is ~1.3 km2. According to geophysical profiling and drilling by the First Hydrogeology and Engineering Geology Team of Gansu Province, the maximum and the average depth of the landslide debris are 70 m and 24 m, respectively, and the landslide volume is estimated to be 30 × 106 m3 (Liu and Wu, 1987; Wu and Wang, 1989). Although its volume is large and its travel distance long, the entire sliding process lasted less than 1 min (Zha, 1983). The velocity of movement was thus extremely rapid, estimated as ~20 m/s. No trigger for this huge landslide is evident. The Loess Plateau of China is a semiarid region and spring is a dry season. No rainfall or earthquake was associated with the sudden catastrophe.

The landslide attracted the attention of numerous researchers, and investigations were carried out. In this chapter we review the results of these investigations and some published papers as they relate to the occurrence, mechanisms, and mobility of the Sale Mountain landslide.

PHYSICAL GEOGRAPHY AND GEOLOGIC BACKGROUND

Topographic features

Sale Mountain is located at the southwest part of the Loess Plateau in northwest China. The Yellow River is to the north and flows eastward. Tributaries of the Yellow River that originate from West Qinling in the south of the plateau dissect this part of the plateau into hilly terrain with pinnate drainage and parallel mountain ridges. The Baxie River is a third-order tributary of the Yellow River. It is to the south of Sale Mountain and flows eastward. The elevation of the mountain ridge varies from 2300 to 2400 m, and the elevation of the river bed below Sale Mountain is 1962 m. The north slope of the mountain ridge is gentle, having a slope angle that varies from 13° to 17°This slope was cultivated into a terraced field (Fig. 3). The south slope is steep, especially above 2100 m elevation, where the slope angle varies from 30° to 35°. The ground surface here is barren, and has almost no vegetation cover. A series of vertical sided gullies are cut into this slope, and two of these gullies later formed the lateral boundaries of the landslide (Figs. 3 and 4; Zhang, 1986).

Figure 3.

Stereopair of Sale Mountain and Baxie River before occurrence of landslide. Photo was taken on June 1, 1979. SM is Sale Mountain; S is Sale Village; X is Xinzhuang Village; K is Kushun Village. Gansu Government photos 05449 and 05450.

Figure 3.

Stereopair of Sale Mountain and Baxie River before occurrence of landslide. Photo was taken on June 1, 1979. SM is Sale Mountain; S is Sale Village; X is Xinzhuang Village; K is Kushun Village. Gansu Government photos 05449 and 05450.

Figure 4.

Topographic map of Sale Mountain area showing topography before landslide, location of villages, outline of source area, and debris and preslide geomorphology of Baxie valley. Key: 1, spring; 2, outline of landslide; 3, scarp; 4, buildings (from Zhang Chengqi, 1986).

Figure 4.

Topographic map of Sale Mountain area showing topography before landslide, location of villages, outline of source area, and debris and preslide geomorphology of Baxie valley. Key: 1, spring; 2, outline of landslide; 3, scarp; 4, buildings (from Zhang Chengqi, 1986).

Although the Baxie River is in a broad valley (Fig. 5), it is practically a small stream. Its discharge may reach a maximum of 40 m-Vs during the flood period, but during times of low water, it drops drastically, to only 0.2 m3/s. Below Sale Mountain, the present channel of the Baxie River is artificial; it is cut into the second terrace level of the south (right) bank. The abandoned old channel is in the middle of the valley floor. This means that, before sliding, the north (left) bank was never undercut by the stream. There are three levels of river terraces within the valley. The first level is the widest, 600–800 m. The surface is covered by 10–15 m of alluvial sand and gravel. After sliding the alluvial sand and gravel was buried under the displaced mass.

Figure 5.

Geologic geomorphologic section of Sale Mountain and Baxie River before sliding. Key: 1, Pliocene mudstone; 2, middle Pleistocene Lishi Loess; 3, late Pleistocene Malan Loess; 4, alluvial loess; 5, surface of rupture of ancient landslide: 6, surface of rupture of old landslide; 7, present surface of rupture; 8, secondary fracture; 9, alluvial sand and gravel; I, II, III, first, second, and third river terraces, respectively; flood plain is absent in this section; IV, old landslide terrace (from Zhang and Wang, 1984).

Figure 5.

Geologic geomorphologic section of Sale Mountain and Baxie River before sliding. Key: 1, Pliocene mudstone; 2, middle Pleistocene Lishi Loess; 3, late Pleistocene Malan Loess; 4, alluvial loess; 5, surface of rupture of ancient landslide: 6, surface of rupture of old landslide; 7, present surface of rupture; 8, secondary fracture; 9, alluvial sand and gravel; I, II, III, first, second, and third river terraces, respectively; flood plain is absent in this section; IV, old landslide terrace (from Zhang and Wang, 1984).

The level of the second terrace is 20 m above river level. Most of the villages are situated on this terrace. At its front margin there is a small cliff facing the river (Fig. 3). The toe of the Sale Mountain landslide rupture surface cropped out just at the foot of this cliff. The third terrace level here is an old landslide buried under the eolian late Pleistocene Malan Loess and alluvial loess. Above the third terrace level there is another ancient landslide terrace, also covered with Malan Loess. The existence of these landslide deposits indicates that, in the long history of the downcutting of the Baxie River, landslides occurred frequently on the left bank in the middle and late Pleistocene.

Climatic features

The climate of this area is semiarid. The average annual precipitation, including rainfall in summer and autumn and snowfall in winter and early spring, is 485 mm. The rainy season is July-September, in which ~80% of the annual precipitation occurs, mostly in the form of rainstorms. Snowfall usually occurs from late October to early February. The curviplanar fracture first appeared behind the peak of Sale Mountain in 1979, which was a very wet year with an annual precipitation of 650 mm. The winter of 1982 was also wetter than usual, the total precipitation from October through December reaching 66 mm. Thus, in the thawing season of the next year, i.e., before March 1983, more thawing water would have infiltrated fractures and joints in loess and the rock mass composing the south slope of Sale Mountain.

The annual precipitation from 1979 through 1982 is shown in Figure 6.

Figure 6.

Monthly precipitation (including rainfall and snowfall) from 1979 to 1982.

Figure 6.

Monthly precipitation (including rainfall and snowfall) from 1979 to 1982.

Geologic features

Sale Mountain is chiefly composed of flat-lying lacustrine mudstones and siltstones of Pliocene age, covered with ~120-m-thick eolian loess of Pleistocene age (Fig. 4). During the Pliocene, the region was a fault-bounded basin, in which 600–700 m of sediments were deposited. These sediments form the Lingxia Group. The group is subdivided into four units that differ in lithology. The lower two units are composed chiefly of conglomerates and sandstones, and the upper two units are composed chiefly of siltstones and mudstones. In the Sale Mountain area, only the upper two units crop out. The stratigraphic descriptions of the units forming the south slope of Sale Mountain are as follows (Liu and Wu, 1987; Wu and Wang, 1989; Fig. 5).

Third Section of Lingxia Group (NL-32)

This unit forms the base of the Sale Mountain section, and is composed predominantly of purplish-red silty mudstone. Its upper middle part is intercalated with layers of grayish silty mudstone and lenses of yellow-green gravel-bearing sandstone. Three or four layers of gray to white marl are usually present. The base of this unit is conglomerate.

Fourth Section of Lingxia Group (NL-42)

This section is composed of pale red clayey siltstones intercalated with silty mudstones, with sandstones and conglomerates at its base; the sandstones are usually cross-bedded. Some early researchers of this landslide misinterpreted this group as lithified loess of early Pleistocene age, so the loess in their geological sections is as thick as 200 m; they suggested that this slide was a loess landslide (Zha, 1983).

Lishi Loess (Q2)

This loess is pale brownish-yellow and dense, 50–70 m thick, and contains calcareous concretions in its lower part. Vertical joints are well developed.

Malan Loess (Q23)

This pale yellow loose porous loess is 15–55 m thick, and exhibits well-developed vertical joints.

The geologic structure of Sale Mountain is simple (Fig. 5); no folds or faults are present. The Neogene strata are more or less flat lying, and dip <5° south or north. Very gentle east-west-trending flexures are present. Two sets of joints trending north-northwest and nearly east-west are predominant: The east-west set strikes from 82°NE to 286°NW and dips 73°–86°S. This attitude is in accordance with that of the main scarp of the Sale Mountain landslide.

Mechanical properties of loess and rocks involved in sliding

All of the preceding strata were involved in the sliding. Samples were taken and laboratory tests were performed: the mechanical properties obtained are listed in Table 1 (Liu and Wu, 1987).

Table 1.

Physico-Mechanical Properties of Loess and Mudstone (From LIU AND WU, 1987)

LithologySymbolPorosityNatural densityPeak cohesionResidual cohesionFriction angle
n(%)ρ(KN/m2)Cp(Kpa)Cr (Kpa)φ(°)
Malan LoessQ2347.3215.3744.3928.13
Lishi LoessQ239.7318.39154.6430.82
SiltstoneNL-4234.1420.54210.7034.39
MudstoneNL-4229.0220.15639.90122.5029.25
MudstoneNL-3221.66687.96397.8826.00
LithologySymbolPorosityNatural densityPeak cohesionResidual cohesionFriction angle
n(%)ρ(KN/m2)Cp(Kpa)Cr (Kpa)φ(°)
Malan LoessQ2347.3215.3744.3928.13
Lishi LoessQ239.7318.39154.6430.82
SiltstoneNL-4234.1420.54210.7034.39
MudstoneNL-4229.0220.15639.90122.5029.25
MudstoneNL-3221.66687.96397.8826.00

Stress-displacement curves, derived from direct shear tests of Lishi Loess and Pliocene mudstone and siltstone, are shown in Figure 7. From these curves it can be seen that the preliminary stages of all these curves approximate to straight lines with very high gradients, especially when the normal stress is low. Their yield strengths are attained after a shear deformation of only 2–3 mm, indicating that, under low confining pressures, they behave like brittle materials.

Figure 7.

Stress-displacement curves derived from direct shear tests on intact block samples of loess and mudstone, test on Pliocene material parallel to bedding. A: Lishi loess. B: Fourth unit of Lingxia group. C, D: Third unit of Lingxia group. s is normal stress; units of s are KPa for A and MPa for B, C, and D.

Figure 7.

Stress-displacement curves derived from direct shear tests on intact block samples of loess and mudstone, test on Pliocene material parallel to bedding. A: Lishi loess. B: Fourth unit of Lingxia group. C, D: Third unit of Lingxia group. s is normal stress; units of s are KPa for A and MPa for B, C, and D.

Another feature of the materials is that the difference between the peak and residual strength is high. The average ratio (Bishop's brittleness index) between these two strengths is 1.92. It means that when the rock mass fails suddenly, the strength decrease may be as high as 48%.

The uniaxial compressive strength of the mudstone with natural moisture content is as high as 20 MPa, but when saturated with water the strength drops drastically. The average saturated value is only 2.2 MPa. These results illustrate clearly the decrease in strength of the mudstone when saturated with water.

MORPHOLOGY AND STRUCTURE OF THE LANDSLIDE

A panoramic photograph of the Sale Mountain landslide is shown in Figure 8. The striking features are its conspicuous high, steep main scarp, its hummocky debris-accumulation zone with a low distal rim, and its long travel distance.

Figure 8.

Panoramic photograph of Sale Mountain landslide taken one month after slide; view is from south to north.

Figure 8.

Panoramic photograph of Sale Mountain landslide taken one month after slide; view is from south to north.

A vertical aerial photograph, geomorphologic map (Liu and Wu, 1987), and geological section (Zhang and Wang, 1984) of the landslide are shown in Figures 9, 10, and 11, respectively.

Figure 9.

Vertical air photograph of landslide taken two weeks after sliding. Gansu Government photo.

Figure 9.

Vertical air photograph of landslide taken two weeks after sliding. Gansu Government photo.

Figure 10.

Geomorphologic map of Sale Mountain landslide. Key: 1, main scarp; 2, outline of displaced mass; 3, toe of rupture surface; 4, depressions; 5, hillocks; 6, minor scarps; 7, hummocks; 8, grooves, mostly longitudinal; 9, mounds; 10, boundary line of different transverse parts; 11, boundary line of different longitudinal lobes (from Liu and Wu, 1987).

Figure 10.

Geomorphologic map of Sale Mountain landslide. Key: 1, main scarp; 2, outline of displaced mass; 3, toe of rupture surface; 4, depressions; 5, hillocks; 6, minor scarps; 7, hummocks; 8, grooves, mostly longitudinal; 9, mounds; 10, boundary line of different transverse parts; 11, boundary line of different longitudinal lobes (from Liu and Wu, 1987).

Figure 11.

Geological section of landslide. Key: 1, undisturbed Pliocene mudstone; 2, Lishi Loess; 3, Malan Loess; 4, displaced rock mass; 5, debris accumulation; 6, surface of rupture; 7, surface of separation; 8, thrust fault; 9, normal fault; 10, talus formed by falls from main scarp; 11, original profile (from Zhang and Wang, 1984).

Figure 11.

Geological section of landslide. Key: 1, undisturbed Pliocene mudstone; 2, Lishi Loess; 3, Malan Loess; 4, displaced rock mass; 5, debris accumulation; 6, surface of rupture; 7, surface of separation; 8, thrust fault; 9, normal fault; 10, talus formed by falls from main scarp; 11, original profile (from Zhang and Wang, 1984).

The plan view of the landslide is glove-like (Figs. 9 and 10) with a maximum length of 1630 m and a maximum width of 1100 m.

The main scarp is 750 m wide and 220–240 m high. The slope angles near its top and near its base are 70° and 45°, respectively; the average slope angle is ~55°. However, this slope angle is not the original one just after sliding. After the slide took place a series of falls occurred from the head scarp. The first and largest one occurred just one minute after sliding, and the tremors when the rock mass hit the ground were recorded by a nearby seismograph (Zha, 1983). These falls resulted in the retreat of the head scaip, and the talus slope formed by the falling fragments of rock and loess buried the foot of the scaip. The slope angle of the main scarp was thus reduced drastically. It can be deduced that before the occurrence of these falls the average slope angle of the main scarp might have been as high as 70°, with a nearly vertical cliff near the top of the scarp (Fig. 11). Apparently, this steep rear portion of the rupture surface was developed along east-west-trending stress relief joint (cf. Barton, 1984).

In the longitudinal direction, the displaced mass can be divided into two parts that differ in morphology and structure, and consequently in movement characteristics.

The surface morphology of the rear part (north part) consists of east-west-trending hillocks separated by elongate depressions. Immediately below the main scarp is the main depression of the landslide, ~250–300 m long and its width 50–60 m wide. The elevation of the bottom of the depression is ~2020 m. In front of the main depression is an elongate hillock. The elevation at its top is 2080 m and it is composed of loess-capped red siltstones and mudstones. The bedding planes of these strata dip northward at ~10°; this back tilting may be interpreted as evidence of rotational sliding along a curved surface of rupture. There are minor scarps in front of this hillock. The hillocks and depressions indicate that this part of the debris is intact slide blocks.

The vast front part of the debris to the south of the hillocks and depressions consists of hummocky terrain consisting of a series of raised transverse ridges, as high as 5–10 m, arranged in a subparallel pattern. There are shallow troughs between the ridges, and some scattered hummocks or raised mounds, especially at the margin of the slide debris. Some longitudinal grooves separate this part into three longitudinal lobes, the east, central, and west lobes. In this front part, the original ground surface was covered by mudstone and siltstone debris and disintegrated loess. The thickness of the debris cover varies; in the east lobe, it is from 26.5 m to 50 m, and is thicker than that of the other lobes (Liu and Wu, 1987). The debris cover is 16–31 m on the central lobe and 8–32 m on the west lobe. The original sand and gravel bed on the first river terrace level was buried, nearly undisturbed, under the debris. This may serve as evidence to indicate that the elevation of the toe of the rupture surface was higher than that of the first terrace level, which appears to crop out just underneath the second terrace. The fact that within such a vast area the undisturbed sand and gravel being buried under debris accumulation, as if being inundated by a debris stream, suggests that the mode of debris movement in this part of the slide is not sliding, but flowing. Some small reverse faults can be found near the front margin of the debris, but in its interior only east-west-trending normal faults can be found (Fig. 11). The stress state within the debris may be deduced from these faults. Tension is predominant in the interior of the debris, but near its front margin, when the advance of the flowing debris is hindered, compression becomes predominant. In places where some obstruction was encountered, the flowing debris was split and advanced along separate routes, forming lobes, and the outline of the resulting deposit is therefore glove-like in plan (Wang et al., 1988).

According to data collected from field investigations, geophysical explorations, and drilling, the rupture surface of the landslide is chair shaped, with a high and steep stress-relief-joint-controlled rear portion and a nearly horizontal bedding-plane-controlled sole portion. These two portions are connected by a gentle curve, not a sharp-radius curvature (cf. Barton, 1984). This rupture surface is illustrated in Figure 11. Its upper part cut through ~120 m of loess, and the lower part of this unit cut across the bedding of the flat- lying Pliocene mudstone. Its sole portion developed along a bedding plane in the mudstone.

MECHANISM OF THE LANDSLIDE

Mechanism of rock-mass deformation on slope

As shown in Figure 11, only the lower part of the rupture surface developed along a preexisting discontinuity, i.e., a bedding plane. The other parts of the rupture surface cut through flat-lying strata, including both loose porous loess and dense loess ~120 m thick and the underlying ~200 m of mudstones and siltstones. It seems remarkable that the rupture surface could cut through such a thick series of sedimentary strata. It is thought unlikely that the entire rupture surface could develop instantaneously, and that a step-by-step progressive process involving a long-term preparatory stage is necessary for the landslide to have occurred. In this stage, certain epigenetic processes were probably involved to promote the generation and propagation of fractures, movement along preexisting discontinuities, and the final linking of these fractures with preexisting discontinuities to form a continuous surface of rupture. Once the continuous surface formed, the slope failed suddenly. Therefore, the preparatory stage is essentially a stage of rock-mass deformation in the slope.

It is hypothesized that, as a result of river downcutting and the release of the original stress, the stress and strain field in the slope also readjusted. Near the surface of the slope, the maximum principal stress (s1) increased in magnitude and reoriented to a direction nearly parallel with the slope surface. The minimum principal stress (s3) decreased to a negligible magnitude and reoriented to a direction perpendicular to the surface, resulting in a horizontal stress-relief condition and a stress differentiation and a shear stress concentration zone near the foot of the slope. Under these conditions the rock mass in the slope was subject to creep under the action of gravitational force. Hundreds of thousands of years must have elapsed since the formation of the valley slope, time enough for time-dependent gravitational creep to play a role in the deformation of rock mass in the valley slope. The deformation gave rise to movement along preexisting discontinuities and to the development and propagation of stress relief joints, and resulted in the redistributing and reconcentrating of stress in the slope. An extremely unequal distribution of shear stress along the resisting elements, such as rock bridges between discontinuities, would occur. In the zone of unfavorable stress concentration, local fracturing or shearing might start and, in turn, increase the stress reconcentration and accelerate the creep process. Thus, gradual progressive failure could occur without the aid of other external processes. As the fractures extended, the unfractured resisting elements diminished in length, while the concentration of shear stress along these elements increased. The ultimate failure occurred abruptly, and a vast volume of rock mass slid down rapidly with strong destructive forces. This is but a summary of the mechanism of rock-mass deformation leading to slope failure. The composite mechanism of the Sale Mountain landslide is extremely slow sliding, tensile fracturing, and final shearing, analyzed in the following.

Extremely slow sliding, tension fracturing, shearing mechanism of Sale Mountain landslide

The most conspicuous manifestation of deformation on the slope before the 1983 slide is the appearance of an curviplanar fracture behind the peak of Sale Mountain (Zhang and Wang, 1984; Wang et al., 1988). It was first observed in March 1979 by local people who went up the mountain to cultivate their cropland. The width of the fracture was ~25 cm at that time. Thereafter, the fracture developed continuously, widening in width and extending in length. These were developments visible at the ground surface. At the same time, there must have been invisible developments beneath the ground surface, extending to depth within the slope. In the winter of 1982, prior to the slide, the fracture was observed to be 180 cm wide. Based on the accounts of two local people, the development of the fracture with time is illustrated as Figure 12 (Zhang and Wang, 1984). Based on experience with other well-investigated catastrophic landslide showing similar mechanisms, such as the Zana landslide in Longyang Gorge reservoir (Liu et al., 1984), the fracture may have extended to a depth equal to H/2, H being the total height of the landslide. For the Sale Mountain landslide H = 320 m; the fracture might have extended to a depth of ~160 m, which is in good agreement with the height of the nearly vertical main scarp after sliding. No doubt, the cause of the fracture widening is by tension stress concentration, but how the stress arose needs to be investigated further.

Figure 12.

Width-time relationship for development of tension fracture on top of Sale Mountain. Key: 1, steady fracturing; 2, accelerating fracturing; 3, fracture closing due to rotational displacement; 4, sliding (from Zhang and Wang, 1984).

Figure 12.

Width-time relationship for development of tension fracture on top of Sale Mountain. Key: 1, steady fracturing; 2, accelerating fracturing; 3, fracture closing due to rotational displacement; 4, sliding (from Zhang and Wang, 1984).

Li and Huang (1989) simulated the gravitational creep deformation of the rock mass in the south-facing slope of Sale Mountain by the finite element method (Fig. 13). They simulated the stress distribution and the occurrence and propagation of a plastic zone (Li and Huang defined this zone as an area where the ratio between shear stress and shear resistance equals 1.0) before and after the development of the rear tension fracture. The result of numerical simulation (Fig. 13) indicated that (1) before the development of the rear tension fracture, a subhorizontal plastic zone occurred underneath the foot of the slope (Fig. 13C); this zone coincides well with the subhorizontal sole portion of the rupture surface of the landslide and its length is ~170 m; (2) after the rear tension fracture occurred and propagated to point A in Figure 13A, the plastic zone extended in length from 170 to 200 m, and another plastic zone developed near the base of the tension fracture. These results of numerical simulation may be interpreted as the following. (1) Before the development of the rear tension fracture, extremely slow sliding movement along a mudstone bedding plane in the foot of the slope occurred due to a shear stress concentration that overcame shear resistance along the bedding plane. (2) The development of rear tension fracture promoted the extremely slow sliding along bedding plane in the foot of the slope and a new shear stress concentration zone appeared at the base of the tension fracture, thus promoting the downward propagation of the fracture. Downward propagation of the tension fracture and inward propagation of the plastic zone reduced the length of the middle locked-up segment (resisting element). At the same time, shear stress concentration along this segment increased progressively.

Figure 13.

Numerical simulation results by finite element method (FEM). A: Geologic section. B: FEM mesh. C: Contour line of ratio between shear stress and shear resistance before development of top fracture. D: Contour line of ratio between shear stress and shear resistance after top fracture extending to point A (in A) (from Li and Huang, 1989).

Figure 13.

Numerical simulation results by finite element method (FEM). A: Geologic section. B: FEM mesh. C: Contour line of ratio between shear stress and shear resistance before development of top fracture. D: Contour line of ratio between shear stress and shear resistance after top fracture extending to point A (in A) (from Li and Huang, 1989).

The Sale Mountain landslide was accompanied by a loud thunderous sound heard by some eyewitnesses. It is suggested that the sound was the acoustic emission effect of the brittle shearing failure of the locked-up segment. Samples collected from the middle part of the rupture surface were observed under optical and scanning electron microscope (Ai et al., 1987; Fang and Ai, 1989). It was found that the crushed shear zone is only 2 mm thick. Muscovite flakes oriented parallel with the surface (Fig. 14) and some glass beads, with wavy extinction under crossed nicois, were present in the shear zone. Based on these phenomena, it is deduced that the temperature of the rupture surface during shearing failure and sliding was as high as 500–600 °C. Under these stress and temperature conditions, kaolinite and montmorillonite in the mudstones were dehydrated and recrystallized and transformed into muscovite flakes; as the recrystailization occurred under shear stress, the flakes were oriented parallel to the direction of shear stress. Some quartz grains were melted to form glass beads with wavy extinctions. The thunderous sound, melted quartz, and recrystallized minerals all might serve as evidence of the final shearing failure of the rock mass composing the south-facing slope of Sale Mountain.

Figure 14.

Recrystallized muscovite flakes oriented parallel to shear stress as observed under scanning electron microscope. Samples were taken from rupture surface near bottom of Sale Mountain landslide (from Ai et al., 1987).

Figure 14.

Recrystallized muscovite flakes oriented parallel to shear stress as observed under scanning electron microscope. Samples were taken from rupture surface near bottom of Sale Mountain landslide (from Ai et al., 1987).

The Sale Mountain landslide was formed through a composite mechanism of extremely slow sliding, tension fracturing, and final shearing. Before final failure, the slope had undergone a long-term time-dependent deformation that could be divided into three stages: elastic rebounding due to lateral stress relief and gravitational creep sliding along the bedding plane near the foot of the slope; tension fracturing on the top of the slope; and final shearing of the locked-up middle part, leading to the linking of the surface of rupture (Liu and Wu, 1987; Zhang et al., 1992; Fig. 15).

Figure 15.

Stages in deformation-failure process. A: Stage of unloading elastic rebound and extremely slow sliding. B: Stage of tension fracturing. C: Stage of final shearing. Modified from Liu and Wu, 1987. Materials as in Figure 5.

Figure 15.

Stages in deformation-failure process. A: Stage of unloading elastic rebound and extremely slow sliding. B: Stage of tension fracturing. C: Stage of final shearing. Modified from Liu and Wu, 1987. Materials as in Figure 5.

KINEMATICS AND DYNAMICS OF THE LANDSLIDE

Displacement vectors

The positions of some objects, such as big trees and farmers' houses, before and after sliding, were determined precisely in the field investigation. The displacement vectors of these objects (Liu and Wu, 1987; Wang et al., 1988; Wu and Wang, 1989) are shown in Figure 16. By analysis of Figure 16, we can derive the following kinematic characteristics of the landslide.

Figure 16.

Displacement vectors of some objects on Sale Mountain landslide. Key: 1, outline of displaced mass; 2, boundary line of different transverse parts; 3, rear part of mass sliding; 4, main debris-flow accumulation; 5, eastern minor debris-flow accumulation; 6, depressions; 7, hillocks; 8, shear fissure en echelon arranged; 9, original house site; 10, house site after sliding; 11, original tree site; 12, tree site after sliding; 13, tree that carried person ~960 m (from Wu and Wang, 1989).

Figure 16.

Displacement vectors of some objects on Sale Mountain landslide. Key: 1, outline of displaced mass; 2, boundary line of different transverse parts; 3, rear part of mass sliding; 4, main debris-flow accumulation; 5, eastern minor debris-flow accumulation; 6, depressions; 7, hillocks; 8, shear fissure en echelon arranged; 9, original house site; 10, house site after sliding; 11, original tree site; 12, tree site after sliding; 13, tree that carried person ~960 m (from Wu and Wang, 1989).

The upper hillock part and the lower hummocky part are entirely different in their kinematic characteristics. The vertical components of displacements of the upper part are significant, usually larger than 100 m. The horizontal components are relatively small, especially near the rear part, where the ratios between the horizontal and vertical components are <1.5. As the horizontal components increase toward the front part of the debris, these ratios increase to ~10. In the lower part, vertical components are small, usually <50 m. However, the horizontal components are relatively large; the ratios between these two components rise to between 30 and 40. The displacements reach a maximum value near the central line of the landslide, and diminish toward lateral margins.

From the geometry of the rupture surface and the difference between displacement vectors of the upper part and lower part, it is apparent that kinematically the Sale Mountain landslide is a compound landslide (Hutchinson, 1988; Kovari, 1988); its bounding rupture surface was partly rotational (the upper part) and partly translational (the lower part). According to the definitions of Hutchinson (1988, p. 10), compound landslides are locked in place as a result of their slip surface geometry and can move only when the slip mass is transformed into a kinematically admissible mechanism by the development of internal displacements and shear. In this case, the internal displacements involve extremely slow sliding along the bedding plane, the tension fracturing on the top of the slope, and shear off step by step of the resisting elements in the middle, locked-up part.

In the lateral margin area, owing to the lateral resistance to the flowing debris, the lower part of the landslide mass moved rotationally in plan, as evidenced by a house yard originally located at the eastern margin that was left-laterally rotated, and the formation of some peculiar landforms, which may be called circular cones with slickensides developed on their surfaces (Fig. 17). The slickensides show left-lateral (counterclockwise) and right-lateral (clockwise) rotation, depending on whether they were located at eastern (left wing) or western (right wing) margins, respectively, of the debris flow (Wang et al., 1988).

Figure 17.

Photograph of circular mudstone cone with slickensides developed on its surface. Cone is surrounded by slide debris.

Figure 17.

Photograph of circular mudstone cone with slickensides developed on its surface. Cone is surrounded by slide debris.

Velocity of the movement

The ground tremor generated by the Sale Mountain landslide was recorded by seismographs at the nearby seismic stations at Lingxia, Lanzhou, and Yongjing. Through seismogram analysis, Zha (1983) determined that the propagation of the tremor was from north to south, the magnitude was 1.4, and the duration of the landslide was 55 s. Another less strong tremor that propagated from east to west was recorded 65.5 s afterward. This second tremor was caused by falls from the main scarp immediately after sliding.

With the available data of displacement vectors and time duration of the sliding, it is possible to calculate the average velocity of the movement. The maximum displacement vector is 1090 m, so the maximum average velocity near the central axial line is 19.8 m/s; i.e., the velocity of the landslide is extremely rapid. The velocities decrease toward the two lateral margins. According to the displacement of three villages situated near the western margin, central line, and eastern margin of the slide, the average velocities are 7.64 m/s, 12–13 m/s, and 8 m/s, respectively (Table 2; Zha, 1983).

Table 2.

Calculation Velocities From the Displacement of Villages

Name of villagePosition relative to the landslideDirection of displacementHorizontal distance of displacement (m)Average velocities of movement (m/s)
SaleNear to the western marginS9°W4207.64
XinzhuangNear to the central lineS5°W66012.0
S22°E74013.45
KushunNear to the western marginS55°E4408.0
Name of villagePosition relative to the landslideDirection of displacementHorizontal distance of displacement (m)Average velocities of movement (m/s)
SaleNear to the western marginS9°W4207.64
XinzhuangNear to the central lineS5°W66012.0
S22°E74013.45
KushunNear to the western marginS55°E4408.0

Mobility

The angle whose tangent is the ratio between the maximum height (Hmax) and maximum horizontal distance (Lmax), Fahrböschung as termed by Heim (1932) and apparent friction angle as termed by Hsii (1975), is only dependent on the kinetic energy and coefficient of kinetic friction. Hence it can express the mobility of the landslide very well. The smaller the ratio, the greater the mobility. Fang and Zhang (1988) designated this ratio as characteristic value of mobility, indicated by m. For the Sale Mountain landslide, the Hmax = 320 m and the Lmax = 1600 m, hence m = 0.2, and the Fahrböschung is 11°.

According to Hsü (1975), the excessive travel distance is related to landslide volume (Fig. 18). From Figure 18 it can be seen that the Sale Mountain landslide has mobility characteristics of a more mobile sturzstrom.

Figure 18.

Mobility of sturzstroms (from Hsu, 1975).

Figure 18.

Mobility of sturzstroms (from Hsu, 1975).

Various processes have been proposed to explain the low apparent friction, such as that involving an air cushion at the base of the debris sheet (Shreve, 1968), acoustic fluidization (Melosh, 1987), mechanical fluidization (Davies, 1982), and the undrained loading involving undrained shear-induced high pore pressure (Sassa, 1988; Sassa et al., 1992). Fang and Zhang (1988) proposed the momentum transmission by collision among flowing debris. The momentum transmission is closely related to the total energy of a landslide, and total energy is determined by the volume of the sliding mass (the bulk weight may be considered as constant) and the dropping height of the center of mass during the movement. The characteristic value of mobility, m, and the total energy, E (in ergs), of 63 known catastrophic landslides without distinct obstructions in their paths of movement were calculated. After regression analysis, a regression equation is derived: 

formula

Its correlation coefficient is 0.96 and standard deviation is only 0.003. The regression curve is illustrated in Figure 19.

Figure 19.

Correlation curve of characteristic value of mobility (m) and total energy (E) (from Fang and Zhang, 1988).

Figure 19.

Correlation curve of characteristic value of mobility (m) and total energy (E) (from Fang and Zhang, 1988).

Conclusions

The following conclusions can be drawn about the Sale Mountain landslide.

1. The 1983 Sale Mountain landslide occurred on the high and steep south-facing slope of Sale Mountain, which is composed of flat-lying Pliocene siltstones and mudstones covered with ~120 m of Pleistocene eolian loess. The volume of loess involved in sliding is less than one-third of the total volume of the sliding mass, so it is not a loess landslide, but a loess-covered soft-rock landslide.

2. The bounding rupture surface is chair shaped, with a straight and steep rear portion controlled by stress-relief joints and a straight and subhorizontal bedding-plane-controlled base, but the middle portion connecting the two and cutting across bedding is a gentle curve.

3. Kinematically it is a compound landslide. The movement of the frontal part is translational debris flowing and that of the rear part is rotational block sliding.

4. Although it occurred suddenly, before its occurrence there was a long-term preparatory stage. In this stage, gravitational extremely slow sliding underneath the foot of the slope and tension fracturing at the top of the mountain were important processes leading to the final abrupt failure. It is suggested by most researchers that the mechanism of the landslide is progressive failure, beginning with mutually promoting extremely slow sliding and tension fracturing, and ending with step-by-step shearing through of the resisting elements in the middle locked-up portion.

5. The velocity of movement was extremely rapid. The average velocity was ~20 m/s. The characteristic value of mobility is 0.2, and Fahrböschung is 11°; it traveled excessively.

References Cited

Nanshan
,
Ai
Xiaomin
,
Fang
Tiande
,
Miao
,
1987
,
Stress induced erosion and its relation to landslides
, in
Proceedings of the China Japan Field Workshop on Landslide
:
Xian-Lanzhou, China
, p.
1
4
.
Barton
,
M.E.
,
1984
,
The preferred path of landslide shear surfaces in over-consolidated clays and soft rocks
, in
Proceedings, 4th International Symposium on Landslides
, Volume
3
,
September 16–21, 1984
:
Toronto, Canada
,
Canadian Geothechnical Society
 , p.
75
79
.
Davies
,
T.R.H.
,
1982
,
Spreading of rock avalanche debris by mechanical fluidization
:
Rock Mechanics
 , v.
15
, p.
9
24
.
Xiaomin
,
Fang
Nanshan
,
Ai
,
1989
,
The microfeature and mechanisms of Sale Shan landslide (in Chinese)
, in
Selected papers on landslides
 :
Chengdu
,
Sichuan Science and Technology Publishing House
, p.
168
173
.
Fang
,
Y.S.
,
1984
,
Preliminary study on the kinematic mechanism of catastrophic landslides and on the prediction of their velocities and travel distances
 
[MA thesis, in Chinese]
:
Chengdu, Sichuan, China
,
Chengdu College of Geology
.
Yushu
,
Fang
Zhuoyuan
,
Zhang
,
1988
,
Kinematic mechanism of catastrophic landslides and prediction of their velocities and travel distances
, in
Bonnard
,
Ch.
, ed.,
Landslides/Glissements de terrain, Proceedings of the 5th International Symposium on Landslides
:
Rotterdam
,
A.A. Balkema
, p.
125
128
.
Heim
,
A.
,
1932
,
Bergsturz und Menschenleben
 :
Zurich
,
Fretz und Wasmuth
,
218
p.
Hutchinson
,
J.N.
,
1988
,
General report; Morphological parameters of landslides in relation to geology and hydrogeology
, in
Bonnard
,
Ch.
, ed.,
Landslides—Glissements de terrain: Proceedings of the 5th International Symposium on Landslides
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
3
35
.
Hsü
,
K.J.
,
1975
,
Catastrophic debris streams, Sturzstroms, generated by rockfalls
:
Geological Society of America Bulletin
 , v.
86
, p.
129
140
.
Kovari
,
K.
,
1988
,
Methods of monitoring landslides
, in
Bonnard
,
Ch.
, ed.,
Landslides—Glissements de terrain: Proceedings of the 5fh International Symposium on Landslides
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
1421
1433
.
Yongyi
,
Li
Dating
,
Huang
,
1989
,
Finite element analysis for the effect of top cracking on slope stability
, in
Selected papers on landslides
 :
Chengdu, Sichuan Science and Technology Publishing House
, p.
271
277
(in Chinese).
Chengyu
,
Liu
Xingming
,
Wu
,
1987
,
On the deformational destruction of Sale Shan landslide and its mechanism of high slipping speed and sliding distance
, in
Proceedings of the China-Japan Field Workshop on Landslides
:
Xian-Lanzhou, China
, p.
19
24
.
Hanchao
,
Liu
Zhuoyuan
,
Zhang
Zuyin
,
Qing
,
1984
,
The mechanism of the landslide in overconsolidated clay beds near the Longyang Gorge damsite
, in
Proceedings, 4th International Symposium on Landslides
, Volume
2
:
Toronto, Canada
, p.
121
126
.
Melosh
,
H.J.
,
1987
,
The mechanics of large rock avalanches
, in
Costa
,
J.E.
Wieczorek
, eds.,
Debris flows/avalanches: Process, recognition, and mitigation
 :
Geological Society of America, Reviews in Engineering Geology
, v.
VII
, p.
41
49
.
Sassa
,
K.
,
1988
,
Geotechnical model for the motion of landslides
, in
Bonnard
,
Ch.
,
Landslides—Glissements de terrain: Proceedings of the 5th International Symposium on Landslides
:
Rotterdam, Netherlands
,
A.A. Balkema, Landslides
, v.
1
, p.
37
55
.
Sassa
,
K.
Fukuoka
,
H.
Lee
,
J.H.
Zhang
,
D.X.
,
1992
,
Measurement of the apparent friction angle during rapid loading by the high-speed high-stress ring shear apparatus: Interpretation of the relationship between landslide volume and the apparent friction during motion
, in
Bell
,
D.H.
,
Landslides: Proceedings of the 6th International Symposium on Landslides, Christchurch, New Zealand
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
545
552
.
Shreve
,
R.L.
,
1968
,
The Blackhawk landslides
:
Geological Society of America Special Paper 108
 , p.
1
47
.
Shitian
,
Wang
Zhuoyuan
,
Zhang
Zheng
,
Zhan
Hanchao
,
Liu
,
1988
,
On the characteristics and dynamics of the catastrophic Mount Sale landslide, Gansu, China
:
Journal of Chengdu College of Geology
 , v.
15
, no.
2
, p.
58
63
.
Weijiang
,
Wu
Shouying
,
Wang
,
1989
,
Mechanism of Sale Shan landslide (in Chinese)
, in
selected Papers on Landslides
 :
Chengdu
,
Sichuan Science and Technology Publishing House
, p.
184
189
.
Xiaogang
,
Zha
,
1983
,
Faulting is the foundation of the landslide at Sale Mountain, water and gravity of soil-body are the decisive factors
:
Bulletin of Soil and Water Conservation
 , no.
3
, p.
27
33
(in Chinese).
Chengqi
,
Zhang
,
1986
,
Analysis of the landslide at Sale Mountain
:
People's Yellow River
 , no.
3
, p.
41
45
(in Chinese).
Zhuoyuan
,
Zhang
Shitian
,
Wang
,
1984
,
On the prediction of the occurrence time and the velocity of a potential landslide
, in
Proceedings, 4th International Symposium on Landslides
, Volume
3
:
Toronto, Canada
, p.
145
146
.
Zhuoyuan
,
Zhang
Shitian
,
Wang
Lansheng
,
Wang
Hanchao
,
Liu
Runqiu
,
Huang
,
1992
,
Recent researches on the mechanism and geomechanic model of huge landslides in China
, in
Bell
,
D.H.
,
Landslides: Proceedings of the 6th International Symposium on Landslides, Christchurch, New Zealand
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
625
634
.

Figures & Tables

Figure 1.

Location of Sale Mountain landslide.

Figure 1.

Location of Sale Mountain landslide.

Figure 2.

Tree, back tilted toward main scarp, that carried person safely for ~960 m.

Figure 2.

Tree, back tilted toward main scarp, that carried person safely for ~960 m.

Figure 3.

Stereopair of Sale Mountain and Baxie River before occurrence of landslide. Photo was taken on June 1, 1979. SM is Sale Mountain; S is Sale Village; X is Xinzhuang Village; K is Kushun Village. Gansu Government photos 05449 and 05450.

Figure 3.

Stereopair of Sale Mountain and Baxie River before occurrence of landslide. Photo was taken on June 1, 1979. SM is Sale Mountain; S is Sale Village; X is Xinzhuang Village; K is Kushun Village. Gansu Government photos 05449 and 05450.

Figure 4.

Topographic map of Sale Mountain area showing topography before landslide, location of villages, outline of source area, and debris and preslide geomorphology of Baxie valley. Key: 1, spring; 2, outline of landslide; 3, scarp; 4, buildings (from Zhang Chengqi, 1986).

Figure 4.

Topographic map of Sale Mountain area showing topography before landslide, location of villages, outline of source area, and debris and preslide geomorphology of Baxie valley. Key: 1, spring; 2, outline of landslide; 3, scarp; 4, buildings (from Zhang Chengqi, 1986).

Figure 5.

Geologic geomorphologic section of Sale Mountain and Baxie River before sliding. Key: 1, Pliocene mudstone; 2, middle Pleistocene Lishi Loess; 3, late Pleistocene Malan Loess; 4, alluvial loess; 5, surface of rupture of ancient landslide: 6, surface of rupture of old landslide; 7, present surface of rupture; 8, secondary fracture; 9, alluvial sand and gravel; I, II, III, first, second, and third river terraces, respectively; flood plain is absent in this section; IV, old landslide terrace (from Zhang and Wang, 1984).

Figure 5.

Geologic geomorphologic section of Sale Mountain and Baxie River before sliding. Key: 1, Pliocene mudstone; 2, middle Pleistocene Lishi Loess; 3, late Pleistocene Malan Loess; 4, alluvial loess; 5, surface of rupture of ancient landslide: 6, surface of rupture of old landslide; 7, present surface of rupture; 8, secondary fracture; 9, alluvial sand and gravel; I, II, III, first, second, and third river terraces, respectively; flood plain is absent in this section; IV, old landslide terrace (from Zhang and Wang, 1984).

Figure 6.

Monthly precipitation (including rainfall and snowfall) from 1979 to 1982.

Figure 6.

Monthly precipitation (including rainfall and snowfall) from 1979 to 1982.

Figure 7.

Stress-displacement curves derived from direct shear tests on intact block samples of loess and mudstone, test on Pliocene material parallel to bedding. A: Lishi loess. B: Fourth unit of Lingxia group. C, D: Third unit of Lingxia group. s is normal stress; units of s are KPa for A and MPa for B, C, and D.

Figure 7.

Stress-displacement curves derived from direct shear tests on intact block samples of loess and mudstone, test on Pliocene material parallel to bedding. A: Lishi loess. B: Fourth unit of Lingxia group. C, D: Third unit of Lingxia group. s is normal stress; units of s are KPa for A and MPa for B, C, and D.

Figure 8.

Panoramic photograph of Sale Mountain landslide taken one month after slide; view is from south to north.

Figure 8.

Panoramic photograph of Sale Mountain landslide taken one month after slide; view is from south to north.

Figure 9.

Vertical air photograph of landslide taken two weeks after sliding. Gansu Government photo.

Figure 9.

Vertical air photograph of landslide taken two weeks after sliding. Gansu Government photo.

Figure 10.

Geomorphologic map of Sale Mountain landslide. Key: 1, main scarp; 2, outline of displaced mass; 3, toe of rupture surface; 4, depressions; 5, hillocks; 6, minor scarps; 7, hummocks; 8, grooves, mostly longitudinal; 9, mounds; 10, boundary line of different transverse parts; 11, boundary line of different longitudinal lobes (from Liu and Wu, 1987).

Figure 10.

Geomorphologic map of Sale Mountain landslide. Key: 1, main scarp; 2, outline of displaced mass; 3, toe of rupture surface; 4, depressions; 5, hillocks; 6, minor scarps; 7, hummocks; 8, grooves, mostly longitudinal; 9, mounds; 10, boundary line of different transverse parts; 11, boundary line of different longitudinal lobes (from Liu and Wu, 1987).

Figure 11.

Geological section of landslide. Key: 1, undisturbed Pliocene mudstone; 2, Lishi Loess; 3, Malan Loess; 4, displaced rock mass; 5, debris accumulation; 6, surface of rupture; 7, surface of separation; 8, thrust fault; 9, normal fault; 10, talus formed by falls from main scarp; 11, original profile (from Zhang and Wang, 1984).

Figure 11.

Geological section of landslide. Key: 1, undisturbed Pliocene mudstone; 2, Lishi Loess; 3, Malan Loess; 4, displaced rock mass; 5, debris accumulation; 6, surface of rupture; 7, surface of separation; 8, thrust fault; 9, normal fault; 10, talus formed by falls from main scarp; 11, original profile (from Zhang and Wang, 1984).

Figure 12.

Width-time relationship for development of tension fracture on top of Sale Mountain. Key: 1, steady fracturing; 2, accelerating fracturing; 3, fracture closing due to rotational displacement; 4, sliding (from Zhang and Wang, 1984).

Figure 12.

Width-time relationship for development of tension fracture on top of Sale Mountain. Key: 1, steady fracturing; 2, accelerating fracturing; 3, fracture closing due to rotational displacement; 4, sliding (from Zhang and Wang, 1984).

Figure 13.

Numerical simulation results by finite element method (FEM). A: Geologic section. B: FEM mesh. C: Contour line of ratio between shear stress and shear resistance before development of top fracture. D: Contour line of ratio between shear stress and shear resistance after top fracture extending to point A (in A) (from Li and Huang, 1989).

Figure 13.

Numerical simulation results by finite element method (FEM). A: Geologic section. B: FEM mesh. C: Contour line of ratio between shear stress and shear resistance before development of top fracture. D: Contour line of ratio between shear stress and shear resistance after top fracture extending to point A (in A) (from Li and Huang, 1989).

Figure 14.

Recrystallized muscovite flakes oriented parallel to shear stress as observed under scanning electron microscope. Samples were taken from rupture surface near bottom of Sale Mountain landslide (from Ai et al., 1987).

Figure 14.

Recrystallized muscovite flakes oriented parallel to shear stress as observed under scanning electron microscope. Samples were taken from rupture surface near bottom of Sale Mountain landslide (from Ai et al., 1987).

Figure 15.

Stages in deformation-failure process. A: Stage of unloading elastic rebound and extremely slow sliding. B: Stage of tension fracturing. C: Stage of final shearing. Modified from Liu and Wu, 1987. Materials as in Figure 5.

Figure 15.

Stages in deformation-failure process. A: Stage of unloading elastic rebound and extremely slow sliding. B: Stage of tension fracturing. C: Stage of final shearing. Modified from Liu and Wu, 1987. Materials as in Figure 5.

Figure 16.

Displacement vectors of some objects on Sale Mountain landslide. Key: 1, outline of displaced mass; 2, boundary line of different transverse parts; 3, rear part of mass sliding; 4, main debris-flow accumulation; 5, eastern minor debris-flow accumulation; 6, depressions; 7, hillocks; 8, shear fissure en echelon arranged; 9, original house site; 10, house site after sliding; 11, original tree site; 12, tree site after sliding; 13, tree that carried person ~960 m (from Wu and Wang, 1989).

Figure 16.

Displacement vectors of some objects on Sale Mountain landslide. Key: 1, outline of displaced mass; 2, boundary line of different transverse parts; 3, rear part of mass sliding; 4, main debris-flow accumulation; 5, eastern minor debris-flow accumulation; 6, depressions; 7, hillocks; 8, shear fissure en echelon arranged; 9, original house site; 10, house site after sliding; 11, original tree site; 12, tree site after sliding; 13, tree that carried person ~960 m (from Wu and Wang, 1989).

Figure 17.

Photograph of circular mudstone cone with slickensides developed on its surface. Cone is surrounded by slide debris.

Figure 17.

Photograph of circular mudstone cone with slickensides developed on its surface. Cone is surrounded by slide debris.

Figure 18.

Mobility of sturzstroms (from Hsu, 1975).

Figure 18.

Mobility of sturzstroms (from Hsu, 1975).

Figure 19.

Correlation curve of characteristic value of mobility (m) and total energy (E) (from Fang and Zhang, 1988).

Figure 19.

Correlation curve of characteristic value of mobility (m) and total energy (E) (from Fang and Zhang, 1988).

Table 1.

Physico-Mechanical Properties of Loess and Mudstone (From LIU AND WU, 1987)

LithologySymbolPorosityNatural densityPeak cohesionResidual cohesionFriction angle
n(%)ρ(KN/m2)Cp(Kpa)Cr (Kpa)φ(°)
Malan LoessQ2347.3215.3744.3928.13
Lishi LoessQ239.7318.39154.6430.82
SiltstoneNL-4234.1420.54210.7034.39
MudstoneNL-4229.0220.15639.90122.5029.25
MudstoneNL-3221.66687.96397.8826.00
LithologySymbolPorosityNatural densityPeak cohesionResidual cohesionFriction angle
n(%)ρ(KN/m2)Cp(Kpa)Cr (Kpa)φ(°)
Malan LoessQ2347.3215.3744.3928.13
Lishi LoessQ239.7318.39154.6430.82
SiltstoneNL-4234.1420.54210.7034.39
MudstoneNL-4229.0220.15639.90122.5029.25
MudstoneNL-3221.66687.96397.8826.00
Table 2.

Calculation Velocities From the Displacement of Villages

Name of villagePosition relative to the landslideDirection of displacementHorizontal distance of displacement (m)Average velocities of movement (m/s)
SaleNear to the western marginS9°W4207.64
XinzhuangNear to the central lineS5°W66012.0
S22°E74013.45
KushunNear to the western marginS55°E4408.0
Name of villagePosition relative to the landslideDirection of displacementHorizontal distance of displacement (m)Average velocities of movement (m/s)
SaleNear to the western marginS9°W4207.64
XinzhuangNear to the central lineS5°W66012.0
S22°E74013.45
KushunNear to the western marginS55°E4408.0

Contents

References

References Cited

Nanshan
,
Ai
Xiaomin
,
Fang
Tiande
,
Miao
,
1987
,
Stress induced erosion and its relation to landslides
, in
Proceedings of the China Japan Field Workshop on Landslide
:
Xian-Lanzhou, China
, p.
1
4
.
Barton
,
M.E.
,
1984
,
The preferred path of landslide shear surfaces in over-consolidated clays and soft rocks
, in
Proceedings, 4th International Symposium on Landslides
, Volume
3
,
September 16–21, 1984
:
Toronto, Canada
,
Canadian Geothechnical Society
 , p.
75
79
.
Davies
,
T.R.H.
,
1982
,
Spreading of rock avalanche debris by mechanical fluidization
:
Rock Mechanics
 , v.
15
, p.
9
24
.
Xiaomin
,
Fang
Nanshan
,
Ai
,
1989
,
The microfeature and mechanisms of Sale Shan landslide (in Chinese)
, in
Selected papers on landslides
 :
Chengdu
,
Sichuan Science and Technology Publishing House
, p.
168
173
.
Fang
,
Y.S.
,
1984
,
Preliminary study on the kinematic mechanism of catastrophic landslides and on the prediction of their velocities and travel distances
 
[MA thesis, in Chinese]
:
Chengdu, Sichuan, China
,
Chengdu College of Geology
.
Yushu
,
Fang
Zhuoyuan
,
Zhang
,
1988
,
Kinematic mechanism of catastrophic landslides and prediction of their velocities and travel distances
, in
Bonnard
,
Ch.
, ed.,
Landslides/Glissements de terrain, Proceedings of the 5th International Symposium on Landslides
:
Rotterdam
,
A.A. Balkema
, p.
125
128
.
Heim
,
A.
,
1932
,
Bergsturz und Menschenleben
 :
Zurich
,
Fretz und Wasmuth
,
218
p.
Hutchinson
,
J.N.
,
1988
,
General report; Morphological parameters of landslides in relation to geology and hydrogeology
, in
Bonnard
,
Ch.
, ed.,
Landslides—Glissements de terrain: Proceedings of the 5th International Symposium on Landslides
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
3
35
.
Hsü
,
K.J.
,
1975
,
Catastrophic debris streams, Sturzstroms, generated by rockfalls
:
Geological Society of America Bulletin
 , v.
86
, p.
129
140
.
Kovari
,
K.
,
1988
,
Methods of monitoring landslides
, in
Bonnard
,
Ch.
, ed.,
Landslides—Glissements de terrain: Proceedings of the 5fh International Symposium on Landslides
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
1421
1433
.
Yongyi
,
Li
Dating
,
Huang
,
1989
,
Finite element analysis for the effect of top cracking on slope stability
, in
Selected papers on landslides
 :
Chengdu, Sichuan Science and Technology Publishing House
, p.
271
277
(in Chinese).
Chengyu
,
Liu
Xingming
,
Wu
,
1987
,
On the deformational destruction of Sale Shan landslide and its mechanism of high slipping speed and sliding distance
, in
Proceedings of the China-Japan Field Workshop on Landslides
:
Xian-Lanzhou, China
, p.
19
24
.
Hanchao
,
Liu
Zhuoyuan
,
Zhang
Zuyin
,
Qing
,
1984
,
The mechanism of the landslide in overconsolidated clay beds near the Longyang Gorge damsite
, in
Proceedings, 4th International Symposium on Landslides
, Volume
2
:
Toronto, Canada
, p.
121
126
.
Melosh
,
H.J.
,
1987
,
The mechanics of large rock avalanches
, in
Costa
,
J.E.
Wieczorek
, eds.,
Debris flows/avalanches: Process, recognition, and mitigation
 :
Geological Society of America, Reviews in Engineering Geology
, v.
VII
, p.
41
49
.
Sassa
,
K.
,
1988
,
Geotechnical model for the motion of landslides
, in
Bonnard
,
Ch.
,
Landslides—Glissements de terrain: Proceedings of the 5th International Symposium on Landslides
:
Rotterdam, Netherlands
,
A.A. Balkema, Landslides
, v.
1
, p.
37
55
.
Sassa
,
K.
Fukuoka
,
H.
Lee
,
J.H.
Zhang
,
D.X.
,
1992
,
Measurement of the apparent friction angle during rapid loading by the high-speed high-stress ring shear apparatus: Interpretation of the relationship between landslide volume and the apparent friction during motion
, in
Bell
,
D.H.
,
Landslides: Proceedings of the 6th International Symposium on Landslides, Christchurch, New Zealand
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
545
552
.
Shreve
,
R.L.
,
1968
,
The Blackhawk landslides
:
Geological Society of America Special Paper 108
 , p.
1
47
.
Shitian
,
Wang
Zhuoyuan
,
Zhang
Zheng
,
Zhan
Hanchao
,
Liu
,
1988
,
On the characteristics and dynamics of the catastrophic Mount Sale landslide, Gansu, China
:
Journal of Chengdu College of Geology
 , v.
15
, no.
2
, p.
58
63
.
Weijiang
,
Wu
Shouying
,
Wang
,
1989
,
Mechanism of Sale Shan landslide (in Chinese)
, in
selected Papers on Landslides
 :
Chengdu
,
Sichuan Science and Technology Publishing House
, p.
184
189
.
Xiaogang
,
Zha
,
1983
,
Faulting is the foundation of the landslide at Sale Mountain, water and gravity of soil-body are the decisive factors
:
Bulletin of Soil and Water Conservation
 , no.
3
, p.
27
33
(in Chinese).
Chengqi
,
Zhang
,
1986
,
Analysis of the landslide at Sale Mountain
:
People's Yellow River
 , no.
3
, p.
41
45
(in Chinese).
Zhuoyuan
,
Zhang
Shitian
,
Wang
,
1984
,
On the prediction of the occurrence time and the velocity of a potential landslide
, in
Proceedings, 4th International Symposium on Landslides
, Volume
3
:
Toronto, Canada
, p.
145
146
.
Zhuoyuan
,
Zhang
Shitian
,
Wang
Lansheng
,
Wang
Hanchao
,
Liu
Runqiu
,
Huang
,
1992
,
Recent researches on the mechanism and geomechanic model of huge landslides in China
, in
Bell
,
D.H.
,
Landslides: Proceedings of the 6th International Symposium on Landslides, Christchurch, New Zealand
:
Rotterdam, Netherlands
,
A.A. Balkema
, p.
625
634
.

Related

Citing Books via

Close Modal
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close Modal
Close Modal