Mapping of contaminant dispersion from a polluted mining area by geochemical and
geophysical methods, Rävlidmyran, northern Sweden
Use of geochemical and geophysical studies to investigate contaminants
Astrid Theander
Natural Resources Engineering, master's 2019
Luleå University of Technology
Department of Civil, Environmental and Natural Resources Engineering
Abstract
After open pit mining, the pit can either be backfilled or be filled with groundwater and become a pit lake. These lakes tend to be acidic and contains high concentrations of metals, which increases the environmental risks in the area. One of these pit lakes is Rävlidmyran in the Skellefte ore district. This problem and this pit lake have created the purpose of this thesis, which is to compare the different water types connected to an open pit, i.e. groundwater, surface water and the actual pit lake water, and to connect the water with geophysical readings. The purpose is also to compare ratios between elements to be able to gain more geochemical information.
To gain information about the groundwater, several groundwater pipes have since a long time back been installed in the area. The groundwater has thereafter regularly been sampled and analysed. The surface water has also been sampled regularly in different spots. In the pit lake, a depth profile has been created by water sampling at different depths in the pit lake. The geophysical measurements used in this thesis are resistivity and induced potential measurements, and measurements were done with the slingram method.
The water sampling indicated that all three types of water contained elevated concentrations of metals. The ratio between (Cu+Zn+Pb)/Na indicated that the pit lake had the highest value. The measurement also indicated that the highest concentration of dissolved metals is found under the chemocline in the pit lake, compared with other water types. Also, the ratio Fe/S were the highest under the chemocline in the pit lake, which indicates e.g. dissolution of pyrite. When it comes to the ratio representing e.g. dissolution of gypsum (Ca/S) it was below 1 all the time, except for in two groundwater pipes. The (Ca+Mg)/Na–ratio had the highest values in the pit lake and that indicates dissolution of e.g. carbonates. This is not very surprising since the pit lake has been and are limed regularly.
The geophysical investigations indicated increased electrical conductivity in a waste rock heap northwest of the pit lake, where the sampling indicated elevated copper concentrations. The readings also show that the groundwater flow direction is against Lake Hornträsket north of the pit lake. They also indicated potential flow paths for the groundwater. These can contain elevated concentrations of ore elements, which can be confirmed by a sampling of the groundwater in that area. From the geophysical data, it is also possible to see that one of the profiles are located along a possible groundwater plume, based on a low resistivity area and the shape of the potential plume.
Sammanfattning
Efter brytning ur ett dagbrott, så kan dagbrotten ur efterbehandlingssyfte till exempel bli återfyllt med antingen gråberg eller med vatten. Ifall dagbrottet blir återfyllt med vatten och en dagbrottssjö skapas, så tenderar denna sjö att vara försurad och innehålla förhöjda halter av metaller. Detta ökar de lokala belastningarna på miljön i området. En av dessa dagbrottssjöar är Rävlidmyran som ligger i Skelleftefältet. Denna dagbrottssjö och dess bieffekter har skapat syftet med detta examensarbete, vilket är att jämföra olika vattentyper kopplade till Rävlidmyran (grundvatten, ytvatten och vatten i dagbrottssjön) samt att jämföra geokemin med geofysiska mätningar. Syftet är också att jämföra molkvoter mellan relevanta element för att i sin tur få mer geokemisk information.
För att få fram mer information om grundvatten så har flertalet grundvattenrör installerats i området sen flera år tillbaka. Därefter har grundvattnet provtagits och analyserats regelbundet. Ytvattnet i området har också provtagits. I dagbrottssjön har en profil tagits fram genom att ta vattenprover på olika djup i sjön. De geofysiska mätningarna som gjorts är resistivitet och inducerad potential mätningar, samt mätningar som utförts med slingram-metoden.
Vattenprovtagningarna indikerade att alla tre vattentyper innehåller förhöjda halter av metaller. Molkvoten (Cu+Zn+Pb)/Na visade att kvoten var högst under kemoklinen i dagbrottssjön i jämförelse med de andra två vattentyperna. De kemiska analyserna av vattenproverna av de olika vattentyperna pekade också på att vattnet under kemoklinen i dagbrottssjön innehöll en större mängd lösta joner. Kvoten Fe/S var också högst i detta vatten. När det kommer till molkvoten som bland annat representerar gipsutfällning och gipsupplösning – Ca/S – så var det under 1 i hela området, med undantag för två grundvattenrör. Kvoten (Ca+Mg)/Na var högst i dagbrottssjön och det indikerar upplösning av exempelvis karbonater. Detta är inte helt oväntat, då dagbrottssjön kalkats regelbundet sedan en lång tid tillbaka.
De geofysiska utredningarna indikerade på förhöjd elektrisk konduktivitet i ett gråbergsupplag nordväst om dagbrottssjön, där vattenanalyserna påvisade en förhöjd koncentration av koppar. Mätningarna visade även att grundvattnets riktning var mot sjön Hornträsket, som är lokaliserad norr om dagbrottssjön. De indikerade även potentiella flödesvägar för grundvatten. Dessa flödesvägar kan innehålla förhöjda halter av metaller, vilket kan bekräftas med hjälp av provtagningar av grundvatten i området.
Från de geofysiska mätningarna är det också möjligt att se att en av profilerna är placerad längs med en möjlig grundvattenplym, baserat på en låg resistivitet och den potentiella plymens form.
0
Contents
1. Introduction ... 1
1.1. Background ... 1
1.2. Purpose ... 1
2. General theory ... 2
2.1. Mine waste ... 2
2.1.1. Waste rock ... 2
2.1.2. Tailings ... 2
2.2 Groundwater ... 2
2.3 Acidic mine waters ... 3
2.4 Pit lakes ... 5
2.5 Groundwater pollution ... 7
2.5.1 Pollution spreading ... 7
2.6 Geophysical methods and polluted groundwaters ... 7
3. Site description ... 10
3.1. General description of the area ... 10
3.1.1. Rävlidmyran Pit Lake... 11
3.1.2. Hydrology ... 13
3.2. Previous remediation actions ... 14
4. Methodology ... 15
4.1. Geochemical methods ... 15
4.1.1. Element concentrations ... 15
4.1.2. Element ratios ... 15
4.2. Geophysical methods ... 16
5. Results and discussion ... 18
5.1. Geochemical results ... 18
5.1.1. Groundwater ... 18
5.1.1.1. pH ... 18
5.1.1.2. Electrical conductivity ... 19
5.1.1.3. Sulphur ... 21
5.1.1.4. Calcium ... 22
5.1.1.5. Iron ... 23
5.1.1.6. Arsenic ... 24
1
5.1.1.7. Zinc ... 25
5.1.1.8. Lead ... 26
5.1.1.9. Copper ... 26
5.1.2. Rävlidmyran Pit Lake... 28
5.1.3. Streams and surface water ... 29
5.1.3.1. pH ... 29
5.1.3.2. Electrical conductivity ... 29
5.1.3.3. Sulphur ... 30
5.1.3.4. Calcium ... 30
5.1.3.5. Iron ... 31
5.1.3.6. Arsenic ... 31
5.1.3.7. Zinc ... 32
5.1.3.8. Lead ... 32
5.1.3.9. Copper ... 33
5.1.4. Comparison of groundwater, Rävlidmyran Pit Lake and surface water... 34
5.2. Element ratios ... 35
5.2.1. Groundwater ... 35
5.2.2. Rävlidmyran Pit Lake... 36
5.2.3. Streams and surface water ... 37
5.3. Geophysical results ... 38
5.3.1. Resistivity and Induced Potential models ... 38
5.3.2. Slingram models ... 43
5.4. Correlation between geophysical and geochemical data ... 46
6. Conclusions ... 49
7. Acknowledgements ... 50
8. References ... 51 Appendix I – Coordinates
Appendix II – Full sampling results from the groundwater samples Appendix III – Full sampling results from the surface water samples Appendix IV – Full sampling results from the pit lake samples
Appendix V – Element ratios
1
1. Introduction
1.1. Background
Since a long time, back mining operations started in Sweden, i.e. extracting valuable metals from ore deposits in the ground. These mining activities could be called necessary for today’s society, but it has its drawbacks. It causes contamination in nature, both through air and water and it causes large volumes of waste rock. There are different ways to extract the ores from the ground, like underground mining or open pit mining. Open pit mining often leaves a pit lake after closure. The groundwater fills up the pit with water and a pit lake arises. Due to the conditions of the old pit, this lake has a low pH and a high content of dissolved metals, which spreads out into nature. The acidic and metal-rich water spreads out through groundwater and surface water and contaminates the surrounding nature. To decrease the environmental effects an open pit mine has after closure remediation actions are needed and it is important to investigate and evaluate the situation for the mining area.
This thesis is about a closed open pit mine called Rävlidmyran and the thesis are written on behalf of Golder Associates AB, which in turns does it for Boliden AB.
1.2. Purpose The purpose of this study is to
∴ Compare the tree water chemistry types - Pit lake
- Surface water - Groundwater
∴ Compare the ratios - Ca/S - Fe/S
- (Ca+Mg)/Na - (Cu+Zn+Pb)/Na between the three water types.
∴ Evaluate the possibility to connect the geochemical measurements and calculations, with the geophysical measurements done in Rävlidmyran
The aim is to answer the following questions
∴ Are there any differences in the ratios Ca/S, Fe/S, (Ca+Mg)/Na or (Cu+Zn+Pb)/Na?
∴ Is there someplace that is more affected by contaminations from the previous mining activities in the area?
∴ Are the geochemical measurements and calculations relatable to the geophysical measurements?
2
2. General theory
2.1. Mine waste
Mines are used to excavating valuable metals in different ways, but all excavation methods produce mine waste as well. The two major waste types are waste rock (gangue) and tailings, and in the case of sulphide ores, these wastes are associated with acidic mine waters.
2.1.1. Waste rock
The waste rock contains surrounding bedrock that has been excavated to reach the actual ore and low ore-graded material. The waste rock is crushed and blasted into mixed grain sizes, from finely graded material to large boulders. The amount depends on the depth, the ore type and the type of mining. The waste rock is deposited in waste rock dumps above ground. If the waste rock is inert – and not likely to represent a significant pollutant threat to the environment – it can be used in the construction of dams and roads. The waste can also be backfilled after mine closure if that is possible. However, the waste can contain high concentrations of sulphides that may undergo oxidation, which has a negative impact on the environment (See section 2.3.).
2.1.2. Tailings
Tailings are the crushed and milled residual material remaining after processing and extraction of the valuable minerals and metals. It is a rather homogeneous and fine- grained material, such as silt or sand, and it can contain a certain percentage of sulphide, often pyrite. Tailings might contain residual process chemicals, which resist from the ore processing. They are typically pumped together with process water to topographic depressions; tailing impoundments. Within the tailings, segregation might occur during deposition due to different particle sizes and densities.
2.2 Groundwater
Water is essential for life on earth. Water is needed for both vegetation and animal life, including human life. There are multiple sources for freshwater, and the main source lies beneath the earth's surface, known as groundwater. Groundwater is the water that is occupying the voids within the geological stratum, such as bedrocks, soils, sands, clays, and cracks. The size and the shape of polluted groundwater plume are controlled by the geology, the groundwater flow, and human modification of the environment. (Todd &
Mays, 2005).
The main source for groundwater is rainwater, which is originally water vapour from the oceans. This connects groundwater to the hydrological cycle; the cycle describing the continuous movement of water on, above and under the surface of the earth (Figure 1).
3
Figure 1 - Hydrologic cycle with global annual average water balance given in units relative to a value of 100 for the rate of precipitation on land (Fetter, 2001), (Todd & Mays, 2005).
In Figure 1 above, the hydrology cycle is presented. The water in the hydrological system is present in three separate systems, named as the subsurface water system, the surface water system and atmospheric water system. The cycle connecting these three systems starts with water vaporizing from oceans and lakes. This water vapour condenses and precipitates, and thereby leaves the atmospheric water system. The precipitated water can either stay on the surface and discharge as surface runoff or infiltrate to the subsurface.
The water on the surface may be intercepted by plants or become an overland flow, i.e.
become a part of the surface water system. This continues its process as evapotranspiration, in other terms it either evaporates from the leaves of plants or the soil, or it transpires through plant leaves. The water infiltrating in the soil and the subsurface water system can percolate deeper into the ground and recharge the groundwater, which can become seepage into streams, streamflow or spring flow. These different systems are interconnected and together create the hydrological cycle (Todd &
Mays, 2005).
The groundwater quality depends not only on the surrounding bedrock but also on human activities – such as industrial operations, poor treatment of wastes of a different kind, and the over-development of groundwater resources. These human activities have increased during the last century, and in addition, we have created problems with pollution. Hence, groundwater may contain hazardous substances and elements, both from natural sources and pollution caused by human activity. These hazardous substances and elements can cause a threat to health if the water is consumed. It also deteriorates the environment.
2.3 Acidic mine waters
During ore or coal mining, sulphur bearing minerals are exposed. The mine itself exposes these minerals. Waste material such as tailings can contain sulphide minerals and
4
therefore also expose sulphur bearing minerals to the environment. When these minerals encounter water and air, acidic leachate is formed. Acid Mine Drainage (AMD) is a global environmental problem. The exposure leads to oxidation of sulphur [S], which often generates sulphuric acid [H2SO4]. In turn, this leads to increased concentrations of dissolved metals and sulphur in e.g. the groundwater or the surrounding soil. These elevated concentrations of metals and sulphur and the acidic conditions cause a threat to the environment and animal health as well as human health. AMD can also occur naturally in unexposed mineralized areas, but not usually in the same extent as the human-made AMD.
The oxidation of sulphur-bearing minerals can be basic, neutral or acidifying, depending on the properties within the minerals. For example; the oxidation of chalcocite [Cu2S] is acid-consuming, according to the following reaction;
2𝐶𝐶𝐶𝐶2𝑆𝑆(𝑠𝑠) + 5𝑂𝑂2(𝑔𝑔) + 4𝐻𝐻+(𝑎𝑎𝑎𝑎) → 4𝐶𝐶𝐶𝐶2+(𝑎𝑎𝑎𝑎) + 𝑆𝑆𝑂𝑂42−(𝑎𝑎𝑎𝑎) + 𝐻𝐻2𝑂𝑂(𝑙𝑙) (2.1) The oxidation of sphalerite [ZnS] is neither basic nor acidic since the reaction involves no hydrogen ions (Castro & Moore, 2000):
𝑍𝑍𝑍𝑍𝑆𝑆(𝑠𝑠) + 2𝑂𝑂2(𝑔𝑔) → 𝑍𝑍𝑍𝑍2+(𝑎𝑎𝑎𝑎) + 𝑆𝑆𝑂𝑂42−(𝑎𝑎𝑎𝑎) (2.2) One mineral whose oxidation is acidifying is pyrite [FeS2];
2𝐹𝐹𝐹𝐹𝑆𝑆2(𝑠𝑠) + 7𝑂𝑂2(𝑔𝑔) + 2𝐻𝐻2𝑂𝑂(𝑙𝑙) → 2𝐹𝐹𝐹𝐹2+(𝑎𝑎𝑎𝑎) + 4𝐻𝐻+(𝑎𝑎𝑎𝑎) + 4𝑆𝑆𝑂𝑂42−(𝑎𝑎𝑎𝑎) (2.3) Pyrite is known to be one of the most acid-producing minerals. As shown in reaction (2.3), the oxidation of pyrite produces ferrous iron, sulphur and hydrogen ions. This reaction is almost irreversible, so the reaction is drawn towards the right. If the conditions are oxidizing, this ferrous iron might oxidize further, i.e. to ferric iron. Ferric iron will in turn precipitate to ferric hydroxide (2.4), unless the pH is very low (2.5) which allows the ferric iron to stay in solution (Appelo & Postima, 2005).
4𝐹𝐹𝐹𝐹2+(𝑎𝑎𝑎𝑎) + 𝑂𝑂2(𝑔𝑔) + 4𝐻𝐻+(𝑎𝑎𝑎𝑎) ↔ 4𝐹𝐹𝐹𝐹3+(𝑎𝑎𝑎𝑎) + 2𝐻𝐻2𝑂𝑂(𝑙𝑙) (2.4) 𝐹𝐹𝐹𝐹3+(𝑎𝑎𝑎𝑎) + 3𝐻𝐻2𝑂𝑂(𝑙𝑙) ↔ 𝐹𝐹𝐹𝐹(𝑂𝑂𝐻𝐻)3(𝑠𝑠) + 3𝐻𝐻+(𝑎𝑎𝑎𝑎) (2.5) Pyrite can also be oxidized by ferric iron, which is a fast reaction occurring at a low pH.
This occurs according to the following reaction;
𝐹𝐹𝐹𝐹𝑆𝑆2(𝑠𝑠) + 14𝐹𝐹𝐹𝐹3+(𝑎𝑎𝑎𝑎) + 8𝐻𝐻2𝑂𝑂(𝑙𝑙) ↔ 15𝐹𝐹𝐹𝐹2+(𝑎𝑎𝑎𝑎) + 2𝑆𝑆𝑂𝑂42−(𝑎𝑎𝑎𝑎) + 16𝐻𝐻+(𝑎𝑎𝑎𝑎) (2.6) Another factor that can affect pyrite oxidation is pyrite oxidizing bacteria, whose activity can accelerate the oxidation process in natural settings. As a conclusion, we can state that pyrite oxidation is overall an acidifying reaction with many different steps. This means that pyrite is one of the minerals that often contribute to the formation of AMD (Appelo &
Postima, 2005).
Nevertheless, metal-rich water formed from chemical reactions between water and rocks containing sulphur-bearing minerals often are produced at mine sites. This leads to an acidic runoff and leaching of metals to the surroundings. Both the lowered pH and the increased content of metals in the water contribute to the environmental threat. The dissolved metals and other constituents in AMD can be toxic to aquatic organisms and can
5
in the end precipitate and form encrustations that degrade the aquatic habitat. The dissolved metals can also be toxic for vegetation when the acidic and metal-rich water is taken up by vegetation in conjunction with photosynthesis.
2.4 Pit lakes
Pit lakes are artificial lakes that are formed when open pit mining activities cease and the pit – which tends to have a high ratio between depth and surface area – is naturally filled with ground-, surface- and rainwater. The bottom of the pit is usually rather flat with steep sides, due to the open pit mining technique. The size of pit lakes has increased during the last century due to the increased need for resources. Extraction technologies have also become more sophisticated and able to operate at larger scales. During the mining procedure in the open pit, the water level needs to be controlled to prevent a lake from forming during the excavation of the ore. This can be done by pumping, either by perimeter pumping or in-pit pumping. This makes it possible to mine material below the original groundwater water table. A steady water level in a pit lake will be established when the groundwater water surface recovers to the level of the surrounding groundwater table (Figure 2).
Figure 2 - Principle sketch over a pit lake and its different water processes (HydroGeoLogica).
The water quality in a pit lake varies depending on the surrounding bedrock and the catchment interaction, such as connection to groundwater, mine discharge or vegetation.
It also depends on the alkalinity, quantity, and the quality of the incoming water (Lu, 2004). Pit lakes are formed in ore-bearing landscapes that tend to contain a lot of sulphides. When these ores are extracted, these surrounding rocks are exposed to air and water, which accelerates oxidation of the surrounding rocks. The oxidation, in turn, produces acidity which increases the leaching of metals. The acidity creates acid mine drainage (AMD, See section 2.3) and increases the concentration of dissolved metals to levels that can be harmful to both animals and plants (Blanchette & Lund, 2016). The pit
6
lake may contain large volumes of water and the pit lake might contaminate nearby areas by spreading contaminated water from the lake.
Pit lakes are complicated systems, which interact with the surrounding environment.
According to previous studies, it is important to predict the rate of filling of a pit lake. This is to support the geochemical modelling of the water quality in the pit lake when the pit lake has reached a steady state (Lu, 2004). To predict the pit lake water quality, it is important to know the directions of the groundwater flow and the water table elevations, since they affect the characteristics of the pit lake. The original water level in a groundwater aquifer can be used as a prediction/approximation of the water level in the pit lake since it depends on the groundwater levels. A pit lake might transect with multiple groundwater aquifers, which means that the level and the rate and direction of the flow affects the water level in the pit lake (Miller, Lyons, & Davis, 1996). The groundwater quality is linked to the water quality in the pit lake since they approach geochemical equilibrium with each other. The surrounding groundwater quality and its flow and direction are also affected by the conditions in the pit lake.
The difference between a regular lake and a pit lake, except for the fact that the pit lake is artificial, is the shape, occurrence, recharge system, chemical composition, and the depth- area ratio. A pit lake has a higher ratio between the maximum depth and the mean diameter of the lake (10-40 %), while a natural lake has a ratio less than 2 % (Castro &
Moore, 2000). Natural lakes are more dependent on surface runoff for recharge, while a pit lake is recharged by mainly groundwater (Bowell, 2002). A characteristic feature of pit lakes is that they tend to lack shorelines and areas with shallow waters. From this follows that the biological activity in pit lakes is generally low, and pit lakes generally have very thin or non-existing sediment layers where the rooted biological communities can develop (Castro and Moore, 2000; Bowell, 2002; Lu, 2004). If the pit lake has been limed, there is a possibility that a layer of Fe or lime has settled as sediment on the bottom of the lake. The chemical composition influences the possibility of vegetation in pit lakes since the chemical composition differs from natural lakes. Pit lakes tend to be acidic, with high levels of sulphate and dissolved metals. The previous mining in the Rävlidmyran pit lake exposed minerals to oxidizing conditions. These minerals released sulphuric acid that dissolved metals during the stage of rising water in the pit lake. In other words, pit lakes have most likely poor water qualities without treatment or substantial remediation work.
The poor water quality will, in turn, affect the environment negatively. Pit lakes tend to need remediation, but in some cases, nature itself has neutralized the pit lake. If the bedrock has a high content of calcite [CaCO3] or dolomite [CaMg(CO3)2], the dissolution of these two minerals in the acidic water will have neutralizing effects on the lake in general.
Acidic lakes can also be naturally neutralized by sulphate reducing bacteria (Castro &
Moore, 2000).
7 2.5 Groundwater pollution
2.5.1 Pollution spreading
As mentioned in section 2.3, groundwater surrounding sulphide mines is often heavily polluted by metals and sulphate. The extent of groundwater pollution from mines and other sources as well decreases with distance from the source until the concentration of the pollutant is no longer in harmful levels. The spreading and the attenuation rate of pollution depends on different spreading ways and the characteristics of the components (Appelo & Postima, 2005).
Figure 3 - Principle sketch of the decreasing concentration in a contamination plume.
Figure 3 shows a principal sketch of how the concentration of contaminants decreases from a harmful level to a harmless one with increasing distance from a general contamination source. For a pit lake, the spreading of pollutions is more concentrated to sporadic emissions points or sections, and not from the entire pit lake. Pollutions can spread from the source in many different ways, both in the soil, the groundwater, the surface water and in the air (Todd & Mays, 2005). The groundwater flow affects the distribution and shape of the pollution plume. As a result, pollutants are affected by a number of processes during their transport from the source, for example:
∴ Diffusion
∴ Advection
∴ Dispersion
∴ Sorption
∴ Adsorption
∴ Absorption
∴ Ion exchange
∴ Degradation
∴ Co-precipitation
∴ Dissolution
∴ Retardation
∴ Gas-water exchange
∴ Chemical mixing and mass balance
2.6 Geophysical methods and polluted groundwaters
One way to help locate pollutants in the ground is to use geophysical methods. These methods are non-invasive, and they measure physical properties of geological materials, e.g. electrical conductivity and magnetic properties. As mentioned earlier, one of the major environmental problems caused by mining activities is AMD, which is defined as acidic, metal and sulphate-rich water formed from chemical reactions between water, air, and rocks containing sulphur-bearing minerals. Metals are good electrical conductors. In a soil/rock, the electrical conductivity is dependent on both the fluid and the minerals in
8
the soil/rock. If the soil/rock is moist, the electrical conductivity is more dependent on the properties of the fluids, while a dry soil/rock is more influenced by the properties in the minerals (Chitea, Ioane, & Serban, n.d.). Since metals generally are good conductors, the electrical conductivity in the soil/rock can indicate elevated levels of metals, which can be associated with pollutants from a mine or an ore body.
One geophysical method is to measure the resistivity in the ground, also known as the direct current resistivity (DC). Resistivity itself describes the inability of a material to conduct electricity. Materials with a low resistivity, have a high capacity to conduct electricity. In many soils and rocks, the minerals act like isolators. Hence, the amount of water, dissolved ions and their distribution within the material control the resistivity. The lower the resistivity is, the higher is the possibility to find metal-rich orebodies, water or water with higher concentrations of dissolved ions. Resistivity measurements can, therefore, give information about possible pathways for contaminated leakage from e.g.
old mines or landfills (Dahlin, 1996). Briefly, the measurements are made by injecting a current [I] between two different electrodes. The receiver voltage [U] measurements are taken at different distances along a line from the current source. The resistance [R] can then be calculated by Ohm’s law;
𝑈𝑈 = 𝑅𝑅 ∗ 𝐼𝐼 . (2.7)
This gives information about variations in apparent resistivity, i.e. variations in the electrical resistance of geological layers. After measurements are made, the data is processed and models of the resistance are generated. This offers a possibility to determine and identify pollutants, soil storage, groundwater and bedrock levels, mineralized bedrock and filling masses in the bedrock (Perttu, 2018). In groundwater, a decrease in electrical resistivity can be a direct result of an increase of water and in the ionic content of this water. This is typical for inorganic contaminations in a groundwater aquifer, such as contaminations caused by mining activities (Vanhala, 2001).
Another type of geophysical method is induced polarization (IP), which is an imaging technique used to identify the electrical chargeability of subsurface material. It is used to provide additional or more detailed information about lithology than the resistivity method does. This method is therefore suitable for characterizing disseminated mineralization. However, it is sensitive to electrical noise from train traffic and thunderstorms. Induced polarization is based on the ability of a certain main material to take an electrical charge, like minerals in a rock or metal waste. An excitation current pulse is sent down into the subsurface and the voltage is measured. The voltage will not instantly return to zero when the current disappears. Instead, it will show a slow material- dependent decay, which is responsive to low-grade mineralization (Perttu, 2018).
A benefit of using IP measurements and resistivity measurements together is partly because the same instruments are used for both methods, so there is no need for carrying two different kinds of equipment. Another benefit is that the combination of the two types of measurements can provide information helpful to distinguish contaminated groundwater from clay (Sharma, 1997). Groundwater and clay both have a low resistivity, while clay has a higher IP than sand with contaminated groundwater (Parasnis, 1997).
9
A third method used for geophysical investigations is slingram methods. The principle of this method is to generate a primary geomagnetic field through the ground and achieve a secondary field. Different transmitters are placed with a known distance from each other, and one of them sends out currents through the underlying ground. The other transmitters register the strength of the upcoming secondary geomagnetic field caused by the primary field. The anomalies between the primary and the secondary field result in a 2D-mapping of the electromagnetic properties, i.e. electric conductivity and magnetic susceptibility of the ground. This method can help locate conductive bedrock extensions, deformation zones, and potential flow paths for groundwater, the dip of deformation zones, as well as surficial geology- or contaminant mapping.
10
3. Site description
3.1. General description of the area
In the Skellefte ore district in the north of Sweden, an old mine is located; Rävlidmyran.
Rävlidmyran is located approximately 4,5 km northwest of the mine in Kristineberg and 120 km west of Skellefteå (Figure 4). The mine belongs to the Rävlidenfältet area ore bodies, where several volcanic massive sulphide (VMS) deposits have been found. The mining activities in Rävlidmyran started in 1951, with both open pit and underground mining. The mean annual ore production was 230 000 tonnes (Boliden Mineral AB, 1975).
The recipient of both the groundwater and the surface runoff from the mining area is the Hornträsket Lake, which is located north of the Rävlidmyran mine. Today, the site consists of a big, partly filled, pit lake, a couple of smaller backfilled pits, an excavation spot a few hundred meters to the west, an empty industrial site southwest of the pit lake, a large waste rock dump, and an industrial area partly damming the northern side of the pit lake (Figure 4) . The waste rock dump and the northern industrial site are covered by 0.5 m and 0.3 m of glacial till, respectively, and are replanted. The sulphide content in the waste rock is estimated to be approximately 2-5% (Öhlander, Lu, & Alakangas, 2007).
Figure 4 - Location map of Rävlidmyran mine. A: Waste rock dump site and industrial area, covered by till and grass.
B: Rävlidmyran Pit Lake, where the former Sture-ore was located. C: Industrial site with some smaller backfilled open pits. (Modified from Boliden Mineral AB)
The mining activity at Rävlidmyran ceased in 1974 in the open pit and it was filled with approximately 40 m of waste rock. Underground mining ended in 1991. At the same time, the dewatering of the underground mine ended. Thereafter, the underground mine was filled with water sometime between 1994-1995, followed by the water filling of the open pit (Golder Associates AB, 2015).
11
The bedrock in the area consists of black or grey shale, sericite quartzite, chlorite and talcum rich shales, and impure carbonates (dolomite and limestone) (Grip & Frietsch, 1973). The bedrock represents a metamorphic transition sequence from the Skellefte volcanic series to the overlying phyllite series. Rävlidmyran contains several separate VMS ore lenses, such as Sture ore, A-, B-, D-, G- and H-ores. The present pit lake formed when the Sture ore body was mined. This ore occurred in carbonate rocks (limestone) and was a pyrite-rich ore with zinc, copper and lead and smaller amounts antimony, silver, and gold (Roering, 1959). The ore lenses lie in an east-west direction and the lenses are steeply dipping towards the south. The area above the bedrock generally consists of sandy till on top of a clayey till.
3.1.1. Rävlidmyran Pit Lake
Rävlidmyran Pit Lake is the result of the previous open pit mining of the Sture ore. After the mining operations ended, the open pit was naturally filled with water, which in turn formed the pit lake. From the closing of the mine until early 2003, the pit lake was left without any treatment and had a high metal and sulphate content and an average pH of 3.6. After that, the lake was limed with approximately 200 tonnes of lime during a period of 3 weeks. Later the same year, the lake was treated with 300 tonnes of sewage sludge (Lu & Öhlander, 2005). The pit lake is in hydraulic contact with the underground mine, which today is water filled. It is also likely to be in hydraulic contact with Rävliden, a closed mine 1.5 km south of Rävlidmyran.
According to Lu (2004), the lake was at that time meromictic and oligotrophic. The uppermost 5 metres, the mixolimnion, were well mixed and oxygenated. Underneath, there was a chemocline (5-8 metres) distinctively separating the mixolimnion from the monimolimnion. The monimolimnion was characterized as an anaerobic layer with higher density, higher elemental concentrations and a higher conductivity. The density difference between the different layers in the pit lake prevents the water in the lake from cycling (Lu, 2004).
12
Figure 5 - Aerial photos over Rävlidmyran during different years. (Modified from BergAB, 2013).
In the figure above (Figure 5), four aerial photos are shown from different years to provide an overview of the development of the open pit. Figure 6 shows parts of the lake in 2018.
In 2012, during approximately 2 months, the lake water level was lowered by 2 metres and the water was pumped into Lake Hornträsket. Two drawbacks with dewatering a pit lake, are that (1) the side walls of the pit lake are exposed to oxygen again and (2) the surrounding groundwater level might change. The water level was lowered by approximately 5 metres (Golder Associates AB, 2015).
13
Figure 6 - Picture of Rävlidmyran Pit Lake after the lowering of the water level done in 2018. Lake Hornträsket is in the background (2018-10-16).
Before the dewatering of the pit lake, the lakes water body had a volume of approximately 527 000 m3 and a surface area of 49,191 m2. The average water depth was 10.7 metres and the maximal depth was 25.9 metres. During the operation of the mine, around 500,000 m 3 yr -1 was pumped from the pit (Golder Associates AB, 2015).
3.1.2. Hydrology
The groundwater table in the area is located around 2-7 metres below the ground surface and the average flow direction is from south to north (Golder Associates AB, 2018; Perttu, 2018).
14
Figure 7 - Conceptual model of the direction of the groundwater flow in the area (Modified from Golder Associates AB, 2018).
The direction of the groundwater flow is illustrated in Figure 7 above. The pit lake itself does not have a specific outflow. Instead, there is a diffuse runoff through the surrounding glacial till towards Lake Hornträsket. The size of the drainage area is approximately 200 000 m2. The sources for water flowing into the lake are precipitation, surface runoff, snowmelt and groundwater (Golders 2015; BergAB 2013).
3.2. Previous remediation actions
Since the closing of the mine, remediation actions have been done with the purpose to minimize the environmental impact. These actions have not been suitable as long-term solutions, and another remediation method is needed. During 2003, the pit lake was treated with 200 tonnes of lime followed by 300 tonnes of sewage sludge to biodegrade sulphate and precipitate metals. This was not successful. Afterwards, the pit lake has been treated with lime with 30 tonnes of lime in 2006, 65 tonnes of lime during 2007, 25 tonnes of lime in 2008 and 15 tonnes of lime in 2010. Thereafter, the pit lake has regularly been limed and it is an ongoing process. Even though liming is an effective and rapid remediation method, it is not suitable in the long run due to economic aspects (Lu, 2004;
Boliden Mineral AB, 2003).
15
4. Methodology
4.1. Geochemical methods
4.1.1. Element concentrations
During several years since the closing of the mine, groundwater pipes have been installed within the area. Different pipes have been sampled since around the year 2000. The latest samplings from 2018, made by Golder Associates, are based on the geophysical profiles made by GeoVista (See section 4.2). Surface water sampling has also been done for a long time. The groundwater pipes and the sampling points for surface water used in this thesis are viewed in the picture below (Figure 8).
Figure 8 - A map showing sampling points for the different water types; freshwater, groundwater and water from the pit lake.
The different sampling points are illustrated in the map above (Figure 8). The map has been constructed in QGIS and the reference system for coordinates is SWEREF99. In Appendix I are the coordinates for the sampling points presented.
A sampling of groundwater and surface water has been done by Golder Associates AB and BergAB (Golder Associates AB, 2015; Golder Associates AB, 2018; BergAB, 2015).
4.1.2. Element ratios
To simplify the interpretation of the relationship between different elements in the different waters, molar ratios have been used. Ratios may be useful for connecting the ratio between two elements in specific water with a certain chemical reaction, such as a weathering reaction.
Ratios are good to use since they open up a visualization of real changes in metal levels which doesn’t only depend on the dissolution of e.g. rain or melting water.
16
For this thesis, four different molar ratios have been used; Ca/S, Fe/S, (Ca+Mg)/Na and (Cu+Pb+Zn)/Na. The first ratio, Ca/S, shows the ratio between calcium and sulphide, which can be connected to the dissolution or precipitation of gypsum (CaSO4 2H2O). The weathering of pyrite (FeS) can be related to the Fe/S ratio. The last two ratios are normalized with sodium (Na) and if for example, rainwater has negligible concentrations of the analysed metals (e.g. copper), the concentration of copper and sodium will be diluted just as much. This means that rain will not affect the ratio between e.g. copper and sodium. If the ratio sinks, it can’t be dilution that decreases the copper concentration, but it must then be a geochemical process that changed the copper content (but not the content of sodium). This can, for example, be adsorption of copper to hydroxides (Berner
& Berner, 1996).
4.2. Geophysical methods
Geophysical methods used at the Rävlidmyran site have been based on electrical fields, either natural or artificial, employed in in-situ investigations of the ground conditions. The resistivity and IP measurements where done by GeoVista (Perttu, 2018).
Figure 9 – Map showing the resistivity and IP profiles placement (the red lines). In the top right corner, a terrain shading map is showed (Perttu, 2018).
Error! Reference source not found. shows the location of the four profiles for the resistivity and IP readings.
The measurements of the electromagnetic properties (slingram method), were done by Golder Associates AB during 2014, using the measuring model CMD-explorer (Golder Associates AB, 2014).
17
Figure 10 - The planned location of the electromagnetic readings (Golder Associates AB, 2014).
The planned placement of the slingram readings is shown in Figure 10 above.
18
5. Results and discussion
5.1. Geochemical results
In this section, data for the most relevant elements are presented in diagrams and tables.
The entire data set is attached in APPENDIX II-IV. The elements chosen for a deeper study in this report are calcium, sulphur iron, arsenic, zinc, lead, and copper. The metals (Fe, As, Zn, Pb, Cu) have been chosen since they are common in the ore minerals of the Rävlidmyran deposit. Calcium occurs in the secondary mineral gypsum and in neutralizing agents (lime) used in acidic lakes. Sulphur was chosen due to the fact that the area is an old mining area and can give information about weathering of sulphide minerals and dissolution/precipitation of gypsum.
The elements aluminium, iron, manganese, lead, arsenic, nickel, zinc, cadmium and copper are more soluble in acidic conditions, i.e. conditions with lower pH. However, depending on redox conditions, some of them can bind to other ions and create almost immobile compounds such as hydroxides (FeOOH, MnOOH) or sulphides (FeS2, ZnS, PbS, CuS, CdS).
Zinc is more soluble in acidic conditions, but also when pH>11. In neutral conditions, it can be seen as almost insoluble. Cadmium often occurs with Zn, with a geochemical behaviour similar to that of Zn. The geochemical behaviour and mobility of elements thus depend on pH-changes and redox conditions, which is important to have in mind.
Another thing to have in mind is that dissolved inorganic solids, (anions and cations) increase the electrical conductivity of waters. A higher content of dissolved inorganic solids, like dissolved metals, increases the conductivity, which means that waters with a high content of metal contamination usually has a higher conductivity. The conductivity is also affected by temperature, with higher temperatures increasing the conductivity.
Different elements also differ in conductivity. For example, the conductivity for copper is higher than the conductivity for lead, which in turn is higher than the conductivity for mercury.
5.1.1. Groundwater
For the sets of groundwater pipes that have been sampled several times, the diagrams show an average value ± standard deviation. By examining the standard deviation of the different parameters, one can see the statistical measure of how much the different values deviate from the mean value.
5.1.1.1. pH
The pH-values in the different groundwater pipes are on average between 6.00 to 8.00, with some deviations, visualized in Figure 11 below. The full results are presented in Appendix II.
19
Figure 11 – Diagrams over pH values from the groundwater pipes. Diagrams a and b show the standard deviation (error bars) since the pipes have been sampled several times. The pipes in diagram c were sampled once (no standard deviation) Note the difference in scale in the y-axis.
The groundwater pipe with the lowest pH is pipe 13GVR3 in September 2013 (pH=5.11).
In June the following year, the pH in that area had increased to 9. This pH of 9 could be interpreted as a measurement error since such a high pH does not fit in the trend and in a mining area overall. This pH results in the large standard deviation (Figure 11b) and therefore, the average pH is not as trustworthy as for the pipes with a smaller standard deviation. The lowest average pH during 2013 is found in 13GVR9 (pH=6.19). Later, the lowest pH is found in 18GA20GV during 2018 (pH=6.36). This pipe is placed in the same place as 13GVR6 which had a pH of 6.50 during 2014. This means that the average pH in this spot during 2014 has decreased until 2018. Pipe 13GVR3 is on the other hand found on the western side and it is close to 13GVR4 which has a slightly higher pH. These two pipes are also rather close to 18GA09GV–18GA11GV, which had pH >7 in 2018 (Figure 11c).
5.1.1.2. Electrical conductivity
The electrical conductivity for all the groundwater pipes is presented in Figure 12 below.
The conductivity is affected by the dissolved ion concentrations. The elements that generally are affecting the conductivity the most by the analysed ones are copper, aluminium, and zinc, but in this case, calcium and sulphur affect the electrical conductivity the most, since the concentration of those elements is at such high levels.
20
Figure 12 – Diagrams showing conductivity from the groundwater pipes. Diagrams a and b shows the standard deviation (error bars) since the pipes have been sampled several times. The pipes in diagram c were sampled once (no standard deviation). Note the difference in scale on the y-axis.
Looking at the three diagrams above, one can see that the conductivity varies a lot. In pipe 18GA20GV, the conductivity is high, and the pH is low (Figure 11), which is a common relationship. This relationship can be seen in pipe 13GVR9 as well. Both these are sampled during different times, but both show the same trend when it comes to pH and conductivity.
21 5.1.1.3. Sulphur
Figure 13 – Concentrations of sulphur in all groundwater pipes, with the standard deviation shown as error bars. In diagram d, the pipes with the highest sulphur concentrations presented in diagrams b and c are presented. Note the difference in scale on the y-axis in all the diagrams.
The groundwater sulphur concentration in the area has decreased from 2010 and earlier to 2013 and onwards. The overall concentration of sulphur is high, especially south of the pit lake. The high concentration of sulphur is probably caused by the weathering of pyrite.
The levels of iron are not very high in comparison to sulphur though, which can be explained by the pH level in the groundwater, which is close to 7. This means that iron is immobile and therefore precipitates as iron-hydroxides.
22 5.1.1.4. Calcium
Figure 14 – Concentrations of calcium in all groundwater pipes, with the standard deviation shown as error bars. In diagram d, the pipes with the highest concentrations presented in diagrams a-c are presented. Note the difference in scale on the y-axis in all the diagrams.
The calcium concentration is highest in pipe 13GVR9 (534.5 mg/l) followed by 18GA20GV (386 mg/l) (Figure 14a-14d). Both pipes are placed on old backfilled open pits, south of the pit lake. These open pits are covered by 0.3 metres of glacial till.
In pipes 13GVR6 and 18GA20GV, the concentration of calcium increases with time, just like conductivity and pH.
23 5.1.1.5. Iron
The concentrations of iron [mg/l] are presented in Figure 15.
Figure 15 - Iron concentration in groundwater pipes, with the standard deviation shown with error bars. In diagram d, the pipes with the highest iron concentrations in diagrams a-c are presented. Note the difference in scale on the y-axis in all the diagrams.
The highest iron concentrations are found in groundwater pipes GWP3, 13GVR1, 13GVR9, 18GA17GV and 18GA20GV (Figure 15a-c, 59.5–222 mg/l), and in pipes 13GVR1 and 13GVR9 (Figure 15d, 125-150 mg/l). The standard deviation of 13GVR1 is large (±64.2 mg/l), most likely caused by the decrease in the iron concentration between 2013 and 2018. The conductivity for this pipe 13GVR9 is also high (Figure 12b). The groundwater in that pipe showed a rather low pH compared to the other pipes installed, so the relationship between pH, conductivity and iron is what can be expected in this pipe.
When the first sampling was done (2010), the iron concentration was highest between the Pit Lake and Lake Hornträsket (GWP3). Later, the concentration of iron was highest in the western part of the site (13GVR1). During 2018, the concentration was highest in the east (18GA17GV) and in the south (18GA20GV). The later groundwater pipe is placed at the same point as 13GVR6. Thus, the concentration at this point has increased generally over the years; from 21.0 mg/l to 70.6 mg/l, which both are considered as very high concentrations (SGU, 2013). Overall, the highest concentrations of iron are found in 13GVR1, even though the standard deviation is large (Figure 15d).
Geographically, the higher concentrations of iron are found south and west of the pit lake;
13GVR9, 18GA20GV, 13GVR1.
24 5.1.1.6. Arsenic
Figure 16 –Arsenic concentrations in groundwater pipes, with the standard deviation shown as error bars. In diagram d, the pipes with the highest concentrations in diagrams a-c are presented. Note the difference in scale on the y-axis in all the diagrams.
For arsenic, Figure 16 shows that pipes 13GVR6 and 18GA20GV have the same trend as for iron. With time, the concentration increased from a high average concentration of 6.87 µg/l to a very high concentration of 78.7 µg/l (2013 to 2018). However, these pipes do not reach the same level as 13GVR9, where the highest concentration was 143 µg/l. All these pipes are however installed south of the lake, at open pits refilled with waste rock.
Earlier in 2010, pipe GWP1 had the highest concentration, 18.7 µg/l, considered as very high (SGU, 2013). This pipe is located south of the lake, next to a backfilled open pit.
Just like iron, the location of the highest concentrations of arsenic are found in 13GVR9, 18GA20GV, 13GVR1, i.e. south and west of the pit lake.
25 5.1.1.7. Zinc
Figure 17 –Zinc concentrations in groundwater pipes, with the standard deviation shown as error bars. In diagram d, the pipes with the highest concentrations in diagrams a-c are presented. Note the difference in scale on the y-axis in all the diagrams.
Just as for iron and arsenic, the zinc concentration follows the same trend in 13GVR6 and 18GA20GV (Figure 17). The concentration increased sharply between 2013 and 2018 (405 µg/l to 54000 µg/l). Earlier, the concentration of zinc in the groundwater was lower (GWP1-GWP4) and the highest concentration was found in GWP2 (3968 µg/l), but it was not as high as later (16500 µg/l) in 13GVR3. Both these groundwater pipes are located west of the pit lake. The concentration of zinc in the groundwater during all sampling occasions has been according to SGU (2013) very high (≥1 µg/l). They also claim that the average background concentration of zinc in the groundwater in Sweden is 4.3 µg/l, thus the concentration of zinc in the Rävlidmyran area is above the overall background concentrations (APPENDIX II). Though, over the years the concentration of zinc in the groundwater has decreased (GWPX, 13GVRX vs. 18GAXXGV), with one exception. The concentration in 13GVR6 and 18GA20GV, as mentioned before.
Location wise, the pipes with the highest concentrations of zinc, are located south of the lake, in 18GA20GV/13GVR6 and 13GVR9. In 13GVR9, the concentration reaches 116000 µg/l.
26 5.1.1.8. Lead
Figure 18 –Lead concentrations in groundwater pipes, with the standard deviation shown as error bars. In diagram d, the pipes with the highest lead concentrations in diagrams a-c are presented. Note the difference in scale on the y-axis in all the diagrams.
Lead follows the same trend as iron, arsenic and zinc. The concentration in 13GVR6 and 18GA20GV increases with time and the highest concentration of lead is found in 13GVR9 (70.9 µg/l) in June 2014. The location of pipe 13GVR9 is southeast of the lake, and with the suggested groundwater flow, the groundwater passes through pipe 13GVR7. Despite this, the lead concentration in this pipe (0.02-1.14 µg/l) is much lower than in 13GVR9.
When the set of groundwater pipes named 13GVRX was sampled, the lowest pH and the highest conductivity were measured in 13GVR9, i.e. a pipe with groundwater with low pH, high conductivity and a high content of iron, arsenic, zinc, and lead. This pipe is placed in an old open pit filled with waste rock and covered with 0.3 metres of glacial till.
The highest concentrations are for lead found south of the lake, more specifically in GWP1, 13GVR9 and 18GA20GV.
5.1.1.9. Copper
27
Figure 19 – Copper concentrations in all groundwater pipes, with the standard deviation shown as error bars. In diagram d, the pipes with the highest concentrations presented in diagrams a-c are presented. Note the difference in scale on the y- axis in all the diagrams.
In the sampling of the groundwater in the Rävlidmyran area, copper does not follow the same trend as the metals presented previously (Fe, As, Zn and Pb). Instead, the highest concentration of copper is found west of the pit lake, in groundwater pipe 13GVR3 in June 2014 (529 µg/l). In 2018, the level of copper in this pipe decreased to less than 0,1 µg/l, which is considered as a very low level instead of the moderate level it had in 2014 (SGU, 2013). This could be a result of the dewatering of the pit lake, which started in 2018.
Southern of 13GVR3, 13GVR4 is placed. Both these pipes are on one side each of a sealed ventilated shaft. Pipe 13GVR4 has a low concentration of copper during 2013 and 2014 (0.1-2 µg/l) that increases until 2018 (260 µg/l).
The high standard deviation in 13GVR4 is due to an increase of copper in 2014 followed by a sharp increase in the concentration of copper in 2018 (0.1 µg/l to 260 µg/l). In 13GVR3, the large standard deviation is caused by the sharp decrease (529 µg/l to 0.1 µg/l).
