Study of the Mining Depth of Tailings Considering the Stability of Existing Open-Pit Slopes

Abstract

The recovery and comprehensive utilization of tailings resources can effectively mitigate or eliminate safety hazards in the upper zones of open-pit mines. To ensure the safe recovery of accumulated tailings and enhance resource utilization efficiency, this study establishes a two-dimensional model based on the Discrete Element Method (DEM) for the overall stability of tailings recovery, which is integrated with the existing slope and ore pillar models of the open-pit mine. Leveraging the mechanical parameters of tailings and waste rock obtained from laboratory tests, this study systematically investigates the effects of tailings recovery on the stability of existing slopes. Results show that due to differences in fracture characteristics and tailings reserves, complete tailings extraction causes no landslides in some sections, but large-scale or varying landslides occur on southern/northern flank slopes in specific sections at certain excavation depths or after full extraction. Targeted recovery recommendations are proposed: “segmented excavation with bench reservation” prevents overall landslides on southern flank slopes of landslide-prone sections; 35° slope cutting ensures stability of northern flank slopes in all sections. Further field verification considering rainfall and seismic loading factors is required for practical applications.

1. Introduction

Conventional tailings are typically regarded as waste products of mining operations; however, they often contain substantial amounts of valuable elements or minerals. The recovery and comprehensive utilization of tailings resources not only convert waste into valuable assets but also alleviate pressures on natural resources and the environment, ultimately mitigating or eliminating safety hazards in the upper zones of mining pits. With appropriate recovery technologies and utilization strategies, these once-valueless materials can be transformed into economically viable resources. This approach not only improves overall resource utilization efficiency but also generates additional economic benefits for mining enterprises. Furthermore, tailings accumulation frequently causes environmental pollution, such as soil and water contamination [1]. Reusing and processing tailings can significantly reduce pollutant emissions, thereby reducing the environmental burden.

Tailings storage facilities have long been essential components of mining operations, with their structural stability closely linked to mine safety, environmental protection and the protection of nearby populations. In recent years, driven by the continuous expansion of mining activities, the scale and elevation of such facilities—particularly ultra-high tailings storage facilities—have steadily increased [2]. This expansion has heightened the prominence of slope stability issues, accompanied by a marked rise in the frequency of landslides and other failure incidents [3,4,5]. Consequently, the analysis and control of dump slope stability have become key research areas in rock mechanics, geotechnical engineering and disaster risk mitigation.

The stability analysis of tailings storage facilities is inherently interdisciplinary, encompassing geotechnical engineering, structural mechanics and hydrogeology. Traditional theoretical methods—such as the limit equilibrium method and finite element method—have been widely applied in slope stability assessment [6]. However, advances in monitoring technologies and numerical simulation techniques have driven integrated analytical approaches to become increasingly mainstream; these combine GNSS-based monitoring [7], displacement and deformation measurements [8] and numerical modeling tools (e.g., FLAC2D and coupled seepage-stress models). For instance, GNSS technology enables real-time tracking of slope deformations, providing high-precision data for landslide early warning systems, while tools like FLAC2D facilitate dynamic simulation of slope deformation mechanisms under complex conditions (e.g., rainfall infiltration [9], seismic loading [10] and surcharge variations [11]).

In recent years, extensive research has explored factors influencing the stability of tailings storage facility slopes. Rainfall infiltration is widely recognized as a critical external factor, significantly altering soil water content and pore water pressure to reduce geomaterial shear strength and trigger slope failures [12,13,14,15,16,17]. Moreover, rainfall intensity and duration directly affect the likelihood of slope instability [18,19]. Beyond external triggers, internal structural conditions—including weak basal layers [3,20], stacking configuration, overburden thickness [21,22] and tailings liquefaction susceptibility [23,24]—also play essential roles in regulating slope stability.

In summary, stability analysis of tailings dam dumps and slope slip analysis have become important research directions in mine geohazard prevention and control. To ensure safe tailings recovery, improve resource utilization efficiency and efficiently recover valuable metals (e.g., copper, cobalt) from tailings, this study takes a specific open-pit dump as the engineering background and relies on tailings shear strength parameters under different moisture contents. Using a combined approach of physical experiments and numerical modeling, a two-dimensional DEM-based model of the tailings ore body slope is established to analyze the impact of tailings mining on dump slope stability. Reasonable tailings mining recommendations are proposed to address gaps in existing tailings recovery research, aiming to enhance mine resource utilization efficiency.

2. Engineering Overview

The tailings deposit of the western orebody at an open-pit mine in the Democratic Republic of Congo extends approximately 670 m east–west and 240 m north–south, with a maximum depth of 79 m and an average thickness of ~44.91 m (Figure 1). Four major cracks have developed on the southern flank slope.

Table 1. Field-measured orientations of sliding surfaces.

Borehole ID	Dip Direction/°	Dip Angle/°	Remarks
XKT 6	64	82	Major fracture
XKT 16	334	60	Major fracture
XKT 17~18	4	83	Major fracture
XKT 18~19	52	83	Major fracture
XKT 21	35	57	Major fracture
XKT 22	28	75	Major fracture
XKT 23	70	34	Major fracture
XKT 24	18	79	Major fracture
XKT 25	16	52	Major fracture
XKT 26	156	80	Major fracture
XKT 27	324	85	Major fracture
XKT 28	342	56	Major fracture
XKT 29	346	49	Major fracture
XKT 30	340	82	Major fracture
XKT 34	312	53	Major fracture
XKT 35	16	72	Major fracture
XKT 36	345	78	Major fracture
XKT 37	337	56	Major fracture

summarizes the dip angles and orientations of these cracks based on field surveys. The open-pit slopes of the western orebody were formed long ago; during the transition from open-pit to underground mining, extensive surface movements occurred, accompanied by varying degrees of slope landslides. Following the completion of underground mining, the western orebody experienced prolonged slow ground deformation, resulting in the current self-adaptive stable slope angles of the open-pit slopes.

However, tailings mining will expose the original slope toe, reducing the slope’s anti-sliding capacity and increasing landslide risk. Based on the locations and orientations of monitoring points, a slope fracture characteristics model was established. From west to east, the slope was divided into seven sections at 75-m intervals (Figure 2). The southern flank slope of Section 1 contains two major fractures; Section 2 has three; Section 3 has one; Section 4 has two; Section 5 has one; Section 6 has one; and Section 7 has two major fractures (

Table 2. Attitude parameters of sliding surfaces in each section.

Section Number	Number of Major Fractures	Fracture Dip Angle/°	Fracture Length/m
Section 1	2	80~86	0.5~7
Section 2	3	80~86	0.5~7
Section 3	1	84~96	3~8
Section 4	2	84~96	3~8
Section 5	1	72~116	1~10
Section 6	1	72~116	1~10
Section 7	2	72~116	1~10

).

3. Physical and Mechanical Testing of Tailings

3.1. Experimental Design

Wet drilling was employed during sampling, resulting in the loss of the in-situ water content of the tailings. According to previous studies, the maximum water content of the tailings is approximately 25%; therefore, a 25% water content was adopted for micromechanical parameter calibration. Under original depositional conditions, the water content of the tailings increases approximately linearly with depth. To systematically characterize the physical and mechanical properties of the tailings, laboratory tests were conducted on tailings samples prepared with five water content levels (7%, 10%, 15%, 20% and 25%) (

Table 3. Water content mixing scheme for tailings samples.

Scheme Number	Water Mass/g	Dry Tailings Mass/g	Water Content $\mathit{\omega }$/%
Section 1	35	500	7
Section 2	50	500	10
Section 3	75	500	15
Section 4	100	500	20
Section 5	125	500	25

). Prior to sample preparation, the moist tailings were oven-dried and conditioned for 86 h to convert cohesive wet lumps into loose, dry granular material.

3.2. Results of Shear Strength Tests on Tailings

The direct shear tests on the tailings were conducted using an SDJ-1A hand-cranked electric direct shear apparatus. The equipment consists of a rotating handwheel, shear box, proving ring, dial gauge, transmission rod and sliding steel balls. A vertical load was applied to the upper part of the shear box containing the tailings specimen. Then, the motorized device was engaged to drive the handwheel at a constant horizontal rate, applying shear force to the sample. The test setup is illustrated in Figure 3.

The applied compressive stress ranged from 0.1 to 0.4 MPa, with an increment of 0.1 MPa. A constant shear rate of 0.8 mm/min was adopted. The dimensions of the specimens were consistent with those of the cutting ring, with a diameter of 61.8 mm and a height of 20 mm.

The dynamometer calibration coefficient was determined as 1.89 through initial adjustment. Based on Equation (1), the shear strength of tailings samples under different normal stresses was calculated. In accordance with Standard for Soil Test Methods (GB/T 50123—2019) [14], the Coulomb formula (Equation (2)) was adopted to perform linear fitting between the normal stress and shear strength derived from all test results at each water content; the intercept and slope of the fitted line represent cohesion and internal friction angle, respectively (Figure 4). Shear test results indicate that cohesion is lowest at a water content of 25% and highest at 10%. As tailings water content increases with depth, shear strength decreases toward both the surface and deeper zones (

Table 4. Shear strength parameters of tailings at different water contents.

Test ID	Water Content/%	Compressive Stress/10−3 MPa	Dial Gauge Reading of the Proving Ring/mm	Shear Strength/10−3 MPa	Cohesion/10−3 MPa	Friction Angle/°
1	7	100	22.39	42.32	15.95	14.0
		200	32.49	61.41
		300	50.39	95.23
		400	60.37	114.09
2	10	100	27.78	52.51	17.45	15.9
		200	32.34	61.12
		300	58.46	110.48
		400	69.41	131.19
3	15	100	24.80	46.88	16.5	15.7
		200	35.90	67.86
		300	54.35	102.73
		400	68.08	128.68
4	20	100	22.30	42.14	15.2	15.6
		200	37.21	70.32
		300	54.61	103.21
		400	65.68	124.14
5	25	100	22.26	42.08	14.6	15.8
		200	34.03	64.32
		300	50.75	95.91
		400	63.13	119.31

). This trend is likely attributed to the bonding effect of bound water at moderate moisture levels, which enhances particle cohesion. At higher water contents, tailings transition toward a fluid state, reducing interparticle bonding and thereby lowering cohesion.

$${\tau }_f = C \times R$$

(1)

In Equation (1), ${\tau }_f$ is the shear strength of the soil (${10}^{-3}$ MPa); C is the calibration coefficient of the proving ring (${10}^{-3}$ MPa/0.01 mm); and R is the dial gauge reading of the proving ring (0.01 mm).

$$\tau ={\sigma }_{n}tan\phi +c$$

(2)

In Equation (2), $\tau$ is the shear strength (${10}^{-3}$ MPa); ${\sigma }_n$ is normal stress (${10}^{-3}$ MPa); $\phi$ is internal friction angle (°); c represents Cohesion (${10}^{-3}$ MPa).

4. Assessment of Slope Stability Affected by Tailings Recovery

4.1. Design of Simulation Scheme

To formulate a reasonable mining plan, it is first necessary to determine the feasible mining depth of the tailings body. As illustrated in Figure 5, in Sections 1 and 2, Tailings 1 and 2 are separated by waste rock; in Section 7, Tailings 1 and 3 are similarly separated by waste rock. In Sections 3 to 6, only Tailings 1 is intersected by waste rock. The thickness of Tailings 1 is generally approximately 40 m across most sections, except for Section 6, where it thins to ~23 m.

Considering that tailings materials lack strong bench-forming characteristics during actual excavation, a top-down layered mining approach is adopted. In the numerical simulation, the excavation depth for each layer is set to 5 m. During downward excavation, if slope failure or collapse occurs, excavation is terminated at that depth. To ensure the reliability and stability of the simulation results, each excavation step in the model is simulated for a total of 30,000 time steps.

4.2. Construction of the Discrete Element Model and Calibration of Particle Mechanical Parameters

(1) 

Slope model construction

Based on the profiles shown in Figure 2, discrete element models of the tailings were established.

A two-dimensional DEM-based model was employed, with model lengths ranging from 870 to 1000 m and heights from 310 to 350 m. Given the relatively large model size, using smaller particles would lead to extremely high computational costs, rendering the simulation process challenging. Since this study focuses on the influence of tailings mining depth on slope stability, a 1 m particle size was adopted—sufficient to capture the macroscopic landslide evolution induced by tailings excavation. Continuous and discontinuous fracture features of each section were incorporated into the model via a discrete fracture network (DFN) approach. Consequently, each section model contained approximately 36,000–39,000 spherical particles and ~6000 DFN fractures. Figure 6 presents the DEM models for each section, where solid lines denote the fracture features identified during field investigations on the southern flank slope.

Among the seven sectional models, the southern flank slope of Section 1 contains two major fractures, while Section 2 includes three. These major fractures exhibit relatively steep dip angles and are the primary contributors to southern flank slope landslides during tailings mining. Consequently, particular attention is given to the major fracture characteristics of Sections 1 and 2, as well as their corresponding landslide mechanisms, in subsequent stability analyses.

(2) 

Calibration of microscopic particle parameters

As shown in Figure 7, a uniaxial compression calibration model was established. To ensure the number of particles in the model does not affect the calibration accuracy, the model’s height-to-particle diameter ratio was set to 100:1, and the width-to-particle diameter ratio to 50:1. During the calibration of linear contact model parameters, a trial-and-error method was used to iteratively adjust the parallel bond model parameters. Ultimately, the calibrated mesoscopic particle bonding parameters yielded elastic modulus and compressive strength values closely matching the average values from historical data and mechanical test results, as listed in

Table 5. Calibrated microscopic particle parameters.

Stratum 1	Cohesion/MPa	Tensile Strength to Cohesion Ratio	Bond Modulus/103 MPa	Linear Modulus/103 MPa	Bond Stiffness Ratio	Linear Stiffness Ratio	Friction Coefficient	Bond Friction Angle/°
Br	55.3	2.4	31.3	29.2	2.0	2.2	0.38	42.1
CMN	87.3	2.6	27.2	28.4	2.2	2.2	0.42	41.5
RAT	57.2	1.8	17.8	18.7	2.0	2.0	0.39	40.2
RSC	97.3	2.4	43.1	48.2	2.1	2.0	0.41	41.8
RSF	62.8	1.9	9.0	10.8	2.2	2.1	0.38	42.3
SD	138.2	2.3	34.4	38.1	2.0	2.2	0.38	42.1

¹ Employed to differentiate strata belonging to distinct geological periods.

. The corresponding error values are summarized in

Table 6. Comparison of simulation values and actual values errors.

Stratum 1	Compressive Strength Test Value/MPa	Compressive Strength Simulation Value/MPa	Error/%	Elastic Modulus Test Value/103 MPa	Elastic Modulus Simulation Value/103 MPa	Error/%
Br	80.5	82.47	2.45	51.9	52.5	1.2
CMN	124.17	123.0	0.9	47.85	47.9	0.1
RAT	75.3	75.7	0.5	32.2	31.6	1.8
RSC	135.3	135.8	0.4	79.3	77.4	2.4
RSF	82.0	81.9	0.1	17.1	16.9	1.1
SD	204.5	194.3	4.9	67.1	67.0	0.1

¹ Employed to differentiate strata belonging to distinct geological periods.

, with the maximum error not exceeding 5%.

In addition to calibrating the rock parameters of the aforementioned strata, it was also necessary to calibrate the parameters of tailings and waste rock. Based on the tailings shear strength parameters listed in Table 4, the contact variables of mesoscopic particles were adjusted accordingly. First, a shear test model (Figure 8) was constructed with dimensions of 50 × 50. During the shear tests, normal stresses of 0.1 MPa, 0.2 MPa, 0.3 MPa and 0.4 MPa were applied, and the shear strength under each normal stress was measured. Using the Mohr-Coulomb criterion, the cohesion and internal friction angle of the simulated samples were determined.

The calibrated shear strength parameters of the tailings and waste rock are presented in

Table 7. Mesoscopic parameters of tailings and waste rock.

Sample	Cohesion/MPa	Tensile Strength to Cohesion Ratio	Bond Modulus/103 MPa	Linear Modulus/103 MPa	Bond Stiffness Ratio	Linear Stiffness Ratio	Friction Coefficient	Bond Friction Angle/°
Tailings	154.0	1.0	12.0	14.0	1.1	1.1	0.23	10.9
Waste rock	667.0	2.8	98.0	109.0	1.1	1.1	0.38	29.5

, with the fitting curves shown in Figure 9. The fitted cohesion and internal friction angle of the tailings are $14.8\times {10}^{-3}$ MPa and 14.7°, respectively, while those of the waste rock are $38.5\times {10}^{-3}$ MPa and 24.1°, respectively.

4.3. Analysis of the Tailings Mining Process

For the seven slope models constructed above, calculations were performed based on the proposed 5 m per stage excavation depth. Excavation was halted whenever a landslide occurred, enabling determination of the tailings mining status at the point of slope instability for each section (Figure 10).

In Section 1, Tailings 1 was fully extracted without any landslide occurrence (Figure 10a). In Section 2, when approximately 10 m of Tailings 1 remained, a large-scale landslide developed on the southern flank slope, while the northern waste rock slope exhibited relatively minor landslides with localized sliding depths of up to 4.5 m (Figure 10b).

In Section 3, full extraction of Tailings 1 did not induce landslides on the southern flank slope; however, a significant landslide occurred near the base of the northern waste rock slope, with a sliding depth of ~12 m (Figure 10c). Similarly, in Section 4, after complete mining of Tailings 1, no landslides were observed on the southern flank slope, but a large-scale landslide developed on the northern waste rock slope, with the landslide height closely matching the tailings mining depth (Figure 10d).

In Section 5, the southern flank slope remained stable after full tailings extraction, while the northern waste rock slope experienced some degree of landsliding with a maximum sliding depth exceeding 6 m (Figure 10e). In Section 6, no landslides were observed on either adjacent open-pit slope after full tailings extraction (Figure 10f).

Finally, in Section 7, following complete extraction of Tailings 1, the southern flank slope remained stable, but the northern waste rock slope exhibited substantial sliding with a maximum sliding depth exceeding 12 m (Figure 10g).

5. Study on Mitigation Measures for Unstable Zones

Based on the analysis of slope responses to tailings recovery, the following stabilization measures are proposed for the southern and northern flank slopes:

Section 1:

After the complete extraction of Tailings 1, no slope failure was observed on either the southern or northern flank slopes. Consequently, the urgency for stabilization measures in this section is considered low.

Section 2:

Prior to the full recovery of Tailings 1, a large-scale translational failure occurred along the slip surface of the southern flank. To mitigate this issue, a certain volume of tailings may be intentionally retained at the toe of the southern flank slope to serve as a stabilizing buttress. Accordingly, it is necessary to determine the optimal quantity of retained tailings and the corresponding slope angle to ensure adequate resistance against further displacement.

Northern Flank Slopes in All Sections:

Slope failure was identified along the northern waste rock flank slope across all sections. To address this, the critical stable slope angle for the waste rock material should be assessed. Any segment exceeding this critical angle should be excavated or reprofiled to restore long-term stability.

5.1. Simulation Scheme for the Determination of Safe Slope Angles of Waste Rock and Tailings

To determine the critical safe slope angles for waste rock and tailings, a numerical model was established, as shown in Figure 11. The particle-scale mechanical properties of waste rock and tailings were adopted from Table 7. The model dimensions were set to 400 m in width and 160 m in height. After applying gravitational loading and achieving static equilibrium, the internal force chain distribution within the particle assemblies was obtained, which served as the basis for subsequent stability analysis. Additionally, it should be noted that the present numerical simulation is based on static DEM, with the effects of rainfall and seismic loading not temporarily considered.

Given that waste rock generally exhibits superior mechanical strength compared to tailings, it is expected to sustain steeper and higher slopes under similar conditions. Therefore, four slope angles (20°, 30°, 35° and 40°) were selected for the waste rock slope analysis. For each angle, various slope heights were considered, resulting in 16 unique slope geometry combinations (

Table 8. Design schemes for waste rock slope configurations.

Slope Height (H)/m	Slope Angle (α)/°
20	20	30	35	40
40	20	30	35	40
60	20	30	35	40
80	20	30	35	40

). These simulations aimed to identify the critical angle-height combinations beyond which slope failure is initiated.

For tailings, which exhibit lower shear strength and cohesion, more conservative slope geometries were adopted. Slope angles of 15°, 20°, 25° and 30° were selected, combined with slope heights of 5 m, 10 m, 15 m and 20 m. This also resulted in 16 simulation scenarios, as summarized in

Table 9. Tailings slope design schemes.

Slope Height (H)/m	Slope Angle (α)/°
5	15	20	25	30
10	15	20	25	30
15	15	20	25	30
20	15	20	25	30

. The objective was to evaluate the critical slope angles and corresponding heights for tailings under the current site-specific material and geometric conditions.

5.2. Analysis of Simulation Results for Safe Slope Angles of Waste Rock and Tailings Slopes

Figure 12 presents the simulation results of slope failure for waste rock slopes under varying slope heights and angles. When the slope angle was 20°, no failure signs were observed for any tested slope height (Figure 12a). At a slope angle of 30°, slopes with heights of 20 m and 40 m remained stable without visible deformation. However, at a height of 60 m, noticeable displacement was detected near the slope toe and mid-slope surface, with a maximum displacement of ~0.15 m. Although deformation occurred, its magnitude was relatively small and unlikely to compromise operational safety. When the slope height increased to 80 m, a more extensive sliding zone developed, with a maximum displacement of ~0.6 m (Figure 12b).

For slopes at v, stability was maintained for heights below 60 m. At 60 m, a large-scale shear zone formed within the slope body, with a maximum displacement of 0.45 m. As the slope height further increased to 80 m, the entire slope exhibited significant instability characteristics, with a maximum displacement of up to 0.8 m (Figure 12c).

When the slope angle was increased to 40°, global slope failure occurred when the slope height exceeded 60 m. In addition to widespread deformation, localized particle detachment from the slope mass was observed, indicating structural breakdown and loss of overall slope integrity (Figure 12d).

Based on the above analysis, the simulated waste rock slope at a 30° angle can support a maximum height of up to 60 m. When the slope angle increases to 35°, the maximum sustainable height decreases to less than 60 m. However, considering that after stripping at a 35° slope angle, the waste rock height on the northern flank slope does not exceed 60 m, this angle is deemed acceptable. Conversely, at a 40° slope angle, significant slope displacement occurs, making this option unsuitable. Therefore, by integrating the estimated stripping volume and actual post-stripping slope height, a 35° slope angle is recommended for stabilizing the unstable waste rock slopes.

Figure 13 shows the displacement simulation results of tailings slopes under different slope angles and heights. At a 15° slope angle, slopes taller than 10 m generally experience varying degrees of displacement, with maximum displacements below 0.5 m (Figure 13a). Similar behavior is observed at a 20° slope angle, where slopes exceeding 10 m also exhibit displacement, but maximum displacement remains under 0.5 m (Figure 13b). At 25°, with a slope height of 20 m, displacement zones within the slope become more extensive and maximum displacement increases to 0.6 m (Figure 13c). At 30°, displacement patterns resemble those at 25°, showing large internal movements for slopes of 20 m height (Figure 13d).

Based on the above simulation analysis of tailings slope stability, it is evident that varying degrees of slope displacement occur once the slope height reaches 15 m. Comparing the simulation results for the four slope angle scenarios, when the slope height reaches 20 m, the displacement range of tailings slopes with angles of 15° and 20° is relatively smaller. Considering both tailings extraction efficiency and slope safety, a 20° slope angle is recommended as the retained tailings berm angle for the southern flank slope of Section 2.

5.3. Analysis of Tailings Mining Slope Management Based on the Safe Angles of Waste Rock and Tailings Slopes

(1) 

Slope management simulation for each section

For Section 1, only a small amount of waste rock slope exists on the north wing. After the complete extraction of Tailings Deposit No. 1, no instability was observed on the southern flank. Therefore, no urgent slope treatment is required for Section 1.

For Section 2, slope failure occurs on the north-wing waste rock slope when the tailings mining depth reaches 25 m. As mining continues to a depth of 35 m, large-scale sliding occurs on the southern flank. Consequently, for Section 2, it is necessary to retain a certain amount of tailings as a buttress on the southern flank slope, while the north-wing waste rock slope should be cut back to a 35° slope angle. Three different schemes are proposed for the retained tailings slope on the southern wing for comparison:

Scheme 1: Tailings are mined down to a depth of 20 m, and further mining continues below at a 20° tailings slope angle (Figure 14a).

Scheme 2: Tailings are mined down to a depth of 15 m, and further mining continues below at a 20° tailings slope angle (Figure 14b).

Scheme 3: Tailings are mined down to a depth of 15 m, then mining continues downward for an additional 10 m at a 20° slope angle, followed by the retention of a 10 m wide bench. Mining then continues downward again, repeating this process until the tailings slope reaches the bottom of Tailings Deposit No. 1 (Figure 14c).

The results of the three simulation schemes for Section 2 are shown in Figure 15. Under Scheme 1, when tailings mining reaches a depth of 35 m, a large-scale landslide occurs on the southern flank slope, similar to the pre-treatment sliding condition. Under Scheme 2, after the complete extraction of Tailings 1, the southern flank slope experiences an overall landslide. In contrast, under Scheme 3, no landslide occurs on the slope when mining reaches the bottom of Tailings 1.

Based on the above analysis, Scheme 3—mining tailings to a depth of 15 m, then continuing downward mining at a 20° slope angle for an additional 10 m, followed by retaining a 10 m wide bench, and repeating this process until the tailings slope is retained to the bottom of Tailings Body No. 1—is suitable for mining in Section 2. However, it should be noted that although no landslides occur on the southern flank slope, significant displacements are observed in the retained tailings slope. This is mainly due to the inherent instability of tailings, which impairs slope integrity. Nonetheless, the 10 m benches retained in this scheme help suppress large-scale landslides of the tailings slope. Additionally, in terms of residual tailings volume, Scheme 3 yields nearly 100,000 tons of residual tailings, compared to only 72,000 tons in Scheme 2. Therefore, Scheme 3 provides greater anti-sliding resistance at the slope base.

For the northern flank slopes of Sections 3 to 7, the waste rock slopes were cut back to a 35° slope angle. The simulation results, shown in Figure 16, indicate that no slope failures occurred in any section under this cut-back scheme.

(2) 

Analysis of recoverable tailings volume under mechanical excavation

By acquiring the contour of recoverable tailings zones from Sections 1 to 7 (see Figure 17), a model of the recoverable tailings area under mechanical excavation was established. The estimated recoverable tailings volume via mechanical excavation is approximately 3.45 million m³. Among this volume, about 100,000 m³ of tailings near Section 2, adjacent to the southern flank slope, need to be reserved to serve as a buttress for slope stabilization. The reserved tailings near Section 2 for Tailings Deposit No. 1 are shown in Figure 17.

6. Conclusions

This study evaluates the impact of tailings extraction on slope stability in the western ore body, focusing on the physical and mechanical properties of tailings and waste rock, and proposes suitable slope treatment measures. Direct shear strength tests on tailings with water content ranging from 7% to 25% revealed that the highest cohesion ($17.45\times {10}^{-3}$ MPa) occurred at a 10% water content. This is attributed to the formation of bound water, which enhances the cementing capacity of tailings at moderate moisture levels; in contrast, higher water content reduces shear strength.

Simulations based on seven cross-sectional models were conducted to assess the effect of excavation depth on slope stability. In Section 1, no slope failure occurred after full tailings extraction. In Section 2, large-scale slope failure was observed when excavation reached 35 m. For Sections 3–7, the waste rock slopes on the northern flank experienced failure after reaching a certain excavation depth; this phenomenon is primarily due to the increased slope angles and heights following tailings removal.

Slope treatment strategies were also analyzed in detail. For Section 2, it was found that reserving tailings at the base of the southern flank could effectively prevent slope failure. Applying a 35° slope angle to the northern waste rock slopes also achieved failure prevention. A targeted tailings reservation plan was developed for Section 2: extraction is conducted in 10-m increments at a 20° slope angle, with 10-m-wide platforms reserved at each level until reaching the base of the first tailings deposit. This approach results in a tailings reserve of approximately 100,000 m³ in Section 2.

Overall, this study employs a 2D DEM-based model to conduct stability analysis of tailings slopes, simulates the sliding behavior of tailings under different excavation depths, and provides a reference for slope control during tailings mining. Subsequent research will be extended to 3D models, and the effects of water factors, cohesion and internal friction angle parameters on the stability of tailings slopes will be considered.