Article Numerical Simulation of Seepage and Stability of Dams: A Case Study in Lixi, China

Chen Zhang 1, Junrui Chai 1,2,* , Jing Cao 1 , Zengguang Xu 1, Yuan Qin 1 and Zongjie Lv 1

1 State Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an 710048, China; [email protected] (C.Z.); [email protected] (J.C.); [email protected] (Z.X.); [email protected] (Y.Q.); [email protected] (Z.L.) 2 School of Civil Engineering, Xijing University, Xi’an 710123, China * Correspondence: [email protected]

 Received: 16 February 2020; Accepted: 6 March 2020; Published: 8 March 2020 

Abstract: The purpose of establishing a tailings dam is to safely store tailings to protect the natural environment from damage. Accidents at tailings dams are frequent, however, with serious consequences of not only threats to life and property but also the pollution of the environment. Many tailings dam accidents are caused by seepage failure. In this paper, the object of the case study is the Lixi tailings dam. Three- and two-dimensional finite element models are established. The seepage field of the project under different working conditions is simulated and the position of the phreatic line is obtained. The safety factors under different working conditions are obtained by combining the seepage field with the stable surface. Finally, the influence of different dry and upstream slope ratio on seepage and stability of tailings dam is obtained. The results show that the longer the length of the dry beach, the lower the phreatic line and the greater the safety factor. The higher the upstream slope ratio, the lower the phreatic line and the greater the safety factor.

Keywords: tailings dam; seepage; stability; numerical simulation

1. Introduction With the acceleration of industrialization in China, demand for mineral ores is also gradually increasing [1,2]. Ore tailings are generated, and waste residues are discharged after useful components are extracted [3,4]. Tailings are generally recycled or stored [5]. Tailings used in concrete additives can be recycled [6,7]. A tailings is a good place to store these waste residues [8]. According to survey data, more than 20,000 tailings have been created worldwide, and the larger ones can reach 10 million or even more than 100 million m3. For example, the storage capacity of the Yang Kedutel tailings dam in the United States reaches 150 million m3 [9]. As a kind of artificial high-potential debris flow, tailings dams are different from ordinary dams in terms of the water blocked by them [10]. Leakage of the tailings dam body will cause more serious damage than leakage of ordinary reservoir dams [11,12]. The tailings dam is a mixture of tailings and water, and seepage phenomena are inevitable [13]. The hazards caused by a tailings pond discharge accident rank 18th, next to those caused by earthquakes, cholera, floods, and the explosion of a hydrogen bomb [14]. A dam failure at a copper mine in Zambia killed more than 100 people. The Pure Pierre tailings dam accident in Italy killed 251 people. More than 1000 Guyanese people were killed when drinking water was contaminated after the Omai tailings dam in Guyana collapsed [15]. The Baia Mara mineral processing plant in Romania dumped 100,000 m3 of cyanide wastewater into nearby rivers, resulting in the pollution of the river, affecting the ecological environment of the Black Sea region, and causing the destruction of all the plants and animals in the path of tailings flowing through the region [16]. After the Bang Lake accident

Water 2020, 12, 742; doi:10.3390/w12030742 www.mdpi.com/journal/water Water 2020, 12, 742 2 of 12 in the United States, more than 60,000 m3 of tailings pond drained into a swamp, resulting in the death of all the plants in the swamp [17]. The Karamken gold tailings dam in Russia’s far east crashed in 2009, causing damage to the site and environment [18].About 32 million cubic metres of tailings were released when the Brazilian Fundao tailings dam crashed in 2015.The pollution reached 668 km and extended into the Atlantic ocean [19,20]. China is a major mining country, with more than 1500 tailings dams and 300 million tons of tailings discharged annually. The design height of the largest tailings dam at present is 260 m. The failure of Anhui’s Huang Meishan iron mine tailings dam caused 119 casualties and 282 acres of farmland were damaged by pollution. At present, the most serious tailings dam accident in China was that of the Xinta mining company in the Tanshan mining area, Xiangfen county, Linfen city, Shanxi. That accident caused 258 deaths, 34 injuries, and environmental pollution damage [21]. Most of the research on tailings dams focuses on the seepage, consolidation, stability, and seismic characteristics of tailings dams [22]. At present, most tailings dams are wet-piled, for which the study of seepage control and slope stability is very important [23]. Seepage also influences the control of stability. Damage caused by seepage flow accounts for 20% to 30% of tailings dam accidents [24]. Zandarin et al. conducted a simulation study of the seepage field and capillary water of a tailings dam in Cuba under heavy rain conditions [25]. They believed that the capillary water greatly influences the stability of the tailings dam, with the stability being proportional to the height of the capillary water. Lu and Cui generalized a complex section and simplified topographic conditions, representing the tailings dam by a plane model [26]. Through the analysis and numerical simulation of the seepage flow field, the simplification was found to have little influence on the engineering simulation results. Xu et al. considered the special situation of the tailings dam physics and chemical clogging, and studied its seepage characteristics [27]. Consequently, Liu Chong et al. proposed a new type of drainage system for seepage and stability analysis of tailings dams [28]. This new drainage system reduces the wetting line to meet stable requirements. Pak and Nabipour studied saturation and unsaturation in different drainage systems and simulated the seepage under different conditions [29]. Yin et al. studied the mesoscopic mechanism of the tailings dam and obtained the microscopic changes under the seepage [30]. Yonghao Yang et al. made a case study on the application of geotextile tubes in tailings dam. The results show that this method is effective in construction [31]. Shaohua Hu et al. studied the seepage control of a tailings dam during its staged construction. They believe that a proper drainage system is of great significance to the reduction in the phreatic surface [32]. Many domestic and foreign scholars have studied the stability of tailings dams. Ozer and Bromwell used a limit equilibrium method and a finite element method to do seepage transient analysis of tailings dams under the condition of extremely rapid water level decline, and coupled this with the analysis of slope stability under that condition [33]. Meggyse et al. believe that there are many factors affecting the stability of a tailings dam, including the choice of building materials, tailings pond management, and the tailings slurry composition [34]. Wang et al. conducted a safety evaluation on the Xiangyun Phosphogypsum tailings pond. The keys to safety analysis and assessment—inspection and verification—can be mutually beneficial, and can lead to a better overall result [35]. Sanjay et al. proposed a simplified method to evaluate the seismic stability of the tailings dam [36]. Mahdi et al. numerically simulated the stability of the high line tailings dam under earthquake conditions, and the results showed that the dam under study was unstable under strong earthquake conditions [37]. In this case study, the Lixi tailings dam is taken as the research object. The research mainly focuses on analyzing its seepage field and stability. The dry spreading length and upstream slope ratio of the tailings dam are of concern in this paper. Following implementing the finite element method, the simulation adopts the seepage field and the stable field combination analysis method. The results of the seepage field are sorted, and the positions of phreatic line under different working conditions are obtained. This result is input into the stability analysis to obtain the safety factor under different working conditions. The analysis results can be used as a reference for the control of dry beach length and the upstream slope ratio in the tailings dam project. Water 2020, 12, x FOR PEER REVIEW 3 of 12 Water 2020, 12, 742 3 of 12 The Lixi tailings dam is located east of the Qinling Mountains in Lixi ravine, Jindui town, Huaxian county, Shanxi province (Figure 1). The dam was designed by the Beijing Nonferrous 2. Project Overview MetallurgyWater 2020, 12, Designx FOR PEER and REVIEW Research Institute in 1973 and construction was started by the 10th metallurgy3 of 12 constructionThe Lixi tailings company dam of is the located Ministry east ofof theMetallurgi Qinlingcal Mountains Industry in in Lixi 1977. ravine, Construction Jindui town, was Huaxiancompleted county,and Thethe Shanxi damLixi provincetailingswas put dam (Figureinto is use located1). in The 1983. east dam Theof wasthe dam designedQinling is designed Mountains by the with Beijing in a Lixi final Nonferrous ravine, stacking Jindui Metallurgydam town, with an DesignelevationHuaxian and Researchcounty,of 1300 Shanximeters, Institute province a intotal 1973 storage (Figure and construction capacity 1). The damof was 165 was started million designed by cubic the by 10th meters, the metallurgy Beijing a total Nonferrous constructiondam height of Metallurgy Design and Research Institute in 1973 and construction was started by the 10th metallurgy company164.5 m, of and the a Ministry service life of Metallurgical of 32 years. Originally Industry inconsidered 1977. Construction a second-class was completedtailings pond and (now the dama first- construction company of the Ministry of Metallurgical Industry in 1977. Construction was completed wasclass put tailings into use pond), in 1983. the The initial dam dam is designed was built with as a a permeable final stacking dam dam made with of anrubble elevation and was of 1300 piled m, up and the dam was put into use in 1983. The dam is designed with a final stacking dam with an a totalwith storage directional capacity blasting. of 165 The million dam crest cubic elevation meters, a is total 1176.5 dam m, height the top of length 164.5 m,is 115.0 and a m, service the top life width of elevation of 1300 meters, a total storage capacity of 165 million cubic meters, a total dam height of 32is years. 4.0 m, Originally and the top considered low width a second-class is 157.0 m. tailingsThe dam pond is 40.5 (now m ahigh. first-class The upstream tailings pond), and downstream the initial 164.5 m, and a service life of 32 years. Originally considered a second-class tailings pond (now a first- damslopeclass was tailings ratios built asof pond), athe permeable long-term the initial dam damdam made arewas of1:1.7 built rubble andas a and 1:2.0,permeable was respectively. piled dam up made with There directionalof rubble is a 2.0 and blasting.m washorse piled Thetrail up dam1156.0 crestmwith downstream elevation directional is 1176.5 blasting.and a m,0.7 The the to 1.0 topdam m length crest reverse elevation is 115.0filter m,islaye 1176.5 ther at top them, width the upstream top is length 4.0 end. m, is and 115.0The the later m, top the dams low top width are raised is 157.0byis 4.0 m.the m, Theupstream and dam the istop stacking 40.5 low m width high.method. is The 157.0 The upstream m. sub-bats The dam and in downstreamis each 40.5 periodm high. slopeare The 3 upstream ratiosmeters of high, theand long-term anddownstream the loader dam is areusedslope 1:1.7 toratios and ship 1:2.0, of the the respectively. results.long-term The dam Theretailings are is1:1.7 pipe a 2.0 and mis 1:2.0,la horseid onrespectively. trail the 1156.0 top of There m the downstream sub-is a 2.0dam m forhorse and decentralized atrail 0.7 1156.0 to 1.0 m ore reversedischarge,m downstream filter layerand theand at thetailingsa 0.7 upstream to are1.0 mdeposited end.reverse The filterafter later layenatural damsr at the areclassification upstream raised by end. theto form upstreamThe alater late-stage dams stacking are accumulation raised method. Thedam.by sub-bats the The upstream outer in each slope stacking period of the aremethod. dam 3 m is high, The stepped, sub-bats and the and loaderin the each de is signedperiod used to aretotal ship 3 slopemeters the results.ratio high, is and The1:5. It tailingsthe is loadercovered pipe is with is laidsoil-bearingused on theto ship top gravelthe of the results. and sub-dam hasThe a tailings forgrid-like decentralized pipe drainage is laid ore ravion discharge,thene. topThe of sectional andthe sub- the viewdam tailings isfor shown aredecentralized deposited in Figure ore after 2. The naturallayoutdischarge, classification of the and project the tailings to formis shown are a late-stage deposited in Figure accumulation after 3. natural dam.classification The outer to form slope a oflate-stage the dam accumulation is stepped, and thedam. designed The outer total slope slope of ratio the dam is 1:5. is stepped, It is covered and withthe de -bearingsigned total slope gravel ratio and is has 1:5. a It grid-like is covered drainage with ravine.soil-bearing The sectional gravel and view has is showna grid-like in Figuredrainage2. Theravi layoutne. The ofsectional the project view is is shown shown in in FigureFigure3 2.. The layout of the project is shown in Figure 3.

Figure 1. The location of the tailings dam.

FigureFigure 1. 1.The The locationlocation of the tailings tailings dam. dam.

Figure 2. The dam sectional view.

Water 2020, 12, x FOR PEER REVIEW 4 of 12

Figure 2. The dam sectional view. Water 2020, 12, x FOR PEER REVIEW 4 of 12

Figure 2. The dam sectional view.

Water 2020, 12, 742 4 of 12

Figure 3. The layout of the project. Figure 3. The layout of the project.

3.3. Model Model and and Computational Computational Method Method Figure 3. The layout of the project. 3.1.3.1. Simulation Simulation Model Model 3. Model and Computational Method InIn this this study, study, ADINA ADINA was usedwas toused simulate to simulate the three-dimensional the three-dimensional (3D) seepage (3D) field, seepage and SEEP field,/W and in 3.1.theSEEP/W Simulation Geo-Studio in theModel software Geo-Studio was usedsoftware to simulate was used the to seepage simulate field the of seepage the two-dimensional field of the two-dimensional (2D) maximum profile.(2D) maximum The results profile. were coupledThe results using were SLOPE coupled/W software using SLOPE/W for the safety software coeffi forcient the calculation. safety coefficient In this study, ADINA was used to simulate the three-dimensional (3D) seepage field, and calculation.The model was simplified in the process of modeling. When modeling in Geo-Studio, SEEP/W in the Geo-Studio software was used to simulate the seepage field of the two-dimensional the dimensionalThe model parameters was simplified of the in model the process are measured of modeling. in CAD When from Figuremodeling1. In in the Geo-Studio, X direction, the (2D) maximum profile. The results were coupled using SLOPE/W software for the safety coefficient thedimensional extension directionparameters of of the the stack model dam are is measured divided and in CAD taken. from The Figure boundary 1. In the between X direction, different the calculation. materialsextension of direction the dam of is the distinguished stack dam is by divided a broken and line. taken. The The materials boundary at between different different locations materials of the The model was simplified in the process of modeling. When modeling in Geo-Studio, the damof the are dam reduced is distinguished to bedrock, tail by silt a broken and tail line. medium The sand.materials The at sub-dams different at locations different of elevations the dam are are dimensional parameters of the model are measured in CAD from Figure 1. In the X direction, the simplifiedreduced to as bedrock, a straight tail line silt from and the tail initial medium dam sand to the. The dam sub-dams top. The at elevation different of elevations the foundation are simplified surface extension direction of the stack dam is divided and taken. The boundary between different materials wasas a 1120 straight m, and line the from elevation the initial of the dam simulated to the da thirty-fifthm top. The pile elevation dam was of 1330the foundation m. The length surface of thewas of the dam is distinguished by a broken line. The materials at different locations of the dam are simulation1120 m, and in the the X directionelevation of of the th modele simulated is 2000m. thirty-fifth The curve pile is simplifieddam was to1330 a straight m. The line length connected of the reduced to bedrock, tail silt and tail medium sand. The sub-dams at different elevations are simplified atsimulation different points. in the X The direction resulting of the model model is shown is 2000m. in Figure The curve4. is simplified to a straight line connected as a straight line from the initial dam to the dam top. The elevation of the foundation surface was at different points. The resulting model is shown in Figure 4. 1120 m, and the elevation of the simulated thirty-fifth pile dam was 1330 m. The length of the simulation in the X direction of the model is 2000m. The curve is simplified to a straight line connected at different points. The resulting model is shown in Figure 4.

FigureFigure 4. 4.The The 2D 2D model model in in geo-studio. geo-studio.

WhenWhen modeling modeling in ADIN,in ADIN, the modelingthe modeling scope scope was designed was designed to take intoto take account into theaccount entire the research entire area.research In the area dam. bodyIn the left dam and body rightFigure left bank 4. andThe (Y 2D direction)right model bank in extend geo-studio.(Y direction) outward extend 100m. Atoutward the upstream 100m. ofAt the the damupstream (X direction), of the dam the length(X direction), of the drythe length bench isof approximatelythe dry bench is equal approximately to the length equal of theto the fill length dam. When modeling in ADIN, the modeling scope was designed to take into account the entire Forof the fill convenience dam. For the of modeling, convenience take of themodeling, number take of its the range number as an of integer. its range Therefore, as an integer. the modeling Therefore, . researchrange was area selected In the as dam 2000m body in theleft Xand direction right bank and1600m (Y direction) in the Y extend direction. outward The elevation 100m. At of the the upstreamfoundation of the model dam at (X the direction), starting pointthe length in the of Z the direction dry bench is 1100m. is approximately First, the terrain equal is to divided the length into ofdi thefferent fill dam. sizes For and the these convenience pieces are of modeled. modeling, Then, take the the number Boolean of operation its range isasused an integer. in ADINA. Therefore, These pieces of bedrock are integrated into a complete foundation. The obtained foundation model is shown in Figure5. It is then combined with Figures2 and3. The 3D model had the same partitioning as WaterWater 2020 2020, 12, x12 FOR, x FOR PEER PEER REVIEW REVIEW 5 of5 12 of 12

thethe modeling modeling range range was was selected selected as as2000m 2000m in inthe the X directionX direction and and 1600m 1600m in inthe the Y direction.Y direction. The The elevationelevation of theof the foundation foundation model model at theat the starting starting po intpo intin thein the Z directionZ direction is 1100m.is 1100m. First, First, the the terrain terrain is dividedis divided into into different different sizes sizes and and these these pieces pieces are ar modeled.e modeled. Then, Then, the the Boolean Boolean operation operation is usedis used in in ADINA.ADINA. These These pieces pieces of bedrockof bedrock are are integrated integrated into in ato complete a complete foundation. foundation. The The obtained obtained foundation foundation Watermodelmodel2020 is, 12 shownis, 742 shown in Figurein Figure 5. It5. is It thenis then combined combined wi thwi Figureth Figure 2, Figure2, Figure 3. The3. The 3D 3D model model had had the 5the same of 12same partitioningpartitioning as theas the previous previous 2D 2D one, one, and and the the pile pile dam dam was was also also divided divided into into a bedrock a bedrock layer, layer, a tailings a tailings siltsilt layer, layer, and and a tailings a tailings medium medium sand sand layer. layer. Then, Then, according according to theto the stratification stratification of differentof different sections sections theof ofthe previous the dam, dam, 2Dthe the one,stratification stratification and the pileis established.is damestablished. was alsoTh eTh dividedreste rest of ofdam into dam is a bedrockmodeled.is modeled. layer, Finally, Finally, a tailings through through silt Boolean layer,Boolean andoperationoperation a tailings in mediumADINA,in ADINA, sandsubtracting subtracting layer. Then, the the overlapping according overlapping to pa the rtpa of stratificationrt foundationof foundation of and di andff erentdam dam body, sections body, the ofthe subtracted the subtracted dam, theinterface stratificationinterface is theis the interface is interface established. between between The dam restdam body of body dam and and is su modeled. rroundingsurrounding Finally, bedrock. bedrock. through The The 3d Boolean 3dmodel model of operation theof the tailings tailings in ADINA,damdam obtained subtractingobtained using using the the overlappingthe Boolean Boolean op erationop parteration of foundationis shownis shown in andFigurein Figure dam 6. body, 6. the subtracted interface is the interfaceTheThe between material material damof theof bodythe dam dam andbody body surrounding was was divided divided bedrock. into into thre thre Thee parts:e parts: 3d modeltail tail silt, silt, of tail the tail medium tailings medium sand, dam sand, and obtained and initial initial usingdamdam theand and Boolean bedrock. bedrock. operation The The material material is shown properties properties in Figure of theof6. the different different parts parts are are shown shown in Tablein Table 1. 1.

FigureFigureFigure 5. 5.Foundation Foundation 5. Foundation model. model. model.

FigureFigureFigure 6. 6.The The 6. The 3d 3d model 3dmodel model of of tailings tailingsof tailings dam. dam. dam.

The material of the dam body was divided into three parts: tail silt, tail medium sand, and initial dam and bedrock. The material properties of the different parts are shown in Table1. The size and quantity of mesh are very important in finite element analysis. When the mesh generation is less, the calculated results cannot meet the requirement of accuracy. When the mesh generation is more, the computational workload is large and the time required for processing is Water 2020, 12, 742 6 of 12

Water 2020, 12, x FOR PEER REVIEW 6 of 12 long. Applying a tetrahedralTable element 1. The tomaterial the terrain properties gives of the it different a strong parts. adaptability. Therefore, we use a tetrahedron as a unit for subdivision. In order to meet the requirements of accuracy, the calculation is Name K (m/s) ρ (kg/m3) CPa（） Φ (°) E (Pa) μ not very large. By trial and error, the length of each unit is selected to be 50 m. Mapped and Free-Form Bedrock 3.7 × 10−7 subsections are used. The model after grid generation is shown in Figure7. The bedrock was divided Initial dam 2 × 10−4 1900 1.6 × 104 30 1.8 × 109 0.33 into 90,742Tail units. medium The sand initial dam 3.6 is × divided10−3 into 1810 387 units. 1.5 × The104 tailings26 silt4.1 layer × 109 was0.30 divided in to 9209 units. TailingsTail silt medium sand 2.2 layer × 10−3was divided 1890 in to 1.5 8211 × 10 units.4 21 2.1 × 109 0.33 The size and quantity of mesh are very important in finite element analysis. When the mesh generation is less, the calculatedTable 1. The results material cannot properties meet the of requirement the different parts.of accuracy. When the mesh generation is more, the computational workload is large and the time required for processing is long. Name K (m/s) ρ (kg/m3) C (Pa) Φ ( ) E (Pa) µ Applying a tetrahedral element to the terrain gives it a strong adaptability.◦ Therefore, we use a 7 tetrahedronBedrock as a unit for subdivision. 3.7 10− In order to meet the requirements of accuracy, the calculation × Initial dam 2 10 4 1900 1.6 104 30 1.8 109 0.33 is not very large. By trial and error,× − the length of each unit ×is selected to be 50 meters.× Mapped and Tail medium sand 3.6 10 3 1810 1.5 104 26 4.1 109 0.30 Free-Form subsections are used.× The− model after grid generation× is shown in Figure× 7. The bedrock Tail silt 2.2 10 3 1890 1.5 104 21 2.1 109 0.33 was divided into 90,742 units. ×The −initial dam is divided into× 387 units. The tailings× silt layer was divided in to 9209 units. Tailings medium sand layer was divided in to 8211 units.

FigureFigure 7. 7.The The modelmodel after grid grid generation. generation.

3.2. Boundary3.2. Boundary Conditions Conditions and and Calculation Calculation Conditions Conditions The downstreamThe downstream water water level level boundary boundary isis inin the initial initial dam. dam. After After the the dry dry beach, beach, the upstream the upstream waterwater level level boundary boundary is is added. added. Both of of these these bounda boundariesries are areadded added to the to surface the surface on which on they which are they are located.located. In theIn simulationthe simulation of of di ffdifferenterent working working conditions,conditions, the the effect effect of of dry dry beach beach length length and andupstream upstream slopeslope ratio ratio was was considered. considered. Under Underthe the influenceinfluence of of these these two two factors, factors, the the seepage seepage and andstability stability of of tailings dam are simulated. Under the influence of only the length of the dry beach, the upstream tailings dam are simulated. Under the influence of only the length of the dry beach, the upstream slope ratio is kept constant. The slope ratio of upstream the dry bench is 1:100. The length of the dry slope ratio is kept constant. The slope ratio of upstream the dry bench is 1:100. The length of the dry benches are set as 100m and 200m, respectively. Then, to simulate the effect of the upstream slope benchesratio, are we set keep as 100mthe drying and 200m,length respectively.constant. The Then,length toof simulatethe dry bench the e ffisect 100m. of the The upstream upstream slopeslope ratio, we keepratio the is 1:100, drying 1:200 length and 1:300, constant. respectively. The length of the dry bench is 100m. The upstream slope ratio is 1:100, 1:200 and 1:300, respectively. 4. Case Study Results 4. Case Study Results 4.1. Seepage Field Simulation in ADINA 4.1. Seepage Field Simulation in ADINA The above working conditions were calculated using the temperature field module in ADINA. TheWe determined above working the total conditions head chart were and calculated position of using the phreatic the temperature line. To make field the module resultsin more ADINA. We determined the total head chart and position of the phreatic line. To make the results more intuitive, we selected the appropriate section for the 3d model. The result is presented as a contour map. For the seepage of tailings dam, we are more concerned with the location of the phreatic line. The phreatic Water 2020, 12, x FOR PEER REVIEW 7 of 12

intuitive, we selected the appropriate section for the 3d model. The result is presented as a contour map. For the seepage of tailings dam, we are more concerned with the location of the phreatic line. The phreatic line is also known as the lifeline of the tailings dam. The seepage simulation results under different working conditions can be seen in Figure 8(a) through Figure 8(d). The total head cloud and phreatic line also can be seen in Figure 8. Figure 8(a) and Figure 8(b) are the calculated results of dry beach lengths of 100m and 200m. Figure 8(b), Figure 8(c) and Figure 8(d) are the results of the upstream slope ratios 1:100, 1:200 and 1:300, respectively. In order to obtain the influence of different factors on the phreatic line, we analyzed situations where only a single factor changed. In the Z direction, the starting position of the elevation is the bottom of the model. In the X direction, the position of the starting point is the leftmost boundary of the model. With these two positions as the starting point, the position of the infiltration line is measured. The height of the phreatic line under the ratios of different lengths of dry beach to the upstream slope was measured. The position of the phreatic line under different dry bench lengths is shown in Figure 9. The position of phreatic line under different upstream slope ratios is shown in Figure 10. The height of the phreatic line near the downstream of the dam decreases less. The height of the phreatic line near the upstream of the dam has decreased a lot. The upstream slope ratio changes from 1:100 to 1:200 to 1:300 when the length of the dry stand remains the same. The maximum values of the phreatic line rising near the upstream are 7.5m and 15m, and the minimum values are 4.3m Waterand2020 5m., 12 ,The 742 maximum values of the phreatic line dropping near the downstream are 19 m and 13 m,7 of 12 and the minimum values are 0.1 m and 0.14 m. The phreatic line with a slope ratio of 1:300 is 11 m lower than the phreatic line with a slope ratio of 1:200 in the outflow seepage point. Therefore, when linethe is alsophreatic known line asis at the different lifeline dry of the bench tailings lengths; dam. the The longer seepage the dry simulation bench, the results lower the under height diff oferent workingthe phreatic conditions line. The can height be seen of in the Figure phreatic8a–d. line The near total the headupstream cloud of andthe dam phreatic decreased line also significantly can be seen in Figurewhen 8the. Figure phreatic8a,b line are was the at calculated a different results upstream of dry slope beach ratio. lengths At the upstream of 100 m andof the 200 dam, m. the Figure higher8b–d arethe upstream results of slope the upstream ratio, the higher slope ratiosthe elevation 1:100, 1:200of theand phreatic 1:300, line. respectively. In the downstream of dam, the larger the upstream slope ratio, the lower the height of the phreatic line.

(a). Dry beach lengths of 100m.

Water 2020, 12, x FOR PEER (b).REVIEW Dry beach lengths of 200m (Upstream slope ratio 1:100). 8 of 12

(c). Upstream slope ratio 1:200.

(d). Upstream slope ratio 1:300.

Figure 8. Seepage calculation with drybeach lengths and Seepage calculation with upstream slope Figure 8. Seepage calculation with drybeach lengths and Seepage calculation with upstream slope a b ratio.ratio.( ( )a) Dry Dry beach beach lengths lengths ofof 100100 m; m; (b) ( ) Dry Dry beach beach lengths lengths of 200 of 200 m (Upstream m (Upstream slope slope ratio ratio1:100); 1:100);(c) (c) UpstreamUpstream slope ratio 1:200; 1:200; (d) (d )Upstream Upstream slope slope ratio ratio 1:300. 1:300.

420 100m 200m 410 400 390 380 370 360 Height(m) 350 340 330 320 310 300 400 500 600 700 800 900 1000 1100 Distance(m)

Figure 9. The position of the phreatic line under different dry bench length.

Water 2020, 12, x FOR PEER REVIEW 8 of 12

(c). Upstream slope ratio 1:200.

Water 2020, 12, 742 8 of 12

In order to obtain the influence of different factors on the phreatic line, we analyzed situations where only a single factor changed. In the Z direction, the starting position of the elevation is the bottom of the model. In the X direction,(d). theUpstream position slope of the ratio starting 1:300. point is the leftmost boundary of the model. With these two positions as the starting point, the position of the infiltration line is measured. The heightFigure of8. Seepage the phreatic calculation line under with thedrybeach ratios lengths of different and Seepage lengths calculation of dry beach with to upstream the upstream slope slope was measured.ratio.(a) Dry The beach position lengths of of the 100 phreatic m; (b) Dry line beach under lengths different of 200 dry m bench(Upstream lengths slope is ratio shown 1:100); in Figure(c) 9. The positionUpstream of slope phreatic ratio 1:200; line under (d) Upstream different slope upstream ratio 1:300. slope ratios is shown in Figure 10.

420 100m 200m 410 400 390 380 370 360 Height(m) 350 340 330 320 310 300 400 500 600 700 800 900 1000 1100 Distance(m)

Water 2020, 12, x FOR PEER REVIEW 9 of 12 FigureFigure 9. 9. TheThe position position of of the the phreatic phreatic line line under under different different dry bench length.

1/100 1/200 1/300 440 420 400 380 360

340 Height(m) 320 300 280 260 240 220 200 150 300 450 600 750 900 1050 1200 Distance(m)

Figure 10. TheThe position of phreatic line unde underr different different upstream slope ratios.ratios.

4.2. StabilityThe simulation Calculation results in Geo-Studio of seepage field show that the height of phreatic line is affected by the length of dry beach and the upstream slope ratio. When the upstream slope ratio remains unchanged, the lengthSlope-W of the in drygeo-studio beach changes was used from for 100 the m stability to 200 m. analysis The average of the height above reduction working in conditions. the infiltration The lineseepage is about results 11 were m, the introduced maximum into is about the stability 27 m, and analysis. the minimum The results is aboutare shown 4.5 m. in TheFigure height 11a,d. of The the phreaticblue line linein the near figure the downstream is the phreatic of theline. dam The decreases calculated less. safety The factor height of of slope the phreaticstability linecan nearbe seen the upstreamfrom Figure of the11a–d. dam The has arc decreased in the figure a lot. is The the upstreamfailure surface. slope ratioThe safety changes factor from can 1:100 be used to 1:200 to describe to 1:300 whenthe stability the length of the of slope. the dry The stand larger remains the valu thee, same.the more The stable maximum the dam values slope of tends the phreatic to be. line rising nearFrom the upstream the results are of 7.5 stability m and simulation, 15 m, and the minimumsafety factor values is affected are 4.3 by m the and length 5 m. of The the maximum dry beach valuesand the of upstream the phreatic slope line ratio. dropping When near the upstream the downstream slope ratio are 19 remained m and 13 unchanged, m, and the minimumthe length valuesof the aredry 0.1beach m and changed 0.14 m. from The 100 phreatic to 200m, line and with the a slopesafety ratio factor of increased 1:300 is 11 from m lower 1.69 to than 1.76, the an phreatic increase line of 0.007. When the length of dry beach remains unchanged, the safety factor changes from 1.76 to 1.85 and 1.92, increasing by 0.09 and 0.12, respectively, and the upstream slope ratio changes from 1:100 to 1:200 and then to 1:300. When the safety factor is at different dry bench lengths, the longer the dry bench, the bigger safety factor of the dam. When the safety factor is at a different upstream slope ratio, the higher the upstream slope ratio, the greater the safety factor.

(a). Stability calculation with dry beach lengths of 100m.

(b). Stability calculation with dry beach lengths of 200m (upstream slope ratio 1:100).

(c). Stability calculation with upstream slope ratio 1:200.

Water 2020, 12, x FOR PEER REVIEW 9 of 12

1/100 1/200 1/300 440 420 400 380 360

340 Height(m) 320 300 280 260 240 220 200 150 300 450 600 750 900 1050 1200 Distance(m)

Water 2020, 12, 742 Figure 10. The position of phreatic line under different upstream slope ratios. 9 of 12

4.2. Stability Calculation in Geo-Studio with a slope ratio of 1:200 in the outflow seepage point. Therefore, when the phreatic line is at different dry benchSlope-W lengths; in geo-studio the longer was the used dry for bench, the stability the lower analysis the height of the of theabove phreatic working line. conditions. The height The of theseepage phreatic results line were near introduced the upstream into of the the stability dam decreased analysis. significantly The results are when shown the phreaticin Figure line 11a,d. was The at ablue diff lineerent in upstream the figure slope is the ratio. phreatic At theline. upstream The calculated of the dam,safety the factor higher of slope the upstream stability can slope be ratio,seen thefrom higher Figure the 11a–d. elevation The arc of the in the phreatic figure line. is the In failure the downstream surface. The of safety dam, thefactor larger can thebe used upstream to describe slope ratio,the stability the lower of the the slope. height The of thelarger phreatic the valu line.e, the more stable the dam slope tends to be. From the results of stability simulation, the safety factor is affected by the length of the dry beach 4.2.and Stabilitythe upstream Calculation slope inratio. Geo-Studio When the upstream slope ratio remained unchanged, the length of the dry beachSlope-W changed in geo-studio from 100 wasto 200m, used and for the the safety stability factor analysis increased of thefrom above 1.69 to working 1.76, an conditions.increase of The0.007. seepage When resultsthe length were of introduced dry beach intoremains the stabilityunchanged, analysis. the safety The results factor arechanges shown from in Figure 1.76 to 11 1.85a,d. Theand blue1.92, lineincreasing in the figure by 0.09 is theand phreatic 0.12, respectively, line. The calculated and the upstream safety factor slope of sloperatio stabilitychanges canfrom be 1:100 seen fromto 1:200 Figure and 11thena–d. to The 1:300. arc When in the the figure safety is the factor failure is at surface. different The dry safety bench factor lengths, can be the used longer to describe the dry thebench, stability the bigger of the safety slope. factor The largerof the thedam. value, When the the more safety stable factor the is damat a diffe sloperent tends upstream to be. slope ratio, the higher the upstream slope ratio, the greater the safety factor.

(a). Stability calculation with dry beach lengths of 100m.

(b). Stability calculation with dry beach lengths of 200m (upstream slope ratio 1:100).

Water 2020, 12, x FOR PEER REVIEW 10 of 12 (c). Stability calculation with upstream slope ratio 1:200.

(d). Stability calculation with upstream slope ratio 1:300.

Figure 11. Stability calculation with dry beach lengths and Stability calculation with upstream slope ratio. (a) (a) 100 m; ((b)b) 200 m (upstream slope ratio 1:100); ((c)c) 1:200;1:200; ((d)d) 1:300.1:300. From the results of stability simulation, the safety factor is affected by the length of the dry beach 5. Conclusions and the upstream slope ratio. When the upstream slope ratio remained unchanged, the length of the dry beachThe calculation changed from of seepage 100 to 200m, and st andability the safetyis very factor important increased for the from design 1.69 to and 1.76, construction an increase of 0.007.tailings When dam.the In this length study, of dry the beach engineering remains overview unchanged, of the the Lixi safety tailings factor dam changes was fromprovided. 1.76 toBased 1.85 andon the 1.92, engineering increasing data by 0.09 of Lixi and tailings 0.12, respectively, dam, the finite and element the upstream model slope is established. ratio changes The from seepage 1:100 and to 1:200stability and are then calculated, to 1:300. and When the thetotal safety head factorchart, isph atreatic different line and dry safety bench coefficient lengths, the of longerseepage the under dry different working conditions are obtained. The major conclusions are summarized as follows: 1. In the seepage analysis. The seepage simulation was carried out with dry bench lengths of 100 m, 200 m and different upstream slope ratios, and the corresponding seepage field was obtained. The results of the seepage simulation under different working conditions were compared. It can be seen that with the same upstream slope, the shorter the length of the dry beach, the higher the elevation of the obtained phreatic line. Under the same length of dry beach, the higher the upstream slope is— the higher the phreatic line is in the upstream, and the lower it is in the downstream. 2. In the stability analysis, it can be seen from the comparison of safety factors under different working conditions, that with the same upstream slope ratio, the longer the length of the dry beach, the higher the safety factor of slope stability. With the same length of dry beach, the higher the upstream slope is, the higher the safety factor of slope stability. The size of the dry beach length and upstream slope ratio is proportional to the size of the safety factor. 3. In the project design and management of tailings dam, within a reasonable range, we should make the dry beach as long as possible and the upstream slope as large as possible. This can make the position of phreatic line lower and the safety factor larger, and the operation of tailings dam tends to remain stable.

Author Contributions: Conceptualization, C.Z., J.C. (Junrui Chai), J.C. (Jing Cao), Z.X., Q.Y. and Z.L.; methodology, C.Z.; software, C.Z.; validation, Z.X.; formal analysis, C.Z.; investigation, C.Z. and Z.L.; resources, C.Z.; data curation, C.Z.; writing—original draft preparation, C.Z.; writing—review and editing, C.Z.; visualization, C.Z.; supervision, C.J.; project administration, J.C. (Junrui Chai); funding acquisition, J.C. (Junrui Chai), J.C. (Jing Cao). All authors have read and agreed to the published version of the manuscript.

Funding: This research was funded by the National Natural Science Foundation of China (Grant No. 51679197), the Natural Science Basic Research Program of Shanxi Province-Key Project (Grant No. 2017JZ013), and the Leadership Talent Project of Shaanxi Province High-Level Talents Special Support Program in Science and Technology Innovation (Grant No. 2013KCT-15).

Conflicts of Interest: The authors declare no conflicts of interest.

References

1. Wang, X.; Zhan, H.; Wang, J.; Li, P. The stability of tailings dams under dry-wet cycles: A case study in Luonan, China. Water 2018, 10, 1048. 2. Wang, K.; Yang, P.; Hudson-Edwards, K.A.; Lyu, W.; Yang, C.; Jing, X. Integration of DSM and SPH to Model Tailings Dam Failure Run-Out Slurry Routing Across 3D Real Terrain. Water 2018, 10, 1087. 3. Asghari, M.; Noaparast, M.; Shafaie, S.Z.; Ghassa, S.; Chelgani, S.C. Recovery of coal particles from a tailing dam for environmental protection and economical beneficiations. Int. J. Coal Sci. Technol. 2018, 5, 253–263. 4. Yagüe, S.; Sánchez, I.; Vigil de La Villa, R.; García-Giménez, R.; Zapardiel, A.; Frías, M. Coal-mining tailings as a pozzolanic material in cements industry. Minerals 2018, 8, 46.

Water 2020, 12, 742 10 of 12 bench, the bigger safety factor of the dam. When the safety factor is at a different upstream slope ratio, the higher the upstream slope ratio, the greater the safety factor.

5. Conclusions The calculation of seepage and stability is very important for the design and construction of tailings dam. In this study, the engineering overview of the Lixi tailings dam was provided. Based on the engineering data of Lixi tailings dam, the finite element model is established. The seepage and stability are calculated, and the total head chart, phreatic line and safety coefficient of seepage under different working conditions are obtained. The major conclusions are summarized as follows: 1. In the seepage analysis. The seepage simulation was carried out with dry bench lengths of 100 m, 200 m and different upstream slope ratios, and the corresponding seepage field was obtained. The results of the seepage simulation under different working conditions were compared. It can be seen that with the same upstream slope, the shorter the length of the dry beach, the higher the elevation of the obtained phreatic line. Under the same length of dry beach, the higher the upstream slope is—the higher the phreatic line is in the upstream, and the lower it is in the downstream. 2. In the stability analysis, it can be seen from the comparison of safety factors under different working conditions, that with the same upstream slope ratio, the longer the length of the dry beach, the higher the safety factor of slope stability. With the same length of dry beach, the higher the upstream slope is, the higher the safety factor of slope stability. The size of the dry beach length and upstream slope ratio is proportional to the size of the safety factor. 3. In the project design and management of tailings dam, within a reasonable range, we should make the dry beach as long as possible and the upstream slope as large as possible. This can make the position of phreatic line lower and the safety factor larger, and the operation of tailings dam tends to remain stable.

Author Contributions: Conceptualization, C.Z., J.C. (Junrui Chai), J.C. (Jing Cao), Z.X., Y.Q. and Z.L.; methodology, C.Z.; software, C.Z.; validation, Z.X.; formal analysis, C.Z.; investigation, C.Z. and Z.L.; resources, C.Z.; data curation, C.Z.; writing—original draft preparation, C.Z.; writing—review and editing, C.Z.; visualization, C.Z.; supervision, C.J.; project administration, J.C. (Junrui Chai); funding acquisition, J.C. (Junrui Chai), J.C. (Jing Cao). All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the National Natural Science Foundation of China (Grant No. 51679197), the Natural Science Basic Research Program of Shanxi Province-Key Project (Grant No. 2017JZ013), and the Leadership Talent Project of Shaanxi Province High-Level Talents Special Support Program in Science and Technology Innovation (Grant No. 2013KCT-15). Conflicts of Interest: The authors declare no conflicts of interest.

References

1. Wang, X.; Zhan, H.; Wang, J.; Li, P. The stability of tailings dams under dry-wet cycles: A case study in Luonan, China. Water 2018, 10, 1048. [CrossRef] 2. Wang, K.; Yang, P.; Hudson-Edwards, K.A.; Lyu, W.; Yang, C.; Jing, X. Integration of DSM and SPH to Model Tailings Dam Failure Run-Out Slurry Routing Across 3D Real Terrain. Water 2018, 10, 1087. [CrossRef] 3. Asghari, M.; Noaparast, M.; Shafaie, S.Z.; Ghassa, S.; Chelgani, S.C. Recovery of coal particles from a tailing dam for environmental protection and economical beneficiations. Int. J. Coal Sci. Technol. 2018, 5, 253–263. [CrossRef] 4. Yagüe, S.; Sánchez, I.; Vigil de La Villa, R.; García-Giménez, R.; Zapardiel, A.; Frías, M. Coal-mining tailings as a pozzolanic material in cements industry. Minerals 2018, 8, 46. [CrossRef] 5. Deng, D.Q.; Liu, L.; Yao, Z.L.; Song, K.I.; Lao, D.Z. A practice of ultra-fine tailings disposal as filling material in a gold mine. J. Environ. Manag. 2017, 196, 100–109. [CrossRef][PubMed] 6. Gao, S.; Cui, X.; Zhang, S. Utilization of Molybdenum Tailings in Concrete Manufacturing: A Review. Appl. Sci. 2020, 10, 138. [CrossRef] Water 2020, 12, 742 11 of 12

7. Lv, X.; Shen, W.; Wang, L.; Dong, Y.; Zhang, J.; Xie, Z. A comparative study on the practical utilization of iron tailings as a complete replacement of normal aggregates in dam concrete with different gradation. J. Clean. Prod. 2019, 211, 704–715. [CrossRef] 8. Kossoff, D.; Dubbin, W.E.; Alfredsson, M.; Edwards, S.J.; Macklin, M.G.; Hudson-Edwards, K.A. Mine tailings dams: Characteristics, failure, environmental impacts, and remediation. Appl. Geochem. 2014, 51, 229–245. [CrossRef] 9. Azam, S.; Li, Q. Tailings dam failures: A review of the last one hundred years. Geotech. News 2010, 28, 50–54. 10. Hatje, V.; Pedreira, R.M.; de Rezende, C.E.; Schettini, C.A.F.; de Souza, G.C.; Marin, D.C.; Hackspacher, P.C. The environmental impacts of one of the largest tailing dam failures worldwide. Sci. Rep. 2017, 7, 1–13. [CrossRef] 11. Ormann, L.; Zardari, M.A.; Mattsson, H.; Bjelkevik, A.; Knutsson, S. Numerical analysis of strengthening by rockfill embankments on an upstream tailings dam. Can. Geotech. J. 2013, 50, 391–399. [CrossRef] 12. ICOLD. Tailings Dams—Risk of Dangerous Occurrences, Lessons Learnt from Practical Experiences; Bulletin 121; United Nations Environmental Programme (UNEP), Division of Technology, Industry and Economics (DTIE) and International Commission on Large Dams (ICOLD): Paris, France, 2001. 13. Hudson-Edwards, K.A.; Macklin, M.G.; Miller, J.R.; Lechler, P.J. Sources, distribution and storage of heavy metals in the Rıo Pilcomayo, Bolivia. J. Geochem. Explor. 2001, 72, 229–250. [CrossRef] 14. Kemper, T.; Sommer, S. Estimate of heavy metal contamination in after a mining accident using reflectance spectroscopy. Environ. Sci. Technol. 2002, 36, 2742–2747. [CrossRef][PubMed] 15. Von der Heyden, C.J.; New, M.G. pollution on the Zambian Copperbelt: Deciphering the source and the risk. Sci. Total Environ. 2004, 327, 17–30. [CrossRef][PubMed] 16. Rico, M.; Benito, G.; Salgueiro, A.R.; Díez-Herrero, A.; Pereira, H.G. Reported tailings dam failures: A review of the European incidents in the worldwide context. J. Hazard. Mater. 2008, 152, 846–852. [CrossRef] [PubMed] 17. Rico, M.; Benito, G.; Diez-Herrero, A. Floods from tailings dam failures. J. Hazard. Mater. 2008, 154, 79–87. [CrossRef][PubMed] 18. Glotov, V.E.; Chlachula, J.; Glotova, L.P.; Little, E. Causes and environmental impact of the gold-tailings dam failure at Karamken, the Russian Far East. Eng. Geol. 2018, 245, 236–247. [CrossRef] 19. Do Carmo, F.F.; Kamino, L.H.Y.; Junior, R.T.; de Campos, I.C.; do Carmo, F.F.; Silvino, G.; Mauro, M.L.; Rodrigues, N.U.A.; de Souza Miranda, M.P.; Pinto, C.E.F. Fundão tailings dam failures: The environment tragedy of the largest technological disaster of Brazilian mining in global context. Perspect. Ecol. Conserv. 2017, 15, 145–151. [CrossRef] 20. Mura, J.C.; Gama, F.F.; Paradella, W.R.; Negrão, P.; Carneiro, S.; De Oliveira, C.G.; Brandão, W.S. Monitoring the vulnerability of the dam and dikes in Germano iron mining area after the collapse of the tailings dam of Fundão (Mariana-MG, Brazil) using DInSAR techniques with TerraSAR-X data. Remote Sens. 2018, 10, 1507. [CrossRef] 21. Yong, W.; Li, X.K.; Xin, Z. Discussion on the causes for tailings dam accidents at home and abroad. Met. Mine 2009, 7, 139–142. 22. Lyu, Z.; Chai, J.; Xu, Z.; Qin, Y.; Cao, J. A comprehensive review on reasons for tailings dam failures based on case history. Adv. Civ. Eng. 2019, 2019, 4159306. [CrossRef] 23. Shahriari, M.; Aydin, M.E. Lessons learned from analysis of Los Frailes tailing dam failure. In International Conference on Applied Human Factors and Ergonomics; Springer: Cham, Switzerland, 2017; pp. 309–317. 24. Yin, G.; Li, G.; Wei, Z.; Wan, L.; Shui, G.; Jing, X. Stability analysis of a copper tailings dam via laboratory model tests: A Chinese case study. Miner. Eng. 2011, 24, 122–130. [CrossRef] 25. Zandarín, M.T.; Oldecop, L.A.; Rodríguez, R.; Zabala, F. The role of capillary water in the stability of tailing dams. Eng. Geol. 2009, 105, 108–118. [CrossRef] 26. Lu, M.L.; Cui, L. Three-dimensional seepage analysis for complex topographical tailings dam. Rock Soil Mech. 2006, 27, 1176–1180. 27. Xu, Z.; Yang, X.; Chai, J.; Qin, Y.; Li, Y. Permeability Characteristics of Tailings considering Chemical and Physical Clogging in Lixi Tailings Dam, China. J. Chem. 2016, 2016, 8147845. [CrossRef] 28. Liu, C.; Shen, Z.; Gan, L.; Xu, L.; Zhang, K.; Jin, T. The Seepage and Stability Performance Assessment of a New Drainage System to Increase the Height of a Tailings Dam. Appl. Sci. 2018, 8, 1840. [CrossRef] Water 2020, 12, 742 12 of 12

29. Pak, A.; Nabipour, M. Numerical Study of the Effects of Drainage Systems on Saturated/Unsaturated Seepage and Stability of Tailings Dams. Mine Water Environ. 2017, 36, 341–355. [CrossRef] 30. Yin, G.; Zhang, Q.; Wang, W.; Chen, Y.; Geng, W.; Liu, H. Experimental study on the mechanism effect of seepage on microstructure of tailings. Saf. Sci. 2012, 50, 792–796. [CrossRef] 31. Yang, Y.; Wei, Z.; Cao, G.; Yang, Y.; Wang, H.; Zhuang, S.; Lu, T. A case study on utilizing geotextile tubes for tailings dams construction in China. Geotext. Geomembr. 2019, 47, 187–192. [CrossRef] 32. Hu, S.; Chen, Y.; Liu, W.; Zhou, S.; Hu, R. Effect of seepage control on stability of a tailings dam during its staged construction with a stepwise-coupled hydro-mechanical model. Int. J. Min. Reclam. Environ. 2015, 29, 125–140. [CrossRef] 33. Ozer, A.T.; Bromwell, L.G. Stability assessment of an earth dam on silt/clay tailings foundation: A case study. Eng. Geol. 2012, 151, 89–99. [CrossRef] 34. Meggyes, T.; Niederleithinger, E.; Witt, K.J.; Csovari, M.; Kreft-Burman, K.; Engels, J.; Roehl, K.E. Enhancing the safety of tailings management facilities. Soil Sediment Contam. 2008, 17, 323–345. [CrossRef] 35. Wang, T.; Zhou, Y.; Lv, Q.; Zhu, Y.; Jiang, C. A safety assessment of the new Xiangyun phosphogypsum tailings pond. Miner. Eng. 2011, 24, 1084–1090. [CrossRef] 36. Nimbalkar, S.; Annapareddy, V.S.R.; Pain, A. A simplified approach to assess seismic stability of tailings dams. J. Rock Mech. Geotech. Eng. 2018, 10, 1082–1090. [CrossRef] 37. Naeini, M.; Akhtarpour, A. Numerical analysis of seismic stability of a high centerline tailings dam. Soil Dyn. Earthq. Eng. 2018, 107, 179–194. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).