Bull Eng Geol Environ (2010) 69:63–69 DOI 10.1007/s10064-009-0243-9
ORIGINAL PAPER
Safety assessment of abandoned tailings ponds: an example from Kirki mines, Greece C. Loupasakis Æ G. Konstantopoulou
Received: 20 December 2008 / Accepted: 10 August 2009 / Published online: 24 September 2009 Ó Springer-Verlag 2009
Abstract At the abandoned hydrometallurgy plant at the Kirki mines, Greece, numerous tailings ponds have been left without adequate consideration for their potential for failure or contamination of the ground water. The paper presents the geotechnical study undertaken to assess the factor of safety of the embankments against sliding and proposes measures to isolate the tailings ponds from the environment and to protect them from potential erosion whilst enhancing the aesthetic environment. Keywords Mining tailings ponds Erosion Transitional sliding Geo-membranes ` l’usine abandonne´e hydrome´tallurgique dans Re´sume´ A les mines de Kirki, Gre`ce, de nombreux e´tangs de re´sidus miniers ont e´te´ laisse´s sans examen ade´quat pour leur risque de de´faillance ou de contamination de l’eau souterraine. Le papier pre´sente l’e´tude ge´otechnique entreprise a` e´valuer le facteur de se´curite´ des talus contre coulissant et a` proposer des mesures visant a` isoler les e´tangs de re´sidus miniers de l’environnement et a` les prote´ger contre l’e´rosion potentielle tandis que ame´liorer l’esthe´tique de la localite´. Mots cle´s e´tangs de re´sidus miniers e´rosion glissements transitoires ge´o-membranes
C. Loupasakis (&) G. Konstantopoulou Engineering Geology Department, Institute of Geology and Mineral Exploration, Olympic Village, Thrakomacedones, 13677 Athens, Greece e-mail:
[email protected]
Introduction Mining tailings ponds, although simple geotechnical constructions, have been much studied because of the great environmental impact caused if they fail. The hydrodynamic failure of the embankments, their response to seismic loading, the seepage flows and the infiltration of the acid fluids, the influence of the erosional procedures and the land planning design of the tailing ponds have been studied by many researchers (Guzenkov and Shul’gina 2007; Lee and Lee 2004; Liu et al. 2007; Finagenov et al. 1999). The present study deals with the safety assessment of the existing flotation tailings ponds of the abandoned Agios Philippos (Kirki) mines. This mixed sulphides mine and the old flotation plant are located close to Kirki village, between 12 and 18 km north of the city of Alexadroupolis, North Greece (Fig. 1). The exploration for Pb and Ag in the area of Kirki began late in the 19th century. In 1932 English mining companies started constructing mine galleries and managed to exploit about 20,000 tons of ore. During the Second World War, the German company ‘‘Thrazische Bergwerke’’ continued exploitation and installed a flotation plant with an 80 ton/ day capacity. In the following years exploration for base metals and Ag was carried out by Greek state-owned and private companies. The ‘‘Evros Mining Co’’ exploited and processed roughly 160,000 ton of ore between 1974 and 1980, from both underground and open pit mining. Although there were several long interruptions, ore exploitation and processing continued from 1990 to 1997 with an estimated additional extraction of 70,000 ton of sulphide ore. The past uncontrolled mining and mineral processing has had an important environmental impact in the wider area between the Agios Philippos mine at Kirki and the
123
64
C. Loupasakis, G. Konstantopoulou
Alexandroupolis coastline. The abandoned open pits and mine galleries, the eight tailings ponds located next to the flotation plant, the remains of the mining concentrates and the piles of unprocessed ore, as well as some abandoned stocks of chemicals necessary for the flotation process (e.g. sodium cyanide) are the main sources of toxic metals such as Pb, Zn, Cu, As, Cd. The Greek Institute of Geology and Mineral Exploration is undertaking an ongoing project to identify and measure the impact of acid mine drainage generated by the copper– lead–zinc mines at Kirki and to propose a feasible remediation plan for the area affected (Michael and Dimadis 2006a, b; Romaidis 2007; Liakopoulos 2009; Liakopoulos et al. 2009). As the fine grained material deposited in the pits retains a very high proportion of heavy metals (Arikas et al. 2007a; Triantafyllidis et al. 2007), a possible failure or a continuing leaching and erosion of these materials would extend the contamination over a much larger area (Arikas 2007; Arikas et al. 2007b).
Geological setting The regional geological structure of the broad study area (Eastern Rhodope zone) comprises three geotectonic units (Michael 2004; Dimitrov 1959; Voriadis 1954): 1. 2. 3.
The pre-Mesozoic crystalline basement of highly metamorphic rocks (gneisses, amphibolites, marbles), The Mesozoic crystalline basement of metamorphic rocks (schists, marbles) and The Tertiary basins occupied by volcanic (andesites, dacites, rhyodacites, rhyolites, shoshonites, latites) and sedimentary rocks, intruded by granitic bodies and acid dykes.
The extensive Eocene–Oligocene magmatic activity resulted from the subduction of the African plate beneath the southern margin of the Eurasian plate and is attributed to a slab break-off mechanism. The old flotation plant and the surrounding tailings ponds are founded on Tertiary clastic sediments. As seen in Fig. 1, the southeastern part of the study area is occupied by a breccio-conglomerate of polygenetic origin (gneisses, leptinites, granitic gneisses, quartz etc.) up to 150 mm in size. Molasse-type sediments are found in the north and the west, consisting of alternating layers of breccio-conglomerate, sandstones, shales and marls with a thin lignite horizon. Flysch formations also occur in the northwestern part of the area, consisting mainly of sandstone and shale, with minor intercalations of micro-breccias. There is relatively little tectonic structure, with only a few vertical normal faults and very wide joint sets. As shown on the geological map (Fig. 1), the northern part of the study area
123
is covered by alluvial deposits, consisting of silty sands with gravels and recent loose deposits along the stream beds. Various mining wastes, including tailings pond materials and stock piles of unprocessed ore, have been placed on the underlying Tertiary clastic basin deposits and on the late Holocene colluvial deposits. Based on in situ hydraulic conductivity tests, the permeability of the clastic materials forming the founding level of the tailings ponds vary from 1.47 9 10-7 to 1.2 9 10-6 m/sec. According to piezometric data, the water level of the shallow unconfined aquifer extending along the study area varies between 1.2 and 3.6 m, hence despite the low hydraulic conductivity, uncontrolled infiltration of the surface water along the tailings ponds allows polluted fluids to enter the underground water system (Ghomshei and Allen 2000; Wiertz and Marinkovic 2005).
Geotechnical problems The Kirki mines flotation plant comprises eight tailings ponds located close to the factory. These pits were placed in morphologically favourable locations, requiring the minimum earthworks. They were excavated along valleys or in the flat areas surrounding the two main streams crossing the study area (Kirkalon and Eirini streams), without taking into consideration the effects of surface water flow. The embankments initially were constructed using the uncompacted soil material arising from the excavation of the pits. Subsequently their height was increased using the mud deposited in the tailings ponds. The maximum height of the embankments varies from 5 to 10 m. The area of the tailing ponds varies from 400 m2 (No. 8) to 8,000 m2 (No. 3) and the exact volume of their contents cannot be established without drilling or geophysical investigations. Taking into consideration the construction practices when the tailings ponds were created, the following geotechnical issues arise: a.
The tailings ponds were excavated into the geological formations without taking any measures to protect the aquifers. No impermeable membranes or low permeability clay layers were installed. b. The stability of the embankments is open to question as they were constructed with uncompacted soil material and flotation mud. c. In some cases the embankments were constructed on the top of high slopes (e.g. tailings ponds No. 3 and 4) such that the stability of the slope—embankment system is also questionable. d. The construction of the tailings ponds with no drainage system and in some cases along valleys (e.g. tailings
Safety of abandoned tailings ponds, Greece
65
Fig. 1 Detailed geological map of the old flotation plant area by I.G.M.E. (Michael and Dimadis 2007) modified by the authors
e.
tailings ponds (leakages) are clearly shown on the geological map in Fig. 1. Some tailings ponds are sufficiently close to the streams (e.g. No. 5, 7 and 8) for erosion to threaten their stability.
The study reported here was therefore undertaken to consider how to terminate the mud leakages and ensure the safety of the embankments.
Geotechnical investigation
Fig. 2 View of flotation tailings pond No. 3 flooded with rain water
ponds No. 3 and 6) means that water can collect in the abandoned pits (Fig. 2), causing the mud to overflow (Fig. 3) as well as extensive erosion problems (Figs. 4, 5). The pathways of the mud escaping from the various
In addition to geological mapping, the study comprised continuous core sampling of four boreholes, in situ permeability and SPTs and laboratory testing. Three of the boreholes were drilled on the top of the embankments separating tailings ponds Nos. 1 and 2 and Nos. 3 and 4. The fourth hole was drilled in the middle of tailings pond No. 3 to obtain the mechanical properties of the flotation mud.
123
66
C. Loupasakis, G. Konstantopoulou
The results of the laboratory and in situ tests are given in Table 1 which shows the properties of the three main types of material: A: Embankment material—gravels and sand with silt to silty sand with gravels. The top layer consists of flotation mud—silty clay. B: Flotation mud—based on its grain size distribution it can be characterised as clayey silt but the material does not contain a significant amount of clay minerals. C: Foundation formation—molasses-type sediments (sandstones shales, Marls and breccio-conglomerates).
Fig. 3 Mud overflowing from tailings pond No. 4 extending to pond No. 5
It is clear that the initially uncompacted materials forming the embankments through the years increased in strength due to the dynamic loading of the earthmoving machines, such that their mechanical parameters are now similar to the natural foundation material. The flotation mud contains the remains of the pulverised ore hence appears to be non plastic (N.P.) although its grain size distribution falls within the clay and silt categories.
Safety assessment
Fig. 4 Erosional rill/ditches draining tailings pond No. 5, through the eroded embankment towards the Kirkalon stream
Fig. 5 Erosion channel cut some 2 m high, cut through the embankment of tailings pond No. 5
123
In order to estimate the factor of safety of the embankments and the slopes against transitional sliding, a global stability was undertaken using the overall stability module of the Larix-5 geotechnical engineering program. Larix estimates the safety factor of slopes, dams, excavations etc. with respect to sliding by formulating the global equilibrium of forces acting on a slice according to the well known methods of Janbu (1954, 1957); (Janbu et al. 1956) and Krey (1936). Representative stability analyses were undertaken for all the embankment and slope scenarios occurring across the study area. The simplified theoretical profiles selected for the stability analyses were determined by evaluating the morphological and geological maps as well as the geotechnical data from the four boreholes. Unfortunately, data concerning the accurate morphology and depth of the tailings ponds as well as the morphology of the embankments foundation surface were not available. For this reason the cross sections were designed to take into consideration the worst possible combination of parameters, e.g. maximum possible pond depth, unfavourable dip direction of the foundation surface etc. As noted above, the embankments were mainly founded on molasse-type sediments consisting of breccio-conglomerates, sandstones, shales, and marls. As expected, the mechanical properties of these materials vary significantly, as do the foundation material from which they were obtained. For example, the friction angle values vary from 21 to 42.5° (Table 1). For the safety factor calculations, the
Safety of abandoned tailings ponds, Greece
67
Table 1 Physical and mechanical properties of the formations and the materials related with the tailings ponds
Physical parameters
Mechanical parameters
Formation
(A) Embankment material
(B) Flotation mud
(C) Foundation formation
Gravel content (%)
8–43
0
5–36
Sand content (%)
19–65
39–60
34–49
Silt content (%)
16–47
31–53
24–42
Clay content (%)
4–24
8–9
2–11
Liquidity limit, WL (%)
23.7–35.8
N.P.
24.9–25.3
Plasticity limit, WP (%)
16.1–22.6
N.P.
16.2–16.9
Moisture content, W (%) Bulk density, cb (kN/m3)
9.4–26 1.99–2.17
16.9–19.1 1.99–2.11
11–18.6 1.93–2.13
Dry density, cd (kN/m3)
1.60–1.96
1.70–1.77
1.39–1.98
Shear test, cohesion, c (kPa)
61–131
72–90
45–72
Shear test, friction angle, u (°)
20–42
28–31
21–42.5
Oedometer test, Compression index, Cc
0.064–0.138
0.032–0.080
0.052–0.178
Oedometer test, initial void ratio, eo
0.361–0.699
0.522–0.570
0.35–0.925
Hydraulic conductivity, k (cm/sec)
1.0E-5–3.37E-4
1.24E-5
1.47E-5–2.2E-4
mechanical properties were the minimum values estimated from the laboratory tests, ignoring the variations between the tailings ponds. The analyses were conducted using both the Janbu and Krey methods. According to Krey’s method, circular slip surfaces of radius R are assumed and the failure mechanism is a rigid body rotating about the centre of the slip circle. It assumes equilibrium conditions [RV = 0 (vertical forces) and RM = 0 (moments)] and that both horizontal and shear forces act on the sides of the slices and the equation to determine the safety factor (F) is solved iteratively. In contrast, Janbu’s method involves straight, non-circular, slip surfaces and translational failure with equilibrium conditions RV = 0 (vertical forces) and RH = 0 (horizontal forces). Again both horizontal and shear forces act on the sides of the slices. This method is especially suited to cases where the morphology of the soil layers indicates a less rounded failure surface (e.g. along a layer interface). Taking into consideration the seismicity of the wider region, seismic loading of the embankments was included in the analyses. The seismic loading is specified as a fraction of the acceleration due to gravity g, in the x and y directions. The resulting acceleration has an influence on the determination of the weight of the individual slices of the sliding body. In this case study the horizontal, ax 9 g, and vertical, ay 9 g, earthquake accelerations values were as 0.10g and 0.05g, respectively, following the Hellenic anti-seismic regulation (E.P.P.O. 2000). As an example of the applied procedure, Fig. 6 presents the results of two typical analyses; Fig. 6a showing the results for the slope carrying the embankment of tailings pond No. 3 obtained using the Krey method and Fig. 6b the results for the embankment of tailings pond No. 6 using the
Janbu method. The analyses indicated that all the embankments and slopes were stable against transitional movement, with minimum safety factors of between 1.33 and 4.46.
Fig. 6 a Analysis of the slope supporting the embankment to tailings pond No. 3 using the Krey method and b analysis of the embankment of tailings pond No. 6 using the Janbu method
123
68
C. Loupasakis, G. Konstantopoulou
Fig. 7 a Cross section of a tailings pond showing the proposed installation of the geomembranes. b A typical installation of the geomembranes. c Geo-cells filled with earth materials
Proposed measures
Conclusions
As the study had shown there was little risk of sliding of the slopes and embankments, consideration was given to the risk of erosion and the overflowing of the tailings ponds as well as a location where washed-out flotation mud could safely be placed. The following measures were proposed (Konstantopoulou and Loupasakis 2007).
The rehabilitation of metallurgical discards from abandoned mines and their associated infrastructure can be an extremely complicated problem, due to the variety of environmental, geotechnical and hydrogeological issues involved. The study reports field and laboratory studies of the abandoned mine area at Agios Philippos and concludes that the common practice of re-depositing the contents of tailings ponds in specially designed sites could be avoided by taking advantage of the relatively low permeability of the foundation geology and the technical properties of commercial geo-materials. The latter would also allow a more aesthetically acceptable environment to be created.
a.
The construction of a drainage channel network upslope of the tailings ponds in order to cut off the surface water inflow which would be diverted to the streams. b. Covering the tailings ponds and embankments with geo-membranes, to minimise infiltration of rainwater. The tailings ponds surfaces would be gently inclined to prevent the localised accumulation and stagnation of water (Fig. 7a, b). The gradient could be created using some of the washed-out flotation mud found in quantity across the study area. The visual impact of the area could also be improved by covering the geomembranes with geo-cells filled with earth in which grass could be planted (Fig. 7c). c. The construction of gabion walls along sections of the streams adjoining embankments/slopes to prevent further erosion and the reconstruction of damaged embankments. The sections of the Kirkalon and Eirini streams adjoining tailings ponds Nos. 7 and 8 would be appropriate sites for the installation of these walls. d. The remaining flotation mud found on the site should be deposited in the empty pond No. 1, after a geo-membrane has been placed on the base of the pond.
123
Acknowledgments This study was conducted in the frame of the Third Community Support Framework of IGME (project 65231) and was financed from the E.E.C and from the Hellenic State.
References Arikas K (2007) Mineralogical and geochemical study of the Kirki mining wastes, Evros region, Greece—determination of the environmental impact by toxic and heavy metals. In: Proceeding of 11th international conference of the Geological Society of Greece, XXXX:1343–1353 Arikas K, Asfahani N, Nowak A, Goetz D (2007a) The Kirki mine in Evros Prefecture, NE Greece, and considerations on the environmental impact—Part I: Geochemical and mineralogical study of the tailing ponds and mineral concentrates. Min Metall Ann 17:21–50 Arikas K, Asfahani N, Nowak A, Goetz D, Melfos V (2007b) The Kirki mine in Evros Prefecture, NE Greece, and considerations on the environmental impact—Part II: Determination of toxic elements in soils and river sediments. Min Metall Ann 17:51–69 ¨ bersicht der Metamorphen Komplexe in Dimitrov S (1959) Kurze U Bulgarien. Freiberger Forchungsh, C 57:62–72
Safety of abandoned tailings ponds, Greece E.P.P.O.—Earthquake Planning and Protection Organization of Greece (2000) Hellenic anti-seismic regulation-2000, E.P.P.O., Athens Finagenov OM, Shul’man SG, Luchina VK (1999) Evaluation of the reliability of a tailings dams during an earthquake. Power Technol Eng (formerly Hydrotech Constr) 33:3 Ghomshei MM, Allen DM (2000) Hydrochemical and stable isotope assessment of tailings pond leakage, Nickel Plate Mine, British Columbia. Environ Geol 39:8 Guzenkov SN, Shul’gina NK (2007) Determination of losses sustained from hydrodynamic failure at a tailing pond located in the northern climatic zone. Power Technol Eng (formerly Hydrotech Constr) 41:5 Janbu N (1954) Stability analysis of slopes with dimensionless parameters. Harvard soil mechanics series, No. 46 Janbu N (1957) Earth pressure and bearing capacity calculations by generalized procedure of slices. In: Proceeding of 4th international conference on SMFE, vol 2. 207–212 Janbu N, Bjerrum L, and Kjaernsli B (1956) Veiledning ved Losning av Fundamenterings—oppgraver. N.G.I. publication No. 16 Konstantopoulou G, Loupasakis C (2007) (Unpublished IGME internal Report) Geotechnical stability study of the tailings ponds of the Kirki mine, Evros Prefecture. IGME—Institute of Geology and Mineral Exploration, Athens Krey H (1936) Erddruck. Erdwiderstand und tragfahigkeis des baugraundes. 5th end. Wild Ernst und Sahn, Berlin Lee CH, Lee HK (2004) Environmental impact and geochemistry of old tailing pile from the Sanggok mine creek, Republic of Korea. Environ Geol 46:6–7 Liakopoulos A (2009) (Unpublished IGME internal Report) Environmental study of the former Kirki mines area. Pollution extension et proposed abatement measures. IGME—Institute of Geology and Mineral Exploration, Athens Liakopoulos A, Lemiere B, Drougas I, Crouzet C, Laperche V, Michaels C, Lassin A, Romaidis I (2009) Environmental impacts
69 of unmanaged solid waste at a former base metal mining and ore processing site (Kirki, Greece). In: Proceedings of second international conference on environmental management, engineering, planning and economics (CEMEPE 09) & SECOTOX (in press) Liu H, Li N, Liao X, Wu C, Pan X (2007) Effective stress analysis method of seismic response for high tailings dam. J Central South Univ Technol 14:1 Michael C (2004) Epithermal systems and gold mineralization in Western Thrace (North Greece). In: Proceeding of 10th international conference of the geological society of Greece vol XXXVI:416–423 Michael K, Dimadis E (2006a) (Unpublished IGME internal report) Geological study of the area around the Kirki beneficiation plant. IGME—Institute of Geology and Mineral Exploration, Athens Michael K, Dimadis E (2006b) (Unpublished IGME internal report) Results of geological mapping in scale 1:25.000 in the catchment area of ‘‘Eirini’’ stream. IGME—Institute of Geology and Mineral Exploration, Athens Michael C, Dimadis E (2007) (Unpublished IGME Map) Geological map of Kirki Processing plant area, scale 1:1000. IGME— Institute of Geology and Mineral Exploration, Athens Romaidis I (2007) (Unpublished IGME internal report) Preliminary hydrogeological study of the Kirki mines area. IGME—Institute of Geology and Mineral Exploration, Athens Triantafyllidis S, Skarpelis N, Komnitsas K (2007) Environmental characterization and geochemistry of Kirki, Thrace, NE Greece, abandoned flotation tailings dumps. Environ Forensic 8(4):351– 359 Voriadis G (1954) Geological and mining exploration in Thasos Island. IGME—Institute of Geology and Mineral Exploration, Athens Wiertz JV, Marinkovic FA (2005) Dissolved pollutant transport in tailings ponds. Environ Geol 47:2
123