The Champati Slide

The dynamics, run-out extent and deposition morphology of granular slide (landslide) primarily depend on the physical and geometrical properties of the constituent grains involved and the topography of the slope in which these grains slide.^1,2 Here, we are primarily interested on the scientific question on the dynamics and deposition of Champatis (the stony seed of Lapsi, a native Nepalese fruit) slide in a laboratory experiment: how it flows, where it goes, how far, and with which force.

Lapsi trees are widely found in mid-hills and mountains of Nepal with very high-intensity in (and around) the Kathmandu Valley. However, the Lapsi trees are mainly found in the tropical and subtropical South–East and also East Asia.³ Lapsi fruit (Fig. 1) and Champati (Lapsi seed, Fig. 2) and the bark of Lapsi tree have many medical, economic, and social importance in daily life.^4–6 

The Champati has a very complex anatomical structure. The wide-flat-head and narrow-oval-tail create local slope for each grain, leading to the spin and rolling motion. The outer surface structure plays a dominant role in the dynamics and deposition morphology of the Champati slides. Pericarp (outer) anatomical structure³ makes the dynamics of Champati movement very unusual as compared to the other spherical-type grains such as sand and glass beads.² For the anatomical terminologies and descriptions of Champati, we refer to Hill,⁷ Hill,⁸ and Wannan and Quinn.⁹ 

The botanical and anatomical studies of Champatis have been carried out in the past.^3,7–9 However, no dynamical and depositional behavior has been studied yet. Here, we focus on some novel granular chute experiments composed of native Nepalese granular materials with Champatis, their flow dynamics, separation between particles of different physical properties, and their interactions with different obstacles. Particularly, we present the seminal dynamics and depositional studies of the Champati slides alone and also in combination with a typical Nepalese native food grain (Silam).

Here, we conduct a number of laboratory experiments on the single phase and mixture flows of two different native Nepalese food grains/seeds having different geometrical and physical properties to study their dynamic flow characteristics, responses when the flows interact with obstacles in the flow path, spreading and deposition morphology of grains, and their mixtures. To study the dynamic differences in the flow–obstacle-interactions, we vary the materials, their compositions and the orientations of the obstacle.

Due to its complex geometrical structure, Champati (the seed of a native Nepalese fruit Lapsi, as it is called in Kathmandu Valley, Nepal) is a type of supergrain. Lapsi [Nepalese Hog Plum; Botanical, Choerosopondias axillaris Roxb, Fig. 1(A)] is widely available and used (distinctly in the Kathmandu Valley) as a very special item in Nepalese food (cuisine) mainly as Achar (a pickle or relish article of food) that makes the entire menu very tasty. Lapsi is also cooked with different legumes and vegetable-curries. Alternately, it is used as a souring agent. Lapsi is mainly used to make dried fruit nuggets or fruit leather, popularly known in Kathmandu as Maadaa, Titauraa, and Paaun. These items are both sweet and salted or sweet–salted, and spicy. However, some people directly take it fresh.

The Nepalese postage stamp of Lapsi from 1978 [Fig. 1(C)] indicates its nativeness and importance in Nepal. So, the indigenous Nepalese fruit has important nutrient, medical, industrial, economic, and social values. Until about 1990, Champatis were a type of popular seeds among children (as could be seen in Kathmandu Valley) as playing objects alternative to round marbles. It is an interesting spinning object in its apex side, the main attraction is the spin rate and duration. When in relatively large scale, the Champaties are also used as a cooking fuel. Lapsi is also important in medicinal uses.

Unlike other food-grains, engineering, and geological materials: sugar, salt, soil, gravel, rock, or sand grains, due to its very peculiar geometry and other physical properties, the hypothesis (anticipation) is that the dynamics of the Champati slide will result in a fundamentally different dynamical, spreading, separation, runout, and depositional morphology from the other more rounded or ellipsoidal grains. The potential dynamical and depositional complexity of the Champati slide may also be inferred from their physical and geometrical properties as compared to other industrial and geological grains. Although the results presented here may provide some insights and further understanding of superspreading of the fragmented rock avalanches, the correspondence and applicability of these results to earth science must be carefully studied in detail. Here, our curiosity is mainly driven by its typical physical properties and dynamical flow and depositional behaviors. For this, we run the experiments on Champati slides in the laboratory Nepnova – Innovation Flows at the Kathmandu Institute of Complex Flows, Kageshwori Manohara-3, Bhadrabas, Kathmandu, Nepal.

The main aim here is to reveal how the special physical and geometrical properties of Champatis lead to previously unnoticed dynamical and morphological characteristics as these grains slide down a slope. This may help us to infer some important information associated with the measurable properties of such grains. With this, we may be able to draw some conclusions on what we can expect in the dynamics, deposition, and mobility as guided by the unusual physical and geometrical features of some similar complex grains involved in landslides or industrial flows of granular materials. We seek to understand the dynamics and complex behavior of Champatis alone and when mixed with some other native Nepalese food grains, particularly Silam, not considered previously. Unlike other experiments found in existing literature, as the materials and their unique composition considered here have very special physical and mechanical properties, this may result in fundamentally different dynamics and depositions.

Champati and Silam grains and their mixture are considered for several reasons. (i) They are very common in Nepalese Cuisine. No, or less attention has yet been paid to their flow dynamics, and depositional behaviors. (ii) These grains are the epitome in terms of their size, shape, and physical properties like their grain geometry, friction between the grains (internal friction), frictions between the grains and basal surfaces (basal friction), grain density, and heap properties. (iii) The industrial process engineering may substantially benefit from our novel findings. (iv) The results may be useful to investigate some special properties such as hypermobility, superspreading, and their control in mixture flows of complex geological material.

For this, we perform some simple, but fundamentally new experiments with Champati and Silam and the Champati–Silam mixture down inclined chutes and channels connected to horizontal plane. We run several novel experiments using an undisturbed flow path and also explore flow-obstruction with triangular obstacles, and reveal some important new findings. From these experiments, we aim to obtain some interesting results that may be applied to landslides and avalanches down mountain slopes. The primary scientific purpose is to verify this hypothesis with the observed novel phenomena. So, we are mainly interested in understanding how the slide of Champatis evolves, its run-out, spreading, and depositional morphology as well as how these grains separate from other grains in the mixture flows during motion, and as slides come to a standstill.

We focus on the dynamics, propagation, and deposition morphology as well as flow-redirection by the presence of obstacles on the flow path. Therefore, the main purpose here is to understand the basic dynamics, dispersion, run-out, and deposition morphology of the structurally and physically very special Champati grains and discover the novel understanding of Champati flows for the first time. We aim to understand how the flow dynamics are altered by other grains in the mixture; and the impact and interaction of Champatis and the mixture with both the mobile and immobile obstacles.

After removing the skin (exocarp) and flesh (mesocarp, Fruchtfleisch), we extracted the Champatis (endocarp and seed) from the Lapsi available in the Kathmandu Valley. The Champatis are carefully cleaned with the water-detergent fluid, and washed several times subsequently with warm water and normal water. The cleaned Champatis were then sun-dried for several hours until they were completely dry.

We measured the geometrical and physical properties of the pericarps of Lapsi and Champati (Fig. 2); including different dimensions and mass (Fig. S1 in the supplementary material), volume (Figs. S2 and S3 in the supplementary material), porosity (Fig. S4 in the supplementary material), coefficient of restitution (Fig. S5 in the supplementary material), density and angle of repose (Fig. 3). As the anatomical structure of the Champati controls its dynamics, here, we are mainly interested on the outer endocarp (surface) structure of Champati, which is made of obscure ridges with some pits that are regularly arranged on both of each ridge.^3, Figures 2(A)–2(C) display the lateral, apical, and basal views of a Champati. (A) Lateral view showing the obovate endocarp and obscure ridge with few regularly arranged surface pits. (B) Apical view showing five eyes and seam on the top. (C) Basal view showing five minor oval eyes (pits). For more details on these, see Fu et al.³ The ovoid stone Champati is distinctly characterized by its dominantly five (but occasionally four or six) subapical oval pores (occurring near the apex) covered by fibrous membranous structures.^3,8 We even found Champatis with three and eight subapical oval pores. We call the subapical oval pores the Champati eyes.

The Lapsi with cylindrical-type rounded ends, generally has flattened-ended more oval structure [Fig. 1(A)]. However, its seed, the Champati, has very special, but even more complex asymmetrical truncated-head, oval-type, three-dimensional anatomical structure [Figs. 2(A)–2(C)]. It is largely symmetrical about the plane containing its major axis [Fig. 2(A)], but asymmetrical about a plane of minor axis that is perpendicular to the major axis [Figs. 2(A) and 2(B)]. The Champati grain, with its head, main body and tail appear to generate different frictional resistance. The more circularly head-side (pedicel side) is relatively flat [Fig. 2(B)] whereas the tail-side (apex side) is rounded-ellipsoidal [Fig. 2(C)], and the body is in between [Fig. 2(A)]. The freely horizontally laying Champati rests with its main body along its major axis [Fig. 2(A)]. The head [Fig. 2(B)] usually is (mostly equally spaced) lens-like (main axis oriented) five-eyed (as we introduce this term fitting the appearance of the Champati), but some can be six-eyed, to a lesser extent four-eyed; and extremely rarely three-eyed or eight-eyed.

It is interesting to observe that the mass of the Champati increases as the number of eyes increases (Table S2 in the supplementary material). The tail-side has shifted mirror of often rarely visible eyes of the head-side [Fig. 2(C)]. Figure 2(D) displays the internal structure of the Champati after being cut into two longitudinal halves, while Fig. 2(E) shows its structure when they are further split into four longitudinal sections. Figure 2(F) shows the structure when cut into two transversal half sections. The head is positioned vertically downward on the left, while the tail is oriented vertically downward on the right. The holes inside the grain align with the number of eyes in the longitudinal direction. While cutting the Champati into sections, we observed that the eyes at the head and those at the tail are connected by tubular structures of hard crust and relatively soft core (interior).

Champatis are of brownish-light color or pinkish in some cases. Some other physical properties of Lapsi and Champati are presented in Figs. 1(B), 2, and 3, and also in Figs. S1–S5, in the supplementary material, including their masses and heaps. The Lapsi and Champati grains are also distinct in terms of their other physical properties like basal friction, internal friction, and density. From laboratory measurements, some of the main physical properties of these grains are listed in Table I.

Our detailed measurements reveal that the mass and anatomical structures between Lapsi and Champati differ greatly. Of particular importance is the surface anatomy of Champati that determines its friction, spinning and rolling during the slides composed of Champatis. These features of Champatis are very special. The geometrical structure and anatomy of Champati play vital role in characterizing the flow dynamical, mobility, depositional, and dispersional properties of the Champati slides, the main focus of the study. For this reason, here, we are interested to understand the dynamical behavior of the Champati slide. As the Champati head surface curves inward as it approaches the head side from the central body, its near eye minor diameter decreases slowly. This also characterizes the geometrical complexity of the Champati that we want to explore how this influences the flow dynamics and the deposition. All these properties will be needed and help to explain the very unusual dynamics, superspreading, thinning, and exceptional run-out of the Champatis in the deposition fan.

The dispersion ability of the granular material determines its dynamics, run-out, and spreading. Figure 3(Aa) displays holding of a bulk volume 3465 cm³ of Champatis in a cylindrical bucket of diameter 20 cm. The hyper-dispersion (spreading) is observed when the bucket is lifted [Fig. 3(Ab)], and the apex height of the deposit (56 mm) after spreading. Similar dynamics are shown for Silam [Fig. 3(B)] and gravel [Fig. 3(C)]. The spreading of Champatis is substantially higher, resulting in a much lower apex heights as compared with the apex height of the Silam and gravel. The apex heights for Champati, Silam, and gravel are, respectively, $hc=5.6$ cm, $hs=6.34$ cm, and $hg=7.6$ cm. We define the spreading by taking the ratio between the distance of the distal particle in the deposit from the apex height of the deposit with the apex height. The spreading for the Silam and gravel are $Ss$ = 38.5 cm and $Sg$ = 33.5 cm, respectively. However, the spreading for the Champati is $Sc$ = 86.7 cm, which is incredibly higher than the gravel and sand. So, the respective dispersion intensities are $Sc/hc=15.48$⁠, $Ss/hs=6.07$⁠, and $Sg/hg=4.41$⁠, which is predominantly high for Champati grains, while relatively low for Silam and very low for gravel. The hyper-spreading of Champati is the result of the very typical surface structures and the complex spins of Champatis. This will be explained in more detail with the deposit of the Champatis slide later. For this reasons, Champatis are said to exhibit the hyper-spreading during their motion, deformation and deposition. Once the bucket is lifted, the Champatis in the outer part of the moving/deforming heap roll and spin on the wooden board traveling much longer distance [Figs. 3(Aa) and 3(Ab)] than Silam [Figs. 3(Ba) and 3(Bb)] and gravel [Figs. 3(Ca) and 3(Cb)]. Figures 3(Ac), 3(Bc), and 3(Cc) display the measurements for the angle of repose for the heap of Champati, Silam, and gravel, respectively. As for spreading, the angle of repose for the Champati grain is remarkably low as compared to Silam and gravel, while this angle is remarkably high for gravel.

The Champatis, mostly in the outer ring at the time of lifting the bucket, move quickly and radially outward. The Champatis in the next inner rim also move outward but in a lesser extent and slower speed. The other Champati grains in the further inner rim also tend to move outward, but as their momentum is weaker, they roll and spin in one or another direction, and not moving that much outward. However, no preferential rolling and spin directions could be identified particularly for the Champatis in the front position. Many of these Champatis make a type of circularly zig-zag path around the main (central) heap of intact and continuum (near-) deposition heap. Some of them collide and even coalesce to the main heap, or making some local coalesce-lobes. We observed these complex dynamics even in a simple lifting of grains from a bucket on a horizontal wooden plate that resulted from the special anatomical structure of Champatis. As the Champatis roll, spin and move, their body lie on the surface such that the grains' major axis is in the horizontal direction, and the minor axis is in the perpendicular (to the slope or plane) direction. Importantly, we observed that the spinning and rolling are about the apex (tail) side while the apex is also making circularly zig-zag path less than the circularly zig-zag path of the pedicel (head) side. It was the result of the local slope of the grain decreasing from head to tail that controls the spin and rolling as the particle linear momentum is overtaken by the spin and rolling, resulting in a complex motion.

The chute geometry as shown in Fig. 4 consists of an upper inclined plane of the rectangular wooden plate (plywood). This plane is inclined at an angle of $32°$ and is connected to a horizontal plane. The inclined plane and the runout plane are connected by a transition line. For the measurements in both inclined and horizontal plates, there are adjacent grids of the same color at 10 cm apart while those of different colors (red and black) are just 5 cm apart. The initial mass is released by quickly lifting the bucket toward the direction perpendicular to the slope. Stopwatches are installed to measure the evolution times of the granular slides from which the slide velocities can be obtained. We used the video recording (50 fps) to analyze the dynamics of the Champatis in motion. For it, one camera is positioned at the front of the chute to capture the video footage from the downstream prospective, while the other camera is placed at the back to record the flow dynamics from the upstream. Both cameras are set at a distance of 45 cm from the chute in all experiments. For the recording, we used the cameras: Nikon D750, and GoPro Hero 12 Black.

The Champati and Silam grains and the mixture of Champati–Silam are used as the sliding materials in the experiments. Perilla seeds, scientifically known as Perilla frutescens, are commonly called Silam in Nepali. These seeds are harvested from the Perilla plant, and almost spherical in shape with a diameter of $2−3$ mm. Silam is widely used in both culinary and nutritional purpose because of their vast use as an oil source and distinctive and delicious flavor.

The granular flows may encounter obstacles in the flow path and sometimes they mobilize the obstacles, especially if they have a sufficiently large volume and momentum. Obstacles may be either naturally formed or human-made structures constructed in areas susceptible to landslides that need protection. The granular mass may or may not mobilize the obstacle. As the impact force of a flow overcomes the frictional resistance of an object (obstacle), the object moves as long as the pressure exerted by the flow is greater than its shear resistance. To examine the interaction of the sliding granular mass with obstacles downstream, we consider a triangular pyramid (or a regular tetrahedron) [Fig. 4(B)]. The obstacles in the flow path can deflect a remarkable amount of sliding material and/or retard or stop some portion of the granular mass. As a result, tetrahedral obstacles can be efficient for deflecting and energy-dissipating structures.¹² For the experiments, we consider the obstacles as: stationary (immobile) and movable (mobile) regular tetrahedra of edge 9.5 cm and slant height 8.7 cm.

Employing the experimental setup, we performed several experiments by changing (i) the granular materials, (ii) changing their mixture compositions, (iii) the position of initial release mass, and (iv) interactions with obstacle (immobile and mobile) in the flow path. Experiments were performed repeatedly (five times) under the same settings and conditions unless their dynamics, runout, and deposition morphology became almost identical. Each test was recorded by two video cameras (50 fps). The run-out, spreading, and deposition depths were measured manually after each test using a measuring tape and a Vernier caliper (Dasqua Blu, an electronic digital Vernier caliper that can measure within the range of $0−150$ mm with a tolerance of 0.015 mm). Geometrical and physical properties of the particles are characterized by the measurements and the video images. Experimental videos are provided in the online multimedia for all experiments.

We run several experiments with novel grains (specifically, Champati and Silam) with undisturbed flow path and when the flows are obstructed with some obstacles on their paths. The basic grains for the experiments are the Champati grains. Later, these grains are mixed with other food grains of fundamentally different physical properties (including grain shape, size, density, friction). Experiments are performed on the unconfined chutes and the narrow confined channels so as to understand the flow and depositional behavior in the unconfined and confined geometries, and how they differ.

First, we run the unconstrained flows of Champatis. The initial bulk (grain packing) volume $Vc=3465$ cm³ of Champatis is held in a cylindrical bucket centered at the position of 40 cm upslope measured from the transition line. The inclined and horizontal plates are marked with grids of 5 cm line spacing to ease measurements.

The dynamics of Champati slide is presented in Fig. 5 (Multimedia view). The upper panel describes the motion from inception to spreading in the inclined plane until the Champatis hit the horizontal surface. It takes about $t=0.28$ s. When the cylindrical bucket is lifted at $t=0$ s [Fig. 5(A)], the mass largely deforms downslope at $t=0.22$ s mainly due to the gravity, but also shears in cross slope on the inclined plane resulting from the hydraulic pressure gradient [Fig. 5(B)]. As the hydraulic pressure gradient almost ceases, no further crosswise dispersion develops and the grains quickly shear/move downwards with an acceleration due to the gravitational force, thereby resulting in a rapid decline in the depth of the mass in the main body until $t=0.28$ s [Fig. 5(C)]. The flow body is still intact and smooth, as expected. However, as the substantial mass hits the horizontal plate at $t=0.47$ s [Fig. 5(D)], the behavior is unprecedented, and somehow difficult to predict mainly in and around the frontal periphery and far downstream of the flow. The grains at the rear have almost dissipated their kinetic energy immediately after traveling the inclined plane and start to form the main deposition at the rear at $t=0.85$ s [Fig. 5(E)], but a substantial quantity of the leading Champatis in the flow quickly escape the main body. Here, exactly is the place where the very special physical and geometrical properties (anatomy) of the Champatis play their role in defining and controlling the tail and main body of deposit, the frontal spreading, rapid grain marching, some even moving out of the run-out domain (more than 1.2 m), and the long-time sporadic dancing of the grains lying in between. The major part of the deposition takes place at $t=0.85$ s. However, some detached Champatis from the front move, spin, and roll in an seemingly unpredictable manner over a long distance and time. All the grains come to standstill at $t=4.25$ s [Fig. 5(F)]. The last two phenomena, the relatively longer spreading time and the very long dancing time are the result of special grain surface anatomical characteristics of the Champati.

As the flow impacts the horizontal plate, many of the Champatis in the front of the flow depart immediately from the main body and move, spin, and roll (together we call this phenomenon the Champati-Dance) in a nonlinear way for a long distance and time [Figs. 5(D)–5(F)]. All this resulted from the very special geometrical structure and physical properties, typical of Champatis as explained in Sec. II B. The visual inspection reveals that the dynamics of dancing motions depend on the orientations and velocities of the Champatis as they decouple from the main sliding body.

These characteristics are very peculiar for Champatis. Until they have sufficient linear momenta, the Champatis propagate forward and outward. In the front, many of them even march outside the runout board. However, as soon as the momenta of the majority of the following discrete particles that are between the lead particles and the main body weaken, the movement becomes sporadic, exhibiting a near-circular zig-zag motion for a long time [about $t=4.25$ s, Fig. 5(F)]. As soon as the Champati slide impacts the horizontal plane, the grains behave as if they were parts of a fragmented rock particle.^10,13 With this behavior, one may even think if some kind of force, akin to the dispersive pressure in rock fragmentation,^10,14–16 is involved that ultimately and strongly pushes the grains (mainly in the outer rim of the moving body) quickly outward. However, unlike the fragmented rock clasts, the Champati grains are spreading in a very chaotic manner.

The question is, can we find some analogy between the spreading of the Champati slide in the laboratory experiments with the spreading of the fragmented rock particles in nature? The paths of the individual detached particles appear difficult to predict. This is a challenge associated with the Champati slide, unlike other granular material, a subject of separate research. It seems that there can be some similarities as can be inferred from the hyper-spreading of the Champati grains in the run-out plane. One common property is: first slowly, then, rapidly decreasing gradient, in the downstream direction, of the free-surface of the deposited Champati grains. There appears to be a strong energy transfer mechanism, akin to the rock avalanches. More on this needs to be studied systematically, probably with the recent dynamic rock-avalanche fragmentation model.¹⁶ 

The hypermobility of mass flows is still largely an unsolved problem.^17,18 As seen in Figs. 5(D)–5(F), the Champatis travel unexpectedly longer distance and wider areas; many of them even travel outside the horizontal plate. So, the unexpected dynamics, dispersion, and travel ranges of the Champatis may provide some useful insights in describing the long travel distance and run-out spreading of mass flows, and probably helping resolve some long-standing challenges. Here, the role is played by the very special surface geometry (shape) of the particle, true particle density, the size, and the exceptionally lower frictional energy dissipation due to smooth and ellipsoidal particle surface, both between the Champatis and the Champatis and the sliding surface, known as the internal and basal frictional energy dissipation. However, the particle rolling and spinning play the major role in the unexpectedly longer travel distance. The point is that in rolling and spin, the energy dissipation is much less than the energy dissipation in friction. Here, the rolling and spin mainly took place once the particles in the frontal region of the flow detached from the main body. Thereafter, those particles could largely move with very less contacts with the neighboring particles, meaning the collisional energy dissipation was also nominal. So, the very special convex surface geometry, and the special head–body–tail structure resulted in largely rolling and spinning motion with much less energy dissipation as compared with other particles, e.g., sand and gravel. This led to the superwide spreading of the Champati grains in the run-out plane. Such a steady acceleration for surprisingly long time and travel path for Champati is due to its ability to flow, slide, and roll down the slope and the runout surface without much of the energy dissipation in friction and grain collision. This explains the hypermobility of the Champati grains.

All these justify our choice of granular particles of extraordinarily diverse mechanical response during motion and deformation with very different dynamical behaviors. It appears that the reduced friction resulted in fast motion and the complex grain anatomy led to the sporadic zig-zag motion (dancing). This is amazing to observe such hypermobility even for the mass release from the nominal fall height of just 0.4 m (along the downslope) in a gentle slope of about $30°$⁠, many of the grains travel quite farther than 1.2 m in the downstream from the transition. Probably, this does not happen for the usual geological and engineering materials such as sand and gravel, an important aspect to be further studied and analyzed in-depth, preferably also with the dynamic simulation model.¹⁶ 

As the impact energy or pressure of the flow overcomes the shear-resistance of the object on the way of the flow, the object moves during the flow exerting impact pressure on it. The object moves as long as the pressure exerted by the flow is greater than its resisting shear stress, afterwards the object remains immobile. The interaction between the flow and the obstacles is highly influenced by the obstacle geometry, its nature (mobile or immobile) and also by the property and volume of the sliding material.^2,12,19–21 To examine such flow–structure interactions, experiments are conducted with the dynamical interaction of the flowing granular mass with a forward-facing tetrahedral obstacle mounted on the downstream horizontal plane at 10 cm away from the transition line [Fig. 4(B)]. Such types of obstacles are efficient structures for deflection and energy dissipation.

Undisturbed flow without obstacle in the path as a reference, now, we discuss the Champati slides and interaction with the forward-facing tetrahedron or pyramid which is placed symmetrically about the central line (⁠ $y=0$⁠) with its frontal nose at $x=110$ cm (Fig. 6) (Multimedia view). The dynamics until the flow impacts the obstacle is the same as in Sec. IV A [Figs. 5(A)–5(C)]. So, here, we only analyze the results as the flow impacts the obstacle. The front of the flow starts to impact the horizontal plane at $t=0.28$ s [Fig. 6(A)], and then, starts to impact the pyramid at $t=0.43$ s [Fig. 6(B)], its frontal nose and lateral slanting surface do not only resist the flow but also divert the flow into two elongated streams, thereby producing lateral deflection (diversion) [Fig. 6(C)]. Due to the blockage of the obstacle, the sliding grains are divided into two almost identical lobes, and kinetic energy of the flow is dissipated by the obstacle. This motion is due to lateral pressure gradient as developed by the obstacle impact, rolling, and spinning of the grains. The deflected streams are more pronounced until $t=0.88$ s [Fig. 6(D)] where entire grains completed to travel the inclined plane. The kinetic energy of the Champatis at the rear have almost dissipated by the obstacle and the (basal) friction, so main deposition starts thereafter. However, substantial amount of detached Champatis from the front of the flow roll and move further in the downstream [Fig. 6(E)] and entire hyper-dispersion comes to halt in 4.14 s [Fig. 6(F)].

The lateral deflection of the grains creates a granular vacuum behind the obstacle which becomes increasingly pronounced until $t=0.88$ s. However, afterwards, the granular vacuum becomes weaker due to inward deflection of some grains toward the vacuum as the uneven geometry of head and tail of the Champati creates local slope for each grain at the boundary of the deflected lobes. Due to the unique shape of Champati, complex motions and rollings in different directions are evolved with a substantial amount of Champatis spreading over large area in the middle and frontal region of the run-out plane with long travel distances. About half of the Champatis spread in front of the tetrahedron, while the remaining mass makes an intact deposit heap lying just on and behind the tetrahedron. Ultimately, the main deposition of the grains develops at the front and around the lateral sides of the obstacle with a maximum deposition depth of 27.7 mm at the vicinity of impact (10 cm away from the transition) after 4.14 s, whereas the maximum deposition depth was only 25.6 mm for the undisturbed flow, but at 15 cm away from the transition after 4.25 s.

To study the dynamics of Champati slide, interaction with mobile obstacle and the mobilization length of the obstacle, a cylindrical bucket containing Champati grains with a bulk volume of 3465 cm³ (as in Secs. IV A and V A) is lifted at $t=0$ s. For the flow-mobile obstacle interaction, we place a forward-facing tetrahedron or pyramid symmetrically about the central line (⁠ $y=0$⁠) with its frontal nose at $x=110$ cm (similar to Sec. V A). The dynamics until the flow front transits to the horizontal runout plane is the same as in previous Secs. IV A and V A [Figs. 5(A)–5(C)]. Thereafter, the front of the mass begins to interact with the horizontal plane at $t=0.28$ s, as it is seen in Fig. 7(A) (Multimedia view), and hence, starts to decelerate as the gravitational force ceases, whereas the basal friction becomes dominant on the runout plane. The flowing Champati mass starts to impact the front face of the obstacle at $t=0.43$ s [Fig. 7(B)] where the grains are intact. At $t=0.64$ s, the substantial amount of the mass starts to interact with the obstacle with the impact force produced by the kinetic energy with the higher speed [Fig. 7(C)]. As the force associated with the exerting kinetic energy exceeds the resisting force of the obstacle, the obstacle is mobilized to a distance of 3 cm downstream from its original position due to the impact. As the mass completely leaves the inclined slope at $t=0.85$ s [Fig. 7(D)], the obstacle is further mobilized as the flow continues to push it forward. Then, the kinetic energy of the flow, mainly in the vicinity of the obstacle, is dissipated substantially due to the impact against the obstacle, producing the lateral diversion of the sliding grains into two identical lobes [Figs. 7(C) and 7(D)]. The lateral deflection of the grains creates a granular vacuum behind and at the lee side of the obstacle which is pronounced as time elapses. The mobilization of the tetrahedron slows down as kinetic energy is almost dissipated. Ultimately, the main deposition of the grains develops at the vicinity of impact at $t=1.33$ s [Fig. 7(E)] but some detached Champati grains from the main body roll, spin and move further downstream. The Champati-Dance takes place for quite a while (for 2.87 s more) and reveals the supermobility of the Champatis. The entire mass comes to a halt at $t=4.14$ s [Fig. 7(F)] with a maximum deposition of 26.4 mm (14.5 cm away from the transition) which is slightly higher than the undisturbed flow (Sec. IV A) but lower than the immobile obstacle interaction (Sec. V A).

The mobilization mechanism is as follows: As the kinetic energy of the Champati flow overcomes the frictional resistance of the tetrahedron, the tetrahedron starts to move. The flow mobilizes the tetrahedron as long as this mechanism prevails. The Champatis mobilized the tetrahedron to a distance of 4.5 mm downstream in the horizontal plane where the tetrahedron was placed 15 cm downstream of the transition. So, although the Champati has relatively low true density as compared to gravel, it has higher object mobilization capacity. The main driving forces are associated with the relatively larger grain size, but very low frictional energy dissipation both between the grains and between the grain and the sliding surface.

We also performed the experiments on flow–obstacle interaction of Silam and gravel (Fig. 8) (Multimedia view). However, they could not mobilize the obstacle as the relatively smaller and less dense Silam cannot exert the impact force that exceeds the resisting force of the obstacle [Fig. 8(B)]. Conversely, gravel has higher density but the kinetic energy is quickly dissipated due to its very high internal and basal frictions. As a result, almost all the grains deposited over the inclined plane without mobilizing the obstacle [Fig. 8(D)]. Yet, we note that the initial potential energies of the Champati, Silam, and gravel are 3.97, 3.04, and 11.55 J, respectively. However, the impact pressure provided by Champati was the largest, by gravel was the lowest, and by Silam was in between.

Most natural granular flows are multi-phase, which consist of a heterogeneous mixture of solid particles with different physical properties and rheological behaviors.^22–24 Their movements are influenced by dynamic interactions among the solid constituents of different physical properties, including size, shape, and friction. So, a better knowledge of their dynamics is required to properly understand the flow, runout, and deposition morphology. As such flows are more complex, but more realistic than the single-phase granular flows, now, we deal with granular mixture flows. For this, we consider the Champati as the base granular material, and then add Silam to form a two-phase mixture. The main idea in doing so is to understand how the flow behavior of the supergrain Champati is altered and controlled by the Silam grains in the mixture. In response, how Champati changes the dynamics and deposition of Silam. The questions are: (a) does the dynamics of Champati and Silam mixture lie in-between Champati and Silam, or (b) are there unexpected dynamical behaviors of the extra-ordinarily behaving Champati? To address these questions, we perform experiments for mixture flows without and with obstacles.

There are several mechanical, dynamical and geomorphological reasons of carrying out these more complex experiments: (a) To understand the complex mechanical interactions between the grains of completely different physical behaviors. Because the physical properties of the constituent materials are fundamentally different, we anticipate that the mechanical response and the rheology of the composite material will be different from those constituent materials separately. (b) As the mechanical behavior of the composite material changes during the movement, it will result in the different dynamical response and flow characteristics, i.e., the flow depth and the flow velocity, determining the state of the flow. These dynamical quantities are required to describe the impact force and the inundation area, as needed in hazard mitigation and planning. (c) Most importantly, the inherent mechanical controls of the composite material (e.g., mixture densities, frictions, viscosities, and the volume fractions of the constituents), and the flow dynamical quantities will eventually determine the geomorphology of the deposition, including the run-out, inundation area, spreading, and the local material composition and their structural formations. These information are very important from the environmental, and engineering point of view. (d) The data, thus, produced may be utilized to validate the performance of the physical-mathematical two-phase and multi-phase, or multi-component flow models. This is so, because, the coarse Champati grains may be considered as the Coulomb plastic material, whereas the Silam when mixed with Champati may be considered as Coulomb-viscoplastic materials.²³ (e) Other important aspects of conducting experiments with the composite mixture materials is to understand the mechanism of mixing and separation between the phases (constituents) in the mixture. These data can also be applied to verify the mechanical phase-separation principles.²⁴ 

Dynamics of the mixture flow differ fundamentally when the base material Champati is mixed with different materials due to their different physical and rheological properties. Now, we study the dynamical evolution of the mixture flow when they slide down the inclined plane. For it, we run experiments with the flows of mixture of Champati and Silam with equal volume of 1732.5 cm³ [Fig. 9(A)]. The constituent materials are uniformly mixed at the time of release by carefully blending them manually to obtain the best possible mixing of different grains. Moreover, as the physical and geometrical properties (mainly size) of Champatis and Silam grains do not vary that much from their mean values, in this study, we disregard the grading of the grain size. However, further research is required to include the possible influence of the grain grading in the dynamics and deposition of these mixture flows.

As in the previous experiments, the mixture starts to advect and slide downslope and cross slope due to gravity and hydraulic pressure gradient [Fig. 9(B)] (Multimedia view). When the pressure gradient is weaker on the inclined plane, the mass mainly advects downslope. The front of the mass starts to transit from inclined to horizontal plane at $t=0.33$ s [Fig. 9(C)]. In this scenario, the interesting phenomenon is that, the front and either side of the boundary of the flowing mass are dominated by the coarser particles (Champati), while in the rear, almost all portion is covered by the finer particles (Silam), because larger and coarser particles dominate the flow front as also observed in most of the natural landslides.^11,24

The flow evolution at $t=0.66$ s shows that the rearmost Champati is 15 cm above the transition while the Silam is at 40 cm above it. This indicates the separation of constituent phases with a separation length $40−15=25$ cm [Fig. 9(D)]. The front of the flow is dominated by the Champatis, whereas the tail is dominated by the Silam. Moreover, due to the difference in grain size, the Champatis begin to float on the free surface and Silam remains at the basal surface. At $t=0.78$ s, all the Champati grains arrived the horizontal plane while the substantial Silam grains are still over the slope [Fig. 9(E)]. It means, there is no mixture in the slope and the phase separation is maximum with separation length of about 35 cm. The advection of the mixture in the downslope and runout zone continues with tail almost of Silam and the frontal and lateral boundaries of the mixture body mostly containing Champatis. As the gravity ceases in the runout zone, the mixture starts dispersing. The detached Champatis from the mixture advect and disperse farther away in the runout zone. However, the main mixture body remains still. The detached Champatis roll and dance further and come to a halt at 3.49 s [Fig. 9(F)] which is 0.76 s earlier than only Champati slide in Fig. 5(F). Given the scale of experiments (total material volume and travel distance), the difference is substantial in this time, as confirmed with repeated experimental runs. The phase-separation between Champati and Silam persists not only in the inclined plane, but also in the horizontal runout where a strong outer ring of Silam emerges in the upstream of the near deposit and deposited heap while Champatis exclusively occupy the lateral boundaries of the deposit.

In the mixture, as they accelerate differently, the Champati has the greatest acceleration as it is almost rounded in shape and has tendency to roll quickly as they disperse. In contrast, Silam accelerates quite slowly due to its lower density and high friction. Most interesting point is that, in the mixture flow of Champati and Silam, the dynamics closely resembles those of the Champati slide, indicating that Silam particles were substantially propelled by the Champatis. In response, the Silam particles also exerted substantial control back thrust over the motion of the Champati grains.

In natural mass flows, the local mixture composition changes, resulting in the phase separation between materials of different properties, e.g., coarse and fine particles; particles and fluid. Such phase separation plays an important role in the dynamics of mixture flow, deposition in the run-out plane and in forming lobes, and frontal surge heads.^24,25 The separation of the coarser particles from the finer particles evolves as the uniform mixture slides down in the inclined plane. The phase separation is a good indicator of particle size segregation.²⁶ It is influenced by variations in the material properties like grain-size, internal and basal friction angles, and grain density. In naturally occurring debris flows, the coarse particles are propelled at the front (dominating solid phase) that have high intergranular friction. However, behind the front, liquefied debris or fine sediments (sand) accumulate, forcing the frontal boulders forward and aside to form levees.^{11,24,25,27,28}

In our experiments, phase separation emerges both in the flow front and the rear, conveniently observed dominantly in the rear of the flow and deposition. In the experiment of Champati–Silam mixture, we observe that as time elapses, the Champati volume fraction (coarse particles) substantially dominates that of the Silam in the front. Silam volume fraction strongly dominates the Champati volume fraction in the rear, resembling the phase separation. During this separation process, the Champatis move in the front and float up in the main body until it deposits. Whereas, the Silam is mechanically compelled to travel inside the flow in the main body, and to the back of the propagating body. Due to the difference in their physical properties and geometrical structures, the Champatis separate very efficiently with other food grains (Silam, Phapar, Tori, Gahat) and geological and engineering materials (sand, gravel), results not shown here.

The separation length is determined by the local evolution of the particle concentrations of the phases. In other words, the separation length of the phases in the mixture flow is the length of free of some phase when the region is fully (mostly) occupied by the other phase only. This is the length between the dominating and almost non existential phases. In the Champati–Silam mixture flow, the extraordinary separation between Champati and Silam evolved just in a blitz of the time and in a remarkably very short travel distance after the release of the mixture mass. Such an absolute and tremendous separation between grains of different physical properties is eye-catching and revealed here. As the mass at the rear descends further along the slope, phase separation becomes more intense, causing the increment of the separation length. The separation length becomes the highest when all the coarse particles have completely traveled the slope. Given a small volume of the mixture mass and a short chute length, this separation length is incredibly high, almost equal to the whole length of the traveled distance in inclined plane. Mechanically, either the Silam could not control the Champati motion or the Champati could not substantially thrust the Silam particles in the Champati–Silam mixture. The dynamics of the separation length, in fact, intrinsically characterizes and explains the complex interaction and interfacial momentum exchange between the phases in the mixture. We could obtain such a great separation mainly due to the unique property of the supergrain Champati.

Similar to the previous experiments, when the uniform mixture of Champati and Silam of equal volume is released at $t=0$ s from a cylindrical container, the mixture starts to disperse in all directions due to the pressure gradient. Due to gravity, the mixture starts to travel downslope. The sliding mixture starts to transit at $t=0.33$ s [Fig. 10(A)] (Multimedia view) and begins to impact the stationary obstacle at $t=0.49$ s [Fig. 10(B)]. Figure 10 shows several interesting and seminal phenomena associated with the Champati–Silam mixture slide, its encounter with a tetrahedron and deposit. As the flow impacts the tetrahedron, flow is distributed into two lobes [Fig. 10(C)]. In the front of the lobes, preferentially Champatis are separating quickly from the Silam resulting in the only Champatis-led front [Fig. 10(D)]. As time elapses (⁠ $t=1.06$ s), at the back of the lobes, the Silam separates from the Champatis as the Silam remains virtually deposited (immobile), but the Champatis are still in relatively fast motion [Fig. 10(E)]. During the flow, once the Champatis are brought to the top of the flow by shear and infiltration (percolation) of Silam, ultimately bringing them to the lower layer of the flow, the Champatis remain in the front and at the top of the flow. Consequently, the Silam primarily are forced to remain at the bottom and the back of the flow. Then, the mass deposition begins with an upward propagating shocks in the back of the flow as material exits the inclined slope. Interestingly, the heap of Silam remains stationary in the deposit but the Champatis floating on top of it, slip on it, and propagate much farther [Fig. 10(F)]. The Champatis are detached only from the center of deflected streams, not from the center of the main body as the kinetic energy is dissipated while interacting with the obstacle. Also, the backward deposition over the incline is elongated. The Champatis on the Silam-heap virtually remain there, however, mainly those Champatis already separated from the main heap, quickly move farther downstream. This process continues quite a while (relatively long time as compared to the same for the main flow deposit). The entire material comes to a halt after quite a long time, as the Champatis are still dancing in their wider or narrower local vicinity, in the farther frontal region of the run-out and deposition plane. The Champati (coarse component) separation from the Silam (fine component) is such that a part of it shows advection, but the remaining part floats above Silam.

Mechanically and dynamically many fascinating spectacles are seen in Fig. 10. It shows that as the Champati–Silam mixture is released, immediately, the process of Champati separation from Silam begins already in the inclined slope. Phase separation intensifies as the mass moves downslope. The separation length (the length of complete separation of one material from another in some local region of the flow) increases in time. This separation process results in the frontal dominance of the Champatis and rear dominance of the Silam, both, first partially, then, fully. During the process, the Champatis rise up, move to the front. In turn, the Silam is mechanically forced to move down, inside the flow, and to the back of the propagating body. This results in phase separation²⁴ between Champati and Silam. The separation becomes clearer as the flow moves down, transits to the horizontal run-out, and nears the final deposition. Since the phase separation do not change much with the mobile object, here, we do not present results for phase-separation with mobile objects.

To study the dynamics of Champati–Silam mixture flow in a confined channel followed by unconstrained flow, we place a cylindrical container (containing the uniform mixture of Champati and Silam) with center at 80 cm above the transition. To confine the channel, two wooden plates (normal to the inclined plane) are fixed at 20 cm apart from upstream to the transition line as shown in Fig. 11(A). The flow inception takes place at $t=0$ s [Fig. 11(A)]. As the transverse shearing is controlled by the conduits, the height of the mass is not substantially decreased after inception as compared to the unconfined flow [Fig. 11(B)] (Multimedia view). The flow spreads down-hill in the inclined channel and begins to impact the unconfined horizontal surface at $t=0.57$ s [Fig. 11(C)]. At this moment, the phase separation becomes apparent at the rear, and the front has become Champati dominating while the rear is Silam dominating. The substantial portion of the flow reaches the unconstrained horizontal run-out zone at $t=0.83$ s [Fig. 11(D)], and hence, the dynamics changes from the constrained channelized two-dimensional flow to unconstrained three-dimensional flow over the flat horizontal plane. Then, the flow advances in the run-out with wider front, however, still principally moving in the longitudinal (frontal) direction, flow spreads a bit in the lateral direction, but mainly propagates in the main downstream direction at $t=1.14$ s [Fig. 11(E)]. Material accumulation (early stage of deposition) begins, accumulation intensifies while the frontal Champatis propagate with high speed and much farther. The inclined channel is then entirely occupied by Silam [Fig. 11(E)] while the frontal Champatis are marching farther outwards.

Virtually, the accumulated mass of Silam does not move in the run-out, but the Champatis that are floating on top of Silam-heap quickly slip on top of it in the foot-hill of the deposited Silam to a significant distance in the frontal direction. In the meantime, the Champatis in the frontal area of the flow move farther and farther. A significant number of Champatis are spread in a greater frontal area of the run-out plane, other remaining major part of the Champatis are floating on the central-frontal part of the deposited Silam. About one-third of the close-to-deposit heap (at the rear) of Silam is absolutely free of Champatis, so is the remaining following tail of the-only-Silam in the lower part of the inclined channel. These are amazing phenomena in the flow of the Champati–Silam mixture, revealed for the first time.

As time proceeds, a shock of Silam (Silam-shock) emerges weakly (as compared to previous experiments) just downstream of the transition from the inclined channel to the run-out plane, and propagates upslope in the channel, while the Champatis are moving farther downstream, greatly the Champatis are separated in the frontal region, but also substantially the Champatis are lying on top of the Silam-heap. While after $t=1.14$ s, the Silam-heap is virtually stationary (not moving, not deforming), the motion of Champatis is still remarkable. In the later time, only the fully separated Champatis in the frontal area are still moving and dancing, but the Champati-Dance is now much less than in the only Champati slide in Fig. 5, because the rolling, spinning and dispersion of the coarse Champatis are now controlled and dampened by the fine Silam grains. After the main body (containing the Silam and the floating Champatis on top of it) remained immobile, the entire slide material comes to a standstill at 7.85 s [Fig. 11(F)], slower than in unconstrained flow (Fig. 5). This is because of (i) the higher position of release than in unconstrained flow, and (ii) the constrained motion in the inclined portion of the chute.

From the geological and industrial perspectives, the separation between materials of different physical properties are of fundamental importance. The strong process of separation of Champatis from the Silam particles are observed in Fig. 11. Separation between the Champati and the Silam particles begins right after the flow inception and intensifies as the flow slides down the channel. The degree of separation between these two particles of different physical properties becomes higher and higher already in the inclined channel flow, but further intensifies as the flow enters the unconstrained horizontal run-out zone, where the separation length (a length scale showing clear and strong dominance of one material's volume fraction to the other material's volume fraction in a mixture flow in some region of flow) between the frontal Champatis and the Silam grains increase exceptionally, as does the separation length (a term invented here) between the main body and the exclusively Silam-covered long tail part of the flow in the inclined portion of the channel.

Right after the mixture mass release, the process of separation begins. The Champatis appear on the top of the flow, march to the flow front, and in the meantime, the Silam particles fall to the bottom of the flow and move back. These processes are observed in all the panels. The phase separation between different materials in the composite mixture has been described with the fully mechanical phase separation model with the enhanced separation flux of one material (here, Champati) against the reduced separation flux of another material (here, Silam) by Pudasaini and Fischer.²⁴ Almost immediately after the flow, but mainly as the flow develops in the inclined channel, the mixture behaves as if it is a two-layer flow, the Champatis fast moving to the front and floating on top of the Silam, the Silam falling into the bottom of the flow and largely moving to the tail of the body.

The spreading and spreading intensities of Champati, Silam, and gravel as discussed with reference to the deposition heap may be useful to model and describe the superspreading of the fragmented clasts in the natural rock-avalanches in the distal run-out fan.^10,14,15 There might be some analogy in the dynamics and energy transfer mechanism of the hyper-spreading of the Champati grains and the fragmented rock particles in rock avalanches. The unexpected dynamics, dispersion and travel ranges of the Champatis may provide useful insights in describing the long travel distance and run-out of mass flows, and probably helping resolve some long-standing challenges in gravitational mass transports. The observations and analyses on the dynamics, phase-separation, flow–obstacle interactions, and deposition of the Champati–Silam mixture from the present study can be utilized to validate the performance of the physical-mathematical two-phase and multi-phase, or multi-component flow models,^22,23 impact models, and the phase-separation principles.²⁴ This is so, because, the coarse Champati grains may be considered as the Coulomb plastic material, whereas the Silam when mixed with Champati may be considered as Coulomb-viscoplastic materials.²³ In our experiments, as larger and coarser particles dominated the flow front and top, and the rear and bottom of the flow body by the finer grains as in the natural landslides,^11,24 these results may be useful to further describe the mechanism of flow segregation and levee formation in geophysical mass flows. The phase separation process of different materials with different physical and geometrical properties are of prime importance in industries, especially in purgation, grading and packing.

Moreover, we mention that the Champatis stones are extracted from the fruit grain Lapsi. The Silam grains are widely used in Nepal. So, any food processing industries may use some of our findings in transport and industrial processing, in extracting the flesh of Lapsi by separating Champati from it and producing energy by burning the Champati grains. Similarly, as the Silam grains are used in producing oil and some sorts of powders commonly used in Nepalese cuisine, the basic knowledge gained here may help in understanding the transport and deposition behaviors of such grains and when mixed with the coarse Champati grains in industrial process engineering.

Here, we reported on some novel findings of chute experiments of native Nepalese granular seeds, the Champatis. The anatomical structure and physical properties make Champati an epitomic supergrain leading to the complex frictional, spin and rolling motion. We aimed to understand the complex dynamical and depositional behavior of Champati alone and when it is mixed with the other Nepalese food grain, Silam, not considered yet, but with fundamentally different physical and geometrical properties from Champati. We revealed spectacular dynamics, superwide spreading, flow–obstacle interaction and depositional behavior of the Champati, and the mixture slide emancipating from their very unusual physical properties. We proved our hypothesis that the Champati slide results in a fundamentally different spreading, separation, mobility, runout, and deposition morphology, as well as the way they separate from other grains in the mixture. Particularly, the surface anatomy of Champati played a vital role. Champatis made a circularly zig-zag path around the intact, near-deposition heap. As the dispersion intensity is predominantly high for Champati, these grains exhibit hyper-spreading. The very special properties of Champatis control the tail and the main body of the deposit, the frontal spreading, rapid grain marching, and the sporadic Champati-Dance. The particle rolling, spinning and the exceptionally low friction and grain collision played a major role in the unexpectedly longer travel time and distance, explaining the hypermobility of the Champati grains.

As the Champati mass hits the ground, the behavior is unprecedented. A substantial amount of the leading Champatis quickly escaped the main body and moved, spun, and rolled in a complex, nonlinear manner. As the flow impacts the tetrahedral obstacle, its mobility substantially changes the flow dynamics, spreading, and deposition. Champatis dominate the front and lateral boundary of the Champati–Silam flow, while the rear is largely covered by Silam. As Champatis propel Silam grains, the latter exerts control-back thrust over Champatis. During this phase separation process, Champatis move in the front and float up in the main body until they deposit, whereas the Silam is compelled to percolate inside the flow and travel in the back of the main body. The extraordinarily eye-catching separation of Champati from Silam evolved just in a blitz of time resulting from the unique property of the supergrain Champati. For the Champati–Silam mixture slide in a channel followed by unconstrained flow, a strong process of separation with exceptionally increasing separation length between Champatis and Silam grains begins right after the flow inception and intensifies as the mixture slides down the channel and enters the unconstrained horizontal run-out. These results may be useful in better understanding the hypermobility of complex geological flows, including fragmenting rock avalanches, and separation between materials of different properties, probably helping resolve some long-standing challenges.

See the supplementary material for videos of the physical laboratory experiments (SM1–SM11) associated with this article are included in the Supporting information as online multimedia. Speed of each of the videos is 0.2× that of the original speed. Moreover, details on grain and bulk properties of the Lapsis and Champatis are present.

Shiva P. Pudasaini acknowledges the financial support from the German Research Foundation (DFG) through the research project: “Landslide mobility with erosion: Proof-of-concept and application - Part I: Modelling, Simulation & Validation; Project No. 522097187.” Bekha Ratna Dangol and Chet Nath Tiwari acknowledge the Research Directorate, Rector Office, Tribhuvan University, Kathmandu, Nepal for the financial support through grants with Grant Nos. 36-2078-2079 and 3-2078-2079, respectively. This paper is based on: https://doi.org/10.21203/rs.3.rs-4182617/v1.

The authors have no conflicts to disclose.

Shiva P. Pudasaini: Conceptualization (lead); Formal analysis (equal); Methodology (equal); Supervision (equal); Validation (equal); Writing – review & editing (equal). Bekha R. Dangol: Data curation (equal); Formal analysis (equal); Validation (equal); Visualization (equal); Writing – original draft (equal). Chet N. Tiwari: Data curation (equal); Formal analysis (equal); Investigation (equal); Validation (equal). Jeevan Kafle: Data curation (equal); Formal analysis (equal); Validation (equal); Visualization (equal); Writing – original draft (equal). Puskar R. Pokhrel: Data curation (equal); Supervision (equal); Validation (equal); Visualization (equal). Parameshwari Kattel: Methodology (equal); Formal analysis (equal); Supervision (equal); Writing – review & editing (equal).

The data that support the findings of this study are available within the article and its supplementary material.