Processes Turning Saline Settling Basins into Freshwater Bodies (Selected Examples from the Upper Silesian Coal Basin)

Abstract

There are numerous deep coal mines in the Upper Silesian Coal Basin. Ensuring their proper exploitation requires constant drainage of the rock mass and the transfer of mine waters to rivers. Several technical solutions are used to prevent the adverse effects of saltwater discharge on the river ecosystem. One such solution is adapting the post-mining reservoirs into mine water settling basins. This article characterises two such facilities—the “Gliniok” and “Hubertus I” reservoirs. The physicochemical properties of their waters were analysed both when they served as settling basins and after their decommissioning. During their exploitation, the waters of the settling basins showed high salinity (>10 g/L). It was revealed that these basins turned into freshwater reservoirs very quickly after decommissioning. A sudden decrease in the electrolytic conductivity and the concentration of main cations and anions in the water was observed. The mixing processes also changed. The reservoirs were transformed from meromictic to polymictic. The processes that led to them turning into freshwater basins differed in the studied settling basins. The transformation of the Gliniok settling basin into a freshwater reservoir was a unique process, draining brines into the rock mass through cracks and crevices. The formation of cracks and crevices was a consequence of high-energy mining tremors. It is the first known case of this type in the world.

1. Introduction

There are many deep coal mines in the Upper Silesian Coal Basin. Ensuring their proper exploitation requires constant rock mass drainage and mine water removal. According to the Hydrogeological Dictionary [1], mine waters include water flowing into the workings from the drained rock mass and technological waters, most often introduced with hydraulic backfilling. Much of the mine water in the Upper Silesian Coal Basin is highly saline brine. Most of it shows mineralisation > 70 g/dm³ and ion concentration (Cl⁻ + SO₄²⁻) > 42 g/dm³ [2]. They may also contain significant concentrations of nitrates [3].

The discharge of brines into surface watercourses causes several adverse effects in freshwater ecosystems. An example was the 2022 mass fish kill in the Odra River [4,5]. Such waters also become unsuitable for irrigation in agriculture and as drinking water [6,7]. Several technical solutions are used to prevent the adverse effects of saltwater discharge [8,9]. They are intended to reduce the negative impact of brines on the environment. One is adapting post-mining reservoirs as mine water settling basins [10]. Post-mining reservoirs are created due to flooding workings after the exploitation of mineral deposits (Figure 1), which can then be used as mine water receivers. In these reservoirs, mine water is mainly purified mechanically from suspension. The brine is also slightly diluted by fresh groundwater and rainwater inflow into the reservoirs [10].

Mine water settling basins, including their hydrology [10,11,12], hydrochemistry [12] and biology [13,14], are a very interesting research theme. Some settling basins are meromictic, a unique mixing type rarely found in temperate latitudes [12]. In the Upper Silesian Coal Basin, mine restructuring resulted in the closure of numerous mines and the cessation of mine water pumping. Thus, mine water settling basins were put out of use. Currently, the limnic processes in the discontinued mine water settling basins are poorly understood, and the number of studies in this field is limited. In addition, they only apply to retention-and-dosing reservoirs, not post-mining reservoirs used as settling basins [11,15].

The processes leading to the changes in the physicochemical properties of water that occur in settling basins after decommissioning may depend on several factors. To identify them, this research was undertaken with the primary objectives of:

-

examining the changes in the morphometric parameters and type of water management of settling basins after their decommissioning,

-

examining the changes in the mineralisation and chemical composition of the water in settling basins after their decommissioning,

-

examining the changes in the physicochemical properties of water in the vertical column of the basins,

-

describing the processes (factors) that turned the settling basins into freshwater tanks,

-

indicating development directions for brine settling basins facing decommissioning.

2. Materials and Methods

2.1. Location and Characteristics of the Settling Basins

The tested reservoirs (settling basins) are located in the Upper Silesian Coal Basin (Figure 2). The Gliniok settling basin is a typical post-mining reservoir created in an excavation after the extraction of construction raw materials. It is located in the central part of Katowice (Figure 3). The exploited raw materials were highly weathered Carboniferous clay and shale used to produce construction ceramics. The reservoir is supplied by the drainage from the first aquifer in Carboniferous formations (sandstone reservoir). The Grunfeld settling basin (Figure 3) was selected as a control reservoir for the study. It is a reservoir of the exact genesis with the same type of exploited raw material and is supplied from the same aquifer.

The “Hubertus I” settling basin is a post-exploitation reservoir created in an excavation after exploiting Quaternary sands. It is located in the northern part of Mysłowice, within a larger complex of post-mining reservoirs (Figure 4). The reservoir is fed with waters from the first aquifer developed in Quaternary sands and gravels. The “Hubertus II” sedimentation reservoir was selected as the control reservoir for the study (Figure 4). It is a reservoir of the exact genesis with the same type of exploited raw material. It is supplied from the same aquifer.

2.2. Water Management in the Settling Basins

The studied reservoirs differ in the type of water management carried out during and after their operation as settling basins. In the case of Gliniok, mine water was discharged into the reservoir and, after initial removal of suspension, pumped again to the Wujek Mine as process water (Figure 3). The building of the pumping station is visible in an archival photo from the early 21st century (Figure 5). When the reservoir was used as a settling basin, water level fluctuations reached 6 m. After the end of operation (early 21st century), water pumping was stopped, and the brine remained in the settling basin (endorheic reservoir).

In the case of the Hubertus reservoir, there was a pumping station that, after initial purification of the water, transferred it to the Rawa River (Figure 4). After the settling basin was decommissioned (2016), pumping continued, and water from the reservoir was still transferred to the Rawa River. A constant water level is maintained in the reservoir. The pumped-out water is replenished by the inflow from the Quaternary aquifer formed in sand and gravel. No pumping work has been carried out on the control (Hubertus II) settling basins.

2.3. Research Methods

Mine water settling basins were studied following the methodology of Gutry-Korycka and Werner-Więckowska [16]. A Lowrance Elite Ti2 echosounder (Lowrance, Tulsa, OK, USA) integrated with a GPS receiver was used to obtain data on the depth of the reservoirs. The basins’ bathymetric plans were plotted using Reef Master 2.0. Their morphometric features were measured using a Tru Pulse 200 L laser rangefinder (Laser Technology, Inc., Centennial, CO, USA).

Profile measurepments of electrolytic conductivity [µS/cm] and water temperature [°C] were made in the deepest places of the settling basins. Measurements were made from a pontoon with a 0.5 m interval using the Aquaread Ap-2000 meter (Aquaread, Broadstairs, UK). Water samples were collected in 0.5 L polyethene bottles, transported and refrigerated at 5 ± 3 °C for laboratory analyses. Before analyses, the samples were filtered through a nitrocellulose membrane with 0.45 µm pore diameter (GVS Filtration Inc., Findlay, OH, USA).

Laboratory analyses included the determination of the major cations and anions in the water: Ca²⁺, Mg²⁺, Na⁺, K⁺, SO₄²⁻, Cl⁻, NO₃⁻, PO₄²⁻. The analyses were performed on the ion chromatograph Metrohm 850 Professional IC (anion column A Suup 7-250/4.0, eluent 3.6 mM Na₂CO₃ and a cation column C4-150/4, eluent 0.7 mM dipicolinic acid and 1.7 HNO₃). Bicarbonates (HCO₃⁻) were determined using titration with the alkalinity indicator b–r (blue–red).

The hydrochemical water type was based on the Altowski–Szwiec classification. In this classification, it was assumed that the hydrochemical type would be determined by those ions whose content in the water is greater than 20 ± 3% [meq] in relation to the sum of anions and cations. The name of the water begins with the ion with the highest concentration, regardless of whether it is a cation or anion. Under natural conditions, the six essential ions are present in amounts above 20% meq: Ca²⁺, Mg²⁺, Na⁺, HCO₃⁻, Cl⁻, SO₄². Statistical analyses were performed using the R language and environment [17].

3. Results

3.1. Gliniok Reservoir

Gliniok was a typical undrained reservoir in which thick brine fell to the bottom and formed a meromictic reservoir once the mine water discharge was discontinued. This is well documented by vertical profiles of electrolytic conductivity and water temperature (Figure 6 and Figure 7), frequently observed between 2008 and 2014 [10,11]. The salinity level was exceptionally high for the monimolimnion layer. Its electrolytic conductivity was approximately 36 mS/cm and was constant in all seasons. There was no decrease in salination in the reservoir. The thermal profiles were also characteristic. In summer, the waters heliothermically warmed up in the chemocline layer, and the temperature reached 32 °C. Also, in winter, the waters of the monimolimnion had a very high temperature of about 15 °C. Since 2015, a gradual decrease in the water level in the reservoir has been observed. Currently, the water level is about two meters lower than in 2010. This change is documented by comparing archival photos and a bathymetric plan from 2011 with a contemporary bathymetric plan (Figure 5; Figure 8).

The decreased maximum depth of the reservoir, reduced from 7 to 5 m, was initially associated with climate change. However, such changes were not detected in other post-exploitation reservoirs in the Upper Silesian Coal Basin (e.g., post-exploitation sand, dolomite and limestone workings). However, similar changes were recorded in the nearby Grunfeld reservoir of the same origin (Figure 9). The answer to the question about the nature of the changes in the water level of the Gliniok reservoir was provided by the continuation of research on the vertical profile of its water regarding the physicochemical features and chemical analyses.

The current measurements of the physicochemical properties of the water’s vertical profile showed that meromixis had disappeared in the reservoir. It is particularly well illustrated by the distribution of electrolytic conductivity (Figure 10). Its significant decrease (up to approximately 2 mS/cm) and the lack of water stratification are visible. The temperature distribution is also different. In summers, the heliothermal water heating zone disappears; in winters, the water temperature at the bottom is about 5 °C (Figure 11).

There was also an extreme drop in water salinity, identified by a decrease in the concentration of primary ions (

Table 1. Water chemistry [meq/L].

Reservoir	Ca2+	Mg2+	Na+	K+	HCO3−	Cl−	SO42−	Hydrochemical Type
Gliniok (during operation)	9.7	6.2	61	0.8	1.8	75	6.9	Cl–Na
Gliniok (currently)	3.8	4.5	23	0.5	3.5	26	3.9	Cl–Na
Grunfeld (2008 year)	2.8	2.1	0.6	0.2	2.7	1.9	1.1	HCO3–Ca
Grunfeld (currently)	2.3	2	0.6	0.2	2.4	0.6	2.4	HCO3–Ca–SO4–Cl
Hubertus I (during operation)	18	16.8	78	1.1	5.2	110	8.5	Cl–Na
Hubertus I (currently)	2.5	2.3	3.5	0.2	3.5	3.6	1.2	Cl–HCO3–Na
Hubertus II	2	1.6	1.3	0.17	3.3	1.4	0.4	HCO3–Ca

). However, the hydrochemical type of the water did not change. It is still di-ionic chloride–sodium water (Cl–Na). This issue seemed interesting since reducing salinity in an endorheic settling basin is impossible without pumping its water out. Gliniok and Grunfeld (Figure 3) are located in the zone of a dense fault network in Carboniferous formations [18]. These faults are still active, and their displacement is observed on the surface and deep in the rock mass [19,20,21]. Additionally, mining activities in the area cause numerous seismic shocks, the energy of which may exceed 105 J, and the Upper Silesian Coal Basin area is one of the most seismic coal basins in the world [22]. These tremors in fault areas lead to numerous ground deformations, facilitating easier water migration into the rock mass. This type of situation also occurred in the case of the tested reservoirs. In the Upper Silesian Coal Basin, there are known cases of complete disappearance of small streams whose waters infiltrate the rock mass through a system of rock cracks (caused by mining exploitation) [23]. In the case of the Gliniok reservoir, brine infiltrated the rock mass through a system of numerous (shock-activated) cracks (fissures, Figure 12). The groundwater and rainwater flowing into the reservoir caused the remaining brine to dilute and meromixis to disappear. The reservoir receives 807 mm of precipitation annually. As the reservoir surface is approximately 12,200 m², it gets an inflow of freshwater of ~10,000 m³.

3.2. “Hubertus I” Reservoir

The “Hubertus I” reservoir is also endorheic. However, it has a pumping station transferring water to the Rawa River (Figure 4). During the operation of the reservoir as a mine water settling basin, significant differences in the water physicochemical properties in its vertical profile were observed despite its small depth (Figure 13). The electrolytic conductivity profiles illustrate this nicely. The highest-density brine fell into the deepest zone of the reservoir, leading to a significant difference between the surface and bottom layers (Figure 14). This also applies to the unusual temperature distribution, particularly visible in winter. No ice cover developed in the mine water discharge zone, and warm, dense brine sank into the deepest part of the settling basin (Figure 15). Therefore, the water of the bottom layer showed a very high (for winter) temperature of about 12 °C (Figure 16). After the mine water discharge was stopped, pumping into the Rawa River continued. As a result, the brine was gradually removed and replaced by fresh groundwater from the first aquifer in the Quaternary sands. This process led to a gradual decrease in the basin water salinity, turning it into freshwater (Figure 17). There was a significant decrease in the electrolytic conductivity (<1 mS/cm); the water stratification in the vertical column of the basin disappeared (Figure 18). Thermal profiles in winter are also characteristic of water bodies in moderate latitudes (Figure 19). Thus, the reservoir was transformed from one with difficult water mixing into a polymictic reservoir with multiple water mixing instances throughout the year. In addition to the decrease in electrolytic conductivity, we also observed a reduction in the concentration of the main anions—chloride (Cl⁻) and sulphate (SO₄²⁻) (Figure 20 and Figure 21). However, the water in the reservoirs can still be described as an unusual chloride–bicarbonate–sodium hydrochemical type ((Cl–HCO₃–Na), Table 1). Thus, the reservoir differs from other post-mining reservoirs in sand excavations, as shown in Figure 4 (e.g., settling basins Hubertus II) as they are bicarbonate–calcium type ((HCO–Ca), Table 1).

4. Discussion

The processes turning saline mine water in settling basins into freshwater after they are no longer used are poorly understood. It would seem that in the case of saltwater endorheic reservoirs, such processes cannot occur because there is no possibility of draining the brine. This is how permanent salty endorheic lakes function, maintaining a more or less constant level of salinity [24]. Small changes in salinity in these lakes result from intensive evaporation in summer, causing an increase in salinity, or the inflow of fresh meltwater or precipitation, causing a decrease in salinity [25]. This is also how endorheic mine water settling basins function, which do not turn into freshwater basins after the end of use [10,12].

However, the studied reservoirs transitioned into freshwater basins. The factors that led to this transition are different in the case of the Gliniok and Hubertus reservoirs. A spectacular example is the Gliniok reservoir, where brine escaped into the rock mass. There are known cases of the disappearance of small streams in the Upper Silesian Coal Basin, whose waters infiltrate the rock mass [23]. So far, such a phenomenon has not been observed in the case of a water reservoir. The removal of brine from the reservoirs caused several changes in the physicochemical properties of the water. The most characteristic ones include the disappearance of meromixis and a hindered mixing of waters. In terms of mixing in the reservoirs, we again observe the processes characteristic of the reservoirs of moderate latitudes [10]. The disappearance of meromixis was also observed in the Rontok Wielki settling basin after decommissioning [15]. However, in the case of the Rontok Wielki settling basin, the water outflow was gravitational. In the Gliniok reservoir, the heliothermal water heating also disappeared. However, such changes are not observed in anthropogenic reservoirs with permanently maintained salinity. In the case of the “Hubertus I” reservoir, an abrupt decrease in the electrolytic conductivity and the concentration of chlorides and sulphates was observed in the first year after decommissioning (Figure 17, Figure 20 and Figure 21). Similar changes were observed in the Rontok settling basin [15]. There was a significant decrease in sulphates, although with substantial fluctuations between individual years. This situation resulted from the fact that the source of sulphates was no longer mine water. Their source was seepage water leached from the mining waste used to build the railway embankment in the shore zone of the reservoir.

Additionally, dry and wet depositions are the source of sulphates. High concentrations of sulphates in seepage from mining waste dumps have been found in numerous cases [26]. Molenda and Jankowska-Nitkiewicz [27] write about the significant influence of dry and wet depositions on the concentration of sulphates in water. An analogous situation was also found in the case of the Rontok reservoir, which harbours a mining waste dump in its shore zone [15].

Despite the fact that the sedimentation basins turn into freshwater, they still show a different hydrochemical type than that commonly found in the waters of post-mining reservoirs of the Upper Silesian Coal Basin. The waters of these reservoirs are most often of a hydrocarbonate–calcium (HCO₃⁻–Ca²⁺) hydrochemical type [15]. The decrease in the water level in the control reservoir Grunfeld also influenced the change in the hydrochemical type of water from hydrocarbonate–calcium (HCO₃⁻–Ca²⁺) to hydrocarbonate–calcium–sulphate–chloride (HCO₃–Ca–SO₄–Cl). This situation should be associated with the exposure to the previously submerged Carboniferous rocks containing pyrites. In aerobic conditions, their weathering is an efficient source of sulphates [28]. This had an impact on the hydrochemical type of water.

The pictures taken in the studied water bodies show plants growing in the shallow water and on the embarkments. In Appendix A, we provide a list of the most common plant species growing in and around the studied water reservoirs. Future studies should be focused on the plants and wetland vegetation. The saline habitat conditions can influence the plant physiology, e.g., increasing the synthesis of secondary metabolites.

5. Summary and Conclusions

Taking the settling basins out of use and discontinuing the discharge of mine water led to their transformation into freshwater bodies. In the case of the “Hubertus I” reservoir, the process was completed by pumping out salty water. In its place, poorly mineralised groundwater and rainwater flowed in, turning the reservoir into freshwater. A specific and only known case of turning an endorheic settling basin into a freshwater reservoir without gravitational drainage or pumping brine out is the Gliniok reservoir. In this case, the brine spontaneously discharged into the rock mass through a system of cracks (fissures) after high-energy tremors.

Turning mine water settling basins into freshwater reservoirs (reducing water mineralisation) reduces the concentration of the main cations and anions in the water. To this day, the water in the reservoirs presents a different hydrochemical type than the control reservoirs. Also, the concentration of the main ions is higher than those found in waters of reservoirs of the same origin and kind of exploited raw material and not subject to secondary anthropopressure. The most significant changes occurred in the physicochemical properties of water in the vertical profiles of reservoirs. Most of all, water stratification and meromixis disappeared. Mixing processes have become typical for reservoirs in temperate latitudes.

An interesting process was the change in the hydrochemical type of water in the Grunfeld reservoir caused by the decrease in water level. The observed change may indicate the direction of water transformation in reservoirs where the water level decreases, for example, due to climate change.

Green energy transformation will cause more and more coal mines to be closed in the near future. The vast majority of these mines have settling basins for salty mine waters. The data presented in the article indicate that if pumping operations continue, the settling basins quickly turn into freshwater. Such reservoirs may begin performing new natural, social and recreational functions. Ichthyofauna reappeared in the Gliniok and Hubertus reservoirs. The observed processes indicate that after the end of exploitation, it would be beneficial to pump out the brines. The fresh underground water and rainwater flowing in their place may quickly cause the reservoir to renaturalise.