Slope Stability Analysis of Hamata Tailings Dam, Hidden Valley Mine, Papua New Guinea

 

Las Kuri^1,2, Takashi Sasaoka¹, Hideki Shimada¹, Akihiro Hamanaka¹, Cho Thae Oo^1,3

¹Department of Earth Resources Engineering, Faculty of Engineering, Kyushu University, Fukuoka 819-0395, Japan

²Regulatory Operations Division, Mineral Resources Authority (MRA), Papua New Guinea

³Department of Geology, University of Kyaing Tong, Myanmar

corresponding author:  kurilars@gmail.com 

 ABSTRACT

Construction and management of Tailing dams in Papua New Guinea (PNG) is faced with many challenges such as high altitude with high rainfall (2000-5000 mm/yr), high seismicity and structurally controlled zones which pose threat to the slope stability of tailings dams. Therefore, slope stability analysis is necessary to give confidence to some extent to the stakeholders. The location for this study is at Hamata Tailings (dam) Storage Facility (TSF) at Hidden Valley Mine in PNG which has two rock/earth filled embankments, the main dam and the saddle dam with downstream construction method. Currently the TSF owner is planning to raise the dam height from RL 2000 to RL 2015 with extra 15 Mt storage capacity as the pond water approaching its designed capacity at RL 2000. The objective of this study is to analyse the slope stability of Hamata TSF using phase 2 based on the design basics for the crest expansion from RL 2000 to RL 2015 and beyond and recommend an ideal slope stability under various conditions in terms of shear strength reduction factor ((SSRF). The results obtained in this study is useful for PNG Mining Regulators in comparing company results in the appraisals for tailings dam development proposals and, it will be useful to future researchers in PNG and other similar tropical regions.

 Keywords: Tailings dam, slope stability analysis, Hamata TSF crest expansion, embankment, Shear Strength Reduction Factor, RL-Reduced Level(m).

 INTRODUCTION

Tailing dam construction in PNG are faced with natural factors such as high altitude with high rainfall (2000-5000 mm) coupled with high seismicity zones and geological/geotechnical conditions which pose threat to the stability of tailing dams. One of the learned experience is the case of Ok Tedi tailings dam failure in 1984 (Griffiths et al. 2004). After this incident, the PNG government allowed mining companies to discharge tailings into the river systems and on to the sea floor (deep-sea tailings placement (DSTP)) which pollutes the riverine and ecology within the vicinity of the mine impacted natural environment and communities downstream and the marine lives respectively.  However, the PNG Government amended the Environment Act to abolish riverine tailings discharge and encourage tailings dam construction in PNG.

In compliance with the PNG government’s intention, the Hidden Valley mine and the K92 mine have constructed tailings dams respectively and store their tailings in the facilities overcoming all odds. However, management of the tailings dam under challenging environment is one of the key concerns of the dam owners to make sure the dam is stable throughout the operation till closure and post closure. On the other hand, the mining regulators and the impacted communities downstream also concern about the stability of the dam as it will affect their livelihood in an unlikely event of failure.

 In this study, it is proposed to review and assume Hamata Tailings Dam designs and embankment material properties to evaluate the slope stability conditions of the TSF under various geotechnical/soil parameters.

 Study Location – Hidden Valley Mine

The Hidden Valley(HV) Mine (coordinates: 7⁰27’17” S,146⁰40’24” E) in PNG operates the Hamata Tailings dam.  Hidden Valle Mine is an Open pit gold-silver mine located in Morobe Province, about 210 km North West (NW) of Port Moresby. The Mining Lease was Granted in 2005 for 20 years and renewal upon expiry. The Lease holder is Harmony Gold Ltd. Mine development Construction started in 2007 and commercial production began in September 2010.

 Figure 1 PNG map (Courtesy of Mineral Resources Authority) showing location of Hidden Valley Mine (Circled).

 Mine Layout

The mining lease area has two main mine pits which are about 6 km apart and mining at three main ore bodies which are named as Hidden Valley- Kaveroi(HVK) and Hamata epithermal gold and silver deposits.  The Hidden Valley and Keveroi Ore deposits are close to each other while the Hamata ore body is on its own. Ore mined from HVK is transported via belt conveyor to the processing plant near Hamata pit. 

The mine has a total mineral resources of 68.776 Mt at Hidden Valley Kaveroi deposits with a metal content of 3.307 Moz Au and 57.270 Moz Ag while the Hamata deposit has a total mineral resource of 2.216 Mt ore with metal content of 0.133 Moz Au as of June 2019 (HV Annual Report-2020).

 Figure 2 Hidden Valley Mine Plan (Rynhoud et al ,2017)

 Hamata Tailings (Dam) Storage Facility (TSF)

The Hamata Tailings (Dam) Storage Facility is constructed using the downstream method with two earth and rock filled embankments, the saddle dam and the main dam. The dam Construction commenced in June 2007 and the starter embankment construction was completed in February 2009 (Rynhoud et al, 2017). The embankments are constructed using the waste rock/materials from the two mine pits at HVK and Hamata.

Klohn Crippen Berger Ltd (KCB) is the design engineer for the Hamata TSF, (Rynhoud et al, 2017). The main dam and the saddle dam is designed to a maximum crest elevation of RL 2000 with a storage capacity of about 40 Mt of tailings with a mill throughput of 4.2 Mtpa (Rynhoud et al, 2017). The height of the dam from the main dam is about 145 m at the RL 2000 crest.

Figure 3 Hamata Tailings Dam, Main dam at NW and Saddle dam at SE (Google Image-7°25'36.6"S146°38'32.0"E)

 Problem Statement

Tailings deposition and sedimentation at Hamata TSF result in ponded water approaching dam crest elevation at RL 2000, the miner proposed to raise the dam height to RL 2015 with extra 15 Mt tailings storage capacity.

The foundation of expansion (RL 2015) embankment is likely to begin at RL1960 to RL 1970 of the RL 2000 design. With the pond water seeping through the embankments coupled with high rainfall, the geotechnical parameters are altered over time and displacement of embankment is anticipated under wet conditions and/or seismic conditions and potential dam slope failure is anticipated in a worse case scenario.

The focus of this study is to review available options to minimize significant displacement of embankment under various stress conditions.

 Significance of Study and Research Advancement

Related literatures of slope stability analysis of tailings dam in PNG is rarely available online except the design basics of dam published by Rynhoud et al 2017 and Murray et al 2010. There are also publications of tailings dam about Frieda River Mining Project and Ok Tedi Mine of which most of the data from this study is obtained from all these publications.

Further research can be done beyond this study in terms of slop stability analysis of tailings dam under various geotechnical and seismic conditions in similar tropical regions.

 Objective Of The Study

The objective of this study is to analyse the slope stability of the Hamata Tailings dam construction and operation of crest expansion from RL 2000 to RL 2015 and recommend an ideal risk factor of safety under various stress conditions. The study adopts Finite Element Analysis in Phase2 software to analyse the slope stability conditions of the dam for RL 2015 based on the design basics and material properties of the dam embankment. Design basics are modified for the purpose of modelling and may not represent construction design specifications.

Slope Stability conditions are expressed in terms of Shear Strength Reduction Factor (SSRF or SRF) and the corresponding displacement under stress conditions.

 METHODOLOGY

Study methodology is designed in a way to review related literatures of the past and collect field data including design parameters and proceed with modelling. Results from the model are interpreted to make conclusion and necessary recommendation is anticipated.

In the case of field research which is impossible at hand, related data from other projects in both PNG and abroad are borrowed for the purpose of modelling in this study.

General information regarding mining in PNG are reviewed and adopted some scripts in this paper. Most of the data at hand is obtained from both unpublished and published literatures related to the Mining in PNG and off-course Hamata TSF.

Most of the material property data is expected to be borrowed from Frieda River Mining Project and other publications and reports which are referenced in this paper.

 Design Basics

The design basics are adopted from the published papers by Murray et al, 2010 and Rynhoud et al, 2017 for RL 2000 and assumptions are made for RL 2015 in terms of construction methodology. A combination of upstream, centerline and downstream method of construction is assumed to analyze the slope stability condition of the dam using the phase 2 software. The maiden modified design in the model is shown in Figure 4.

Figure 4 Proposed expansion design (model-cross section, main dam) for RL 2015 (modified from Murray et al, 2010)

 The foundation of expansion embankment is at RL 1960 and RL 1970 in the model design. The expansion (RL 2015) design for the model is modified from the RL 2000 design published by Murray et al, 2010.  

This study adopted material property data from the proposed Frieda River Tailings Dam in PNG which has similar embankment fill materials to that of Hamata TSF. 

 Modeling And Results

Based on the borrowed embankment material property/parameters and design basics data, a maiden model was built in Phase2 and computed to observe the behaviour of the TSF embankment. The material strength parameters used in the model are shown in the Table 1. Mohr-Coulomb failure criteria is used for all plastic materials type computed in the model.

 Table 1 Material strength parameters

Material	Unit Weight (MN/m3)	Cohesion  (MPa)	Friction angle(o)
Bedrock	0.02	0.25	45
Boulder Colluvium	0.018	0	30
Lacustrine(clay-assume)	0.02	0.03	20
Alluvium(Organic)	0.018	0	15
Oxide	0.018	0	25
Random Fill (Oxide)	0.018	0	25
Fresh rock fill	0.023	0.6	25
Gravel filter drain	0.021	0	35
Tailings	0.02	0.001	34

 The results obtained from the maiden model shows that potential failure is anticipated. At the critical SRF of 1.4, the total displacement is 0.153 m at the embankment. Ground water conditions and permeability are not computed in the maiden model but will consider in the preceding models.

  

Figure 5(a) Shear Strain, underground water seepage flow rate at main dam embankment in model. (b) Total Total Displacement. (c) Shear Strength Reduction Curve.

Figure 6 Model Results at various SRF in terms of Shear Strain and Displacement Progression at increasing SRF.

 DISCUSSION and CONCLUSION

Results indicate that shear strain is concentrated along the chimney drain and almost steady at all stages of SRF. Displacement is significant and vary at all stages.  At Critical SRF of 0.48, Maximum Displacement is 0.251m.   Maximum displacement is observed at the foundation of RL2015 expansion.  The    Toe of TSF has insignificant shear strain and is stable but weight of displaced materials can induce stress at the toe to be unstable over time. Thus, it requires more attention in this regard.

The Weight of RL2015 foundation cause the maximum shear strain at the chimney drain/ channel in the model and thus displacement at the crest of RL 2015 and along the slope of the downstream embankment.  Seepage water might cause the saturation of embankment materials and failure is anticipated during wet conditions and/or seismic activity.

Thefore there following measures will be taken in the next phase of this study:

q  Model

ü  Variation of model parameters and input data

ü  Analyze other Sections of the TSF.

q  Review Counter Measures to stabilize the unstable slope conditions:

ü  Construction Method

ü  Fill material variations

ü  Geotechnical support systems –i.e. geogrid

ü  Design parameter variations etc..

 

ACKNOWLEDGEMENTS

The authors would like to acknowledge the Harmony South East Asia and Klohn Crippen Berger Ltd (KCB) for permitting previous researchers to make use of unpublished consulting works dealing with the design and construction of the Hamata TSF at the Hidden Valley mine. Acknowledgment and appreciation is also conveyed to staff at Australian National Committee on Large Dams (ANCOLD) for making available the published papers for further researches.

Acknowledgement and thank you to Japan International Cooperation Agency (JICA) both in PNG and Japan for the opportunity and support towards this research through the KIZUNA program.

Finally, acknowledgement and appreciation to the Mineral Resources Authority of Papua New Guinea for the support in this research in terms of fundamental data for PNG Mining Industry and Hidden Valley Mine.

Note: This publication is a work in progress and several articles will be published If you want full paper of this publication and the advanced information regard then contact  us via contact form.

 REFERENCES

Rynhoud, Mark Stephen, Johns, David and Murray,         Len., The Hamata Tailings Dam Design and Construction at the Hidden Valley Mine in Papua New Guinea (2017).

Rick Friedel, Len Murray, Gerrad Suter, James     Penman, Hidden Valley - Design and Construction of Highland Papua New Guinea First Tailings Dams (2010).

G. Vessia and D. Di Curzio,Lacustrine Deposits, Engineering and Geology. Department (InGeo), University “G. d’Annunzio” of Chieti-Pescara, Chieti Scalo/Chieti, Italy (2018).

J.S. Griffiths, J.N. Hutchinson, D. Brunsden, D.J. Petley & P.G. Fookes, The reactivation of a landslide during the construction of the Ok Ma tailings dam, Papua New Guinea (2004).

Simone Oberhollenzer, Franz Tschuchnigg, Helmut F.Schweiger. Finite element Analyses of Slope stability problems using non-associated plasticity. Journal of Rock Mechanics and Geotechnical Engineering (2018).

Hidden Valley Mine Annual Report 2020. (Unpublished)

Frieda River Copper-Gold Project, PNG. Waste Rock Dump Report - Feasibility Study Report (2016).  (Unpublished)

 

Read More

Waste Manage Management

Waste Management is basically planning and taking the responsibility of where to dispose off what is not needed. Waste is simply something that you don't need it for a particular task or purpose. However, some materials or waste are not really are waste. They can be recycled or used in elsewhere.

So to manage waste, one has to consider, what is really waste, what is recyclable and what is required at other places or purposes. Some waste are burnable and considered waste but then the burnable waste are also given second thought as to whether their remains after burning is usable or not. For example, organic materials are burnable but their remains after burning can be of useful for agricultural compost. 

Other waster are considered recyclable in the likes of metals and hydrocarbon products. For example, can drinks are recyclable as well and pet bottles are recyclable. 

So in managing waste, you have to consider the separation of these waste materials which you don't need but  others they need for recycling or  to be used for other purposes.

Some of the waste can be used for craft making and landform beautification and gardening. People have different ways to utilize waste materials into useful materials. But what that you consider waste must be disposed off in the right place so that the next person might need or will be a waste indeed.

So different countries have different waste management policies and regulations which all the citizens are bound to follow so that  the living environment or the surrounding environment including the natural environment is not polluted or affected by the waste produced or created by human beings.

Read More

Clean up of Mine Waste in Papua New Guinea

 The Papua New Guinea's (PNG) Minister for Environment and Protection stated on the daily news paper dated October30, 2020 that an Hong Kong based company would be engaged to clean-up the mine waste in PNG at no cost to PNG government and the mining operators.

The company targets the river deltas where the Ok Tedi Mine and Porgera Mine dispose their mining waste. And the overseas based company is kind enough to clean up the mine waste in PNG.

The Minister may not be aware thinking that it is a good idea to get rid of the mine waste without realizing the value of the contents in it, Or he may be very well knowingly trying to get these assumed valuable residues from OK Tedi and Porgera which could be full of gold and other minerals. The Map interestingly showing the targets are Porgera and Ok Tedi river deltas. The foreign investor is not a non-sense to collect the mine waste from PNG deltas targeting Ok Tedi and Porgera tailings disposal rivers.  The tailings contain gold other gangue minerals which are unable to extract at the processing plants in both Porgera and Ok Tedi Mining.  And because both mines discharger their tailings into the river system, the sediments of these rivers (Strickland and Ok Tedi River) are rich in alluvial gold which can be extracted with improved techniques that can recover gold that are of fine particles.

 

It is good the company is interested in the mine waste disposal area and PNG should welcome the idea as it is an investing opportunity. however,  the company's approach is quite not right and must re-strategize and must make its hidden intention known by way of exploration proposal or Mining Proposal rather than saying they want to clean-up mine waste. They should apply for mineral tenement/leases and do the right thing with the PNG Mining Regulator which is Mineral Resources Authority.  

 

It must not be opposed but allow them to  progress and re-phrase to mine alluvial gold from  mine waste disposal areas rather then clean-up mine waste at the deltas. The deltas are rich in alluvial gold and industrial minerals if not known. It can only be proven with exploration and sampling.

Screen shot of The National News Paper publication..

 

Read More

Analysis of Flood in Mul District that caused 6 lives and Catastrophic destruction to properties

The flooding of Kuma Creek has caused massive destruction to properties and confirmed six fatalities downstream. Kuma Creek is such a small creek which is  a tributary of Gumanch River which joins with other rivers to form the Wagi River in the Western Highlands Province.

It is unbelievable for such a small creek to cause massive destruction to lives of people and properties downstream. According to preliminary report posted on Facebook dated 4th February 2020 by Stanley Kheel Kewa, it reads: 

"Preliminary reports from Mt Hagen confirm massive scale of destruction by the Kuma river a tributary of the Gumanch river in Mul district of Western Highlands Province. Four adults and five children totaling nine casualties as reported deaths now. More investigations are in progress as surrounding communities are assessing and investigating the magnitude of the destruction.

Local tribes in the area are the Nengka, Munjika & Mele tribes. Locals reporting from Hagen say this is one of the worst natural disasters the community has ever experienced since time immemorial. The Kuma & Gumanch rivers originate from the top peak of the highest mountain range in WHP known as the Mt Hagen range from which the current Hagen city got its name.

The Nengka Kuiprungils, Nengka Oiyambs and Munjika Rapgangils live at the edge of the Hagen range with houses and gardens patched along the Gumanch and Kuma tributaries.

Ken Paul is a local from the area and reports he is in Hagen town trying to mobilize disaster office and news personnel into the area for further investigations and reporting.

This is just a preliminary report with photos of the disaster zone downloaded from fb pages."

Locals on site - photo courtesy of Facebook

Photo Courtesy of The National Newspaper

Debris of flood - Photo courtesy of facebook

MarapanaVillage aftermath - photo by National newspaper

Now,

one would wonder with questions in anticipating superstitions without establishing the facts and without even having a curiosity in mind. The possible cause of the flood can be best explained as follows;

There must be couple of landslips

caused by what is believed to be over saturated water-table/reservoir

contain by permeable rocks

at both steep

sides of the wedge walls/hills

of  Kuma Creek which is indicated on the snapshot below. 

Then the slipped materials must have

formed an embankment or base which blocked the upstream and the water built up at the upper end of the embankment which formed a temporary mini dam. 

As the mini dam rose with altitude, the stress build up also increased until it reached a

bursting failure in which debris of embankment together with other slipped materials along the creek's

pathway were all washed away and flooded the banks of Kuma and Gumanch Rivers which caused the catastrophic destruction to properties and fatality of 6 human lives. 

The mass flow of loose materials which blocked the flowing river which resulted in forming a mini dam were not competent or strong enough to withstand the pressure/stress build up at the upper end of the blockage, it then burst out and flooded the downstream at a greater momentum which is possible for massive destruction.

So sad that  many loved ones lost their lives due to the catastrophic disaster caused by this unusual flood.

Expected failed area

Location Failure is Expected

Marapana Village

Ariel view of Kuma,Tagla Kwip and Marapana

Note  that this analysis is based on opinion only and not substantiated with facts. If someone wants to proof with factual information then someone need to take a walk up the Kuma river and look for any trace of landslip. If that is so then that would be the cause of the flooding. 

To prevent properties and lives, build houses on higher grounds and also build flood walls along the river banks where valuable properties are installed. Do make awareness to kids and matured people to evacuate quick if unexpected signals are given before massive destruction happens again.

Read More