Relationship Between Water Quality and Oil-Shale Mines in Northern Estonia

Different parameters and their combinations are used to evaluate and characterize the quality of the surface water: a) chemometric techniques (Brodnjak-Voncina et al., 2002; Kowalkowski et al. 2006); b) weekly chemical analysis of a wide range of major, minor and trace elements (Neal et al., 2006); c) multiple indicators of conditions (concentrations of compounds, intensity of agriculture and industry) (Cuffney et al. 2000); d) calculated water quality index (Pesce e al., 2000; Sanchez et al., 2007; Štambuk-Giljanovic, 1999).

mine water) (Equeenuddin et al., 2010;Mighanetara et al., 2009;Tiwary, 2001).The underground mines have significant impact of hydrosphere as large amounts of groundwater is pumped out into the bodies of surface water generating changes in hydrological and hydrogeological conditions.
The environmental impact of mines can be evaluated by the analysis of water quality during mining in comparison of water quality before the mining started.It means that it is important to measure many values of different parameters before constructions of mines.It will be good to analyze quite well the surface water sampled in the different places and at different time for the better comparison (Finkelman et al., 1999;Kolker et al., 2009;Natarajan et al., 2006).
If we have already working mines then it is quite problematic to get additional information about the surface water bodies which are under the direct impact of mines.It is possible to analyze the water sampled from the water bodies which are located in the similar climatic and hydrogeological conditions, but this situation has some uncertainty due to the differences of water sources, composition of soil and bedrock and biological diversity of water bodies.On the other hand the closing of mines gives the alternative opportunity to evaluate the environmental impact of mining during the restoration of natural conditions in the mining area.

Specific influence of oil-shale mining on the water quality
In certain mines of base metals and iron where ores have high sulfur content, drainage from mine workings and waste heaps can become highly acidic and can contain high concentrations of dissolved heavy metals.The acid generation from the oxidation of sulfurbearing minerals like pyrites in coal mines occurs only when the minerals react with water and oxygen in the presence of Thiobacillus bacteria to produce sulphuric acid and iron hydroxide or iron sulfate (Banks & Banks, 2001;Equeenuddin et al., 2010;Natarajan et al., 2006;Tiwary 2001).
Common characteristics of acid mine drainage include highly acid water with pH values as low as 2.0 and high sulfate concentration (up to 2 g L¯1) and the high metal loading typically including iron (up to 100 mg L¯1) and manganese (up to 5 mg L¯1) (Banks & Banks, 2001;Equeenuddin et al., 2010;Tiwary 2001).The high acidity results in further dissolution of minerals and the release of toxic metals and other constituents into waterways (Equeenuddin et al., 2010;Tiwary 2001).
Oil shale mining is accompanied by lowering of the water level and discharge of mine water into bodies of surface water (Rätsep & Liblik, 2000, 2001;Rätsep et al., 2002;Perens at al., 2006;Vaht & Rätsep 2009).For example, during the period of 1990-1998 the average amount of mine water directed into the Purtse catchment was 104 million m 3 per year (Rätsep & Liblik, 2000).An important problem is prognostication of hydrological processes after the shutting down of mines.The water in the mining area filtrates from one mine to another and feeds the waters of operating mines (Perens et al., 2006).There is also evidence that the decreasing amount of pumped-out mine water has a negative influence on the quality of the water of some rivers due to the decreasing amount of water in them (Punning et al., 1997).Sampling was made in July when the water level is usually at the lowest; water quality parameters are the most stable and well comparable.The summaries of results of physicochemical analysis are presented in Tables 1 and 2. The measurements of conductivity and alkalinity were included into the program of the complex hydrobiological field works after 2000, and because of that the values of conductivity and alkalinity of water samples are presented for 2005 only.Methodology of sampling and analysis was described in detail in our publication (Selberg et al., 2009) and it is not repeated here.

Spatial changes in the water quality in the Kunda River, directly unconnected with oil-shale mines
The Kunda catchment is located on the limestone base, and it means that the water samples are rather rich in carbonate ions (HCO 3 ¯ + CO 3 2 ¯) resulting in high alkalinity (Table 1).In the present study alkalinity of water samples was calculated in millimoles of bicarbonate ions per m 3 of water (HCO 3 ¯ mmol m¯3) for better comparison of the amounts of different ions in the water samples.The values of pH of water samples from Kunda River (Table 1) were mainly around 8 as typical for the surface water in northern and western Estonia due to the limestone.The alkalinity of water samples from the sampling points of the Kunda catchment was on an average 4700 mmol m¯3, independent of sampling points.Conductivity of the water samples of the Kunda River was on the average 500 microsiemens per centimeter (µS cm¯1), practically independent of the sampling point.
The concentrations of total nitrogen and nitrate nitrogen of water samples from different sampling points of the Kunda River had quite similar values (Table 2), and during ten years (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005) the concentration decreased on the average only ca 20%.In the water samples of the Kunda catchment the concentration of nitrate nitrogen formed more than 70% of the total nitrogen concentration (Selberg et al., 2009).
The water samples of Kunda catchment had quite steady concentration of nitrite (below 1.2 mmol m -3 ) over all sampling area and period (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005) (Table 2).The calculated mean values of concentration of NO 2 − -N decreased during the ten years (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005) as the mean values were 0.53±0.34,0.48±0.18and 0.29±0.1 mmol m − 3 , respectively.The concentrations of ammonia in water sampled from the Kunda catchment had also quite steady (Table 2).The highest concentration of NH 4 + -N up to 3.1 mmol m − 3 was determined in water sampled from sampling point of Lammasmäe in 1995, but it was below 1 mmol m − 3 in the other sampling points.In 2000 the concentrations of ammonia were below 1.7 mmol m − 3 , while in 2005 the highest concentrations of ammonia up to 2.3 mmol m − 3 were determined in water sampled from sampling points of Kulina and Kohala.The calculated variations of concentrations of different forms of nitrogen showed that concentrations of ammonia even were grown during the ten years and it can be connected with development in agriculture and settlement.
The fluxes of inorganic nitrogen formed more than 85% from the fluxes of total nitrogen in water of the Kunda River in 1995 and fluxes of toxic NO 2 − -N and NH 4 + -N formed less than 1% of inorganic nitrogen (Table 3).In 2005 the fluxes of nitrogen were lower than in 1995, but the discharge of water was also lower in the upper reach of the Kunda River.The inorganic nitrogen formed more than 85% from the fluxes of total nitrogen in the upper reach, but it was below 75% in the lower reach of the Kunda River in 2005.The fluxes of toxic NO 2 − -N and NH 4 + -N formed less than 2% of inorganic nitrogen.The fluxes of different forms of nitrogen increased further downstream of the Kunda River indicating the continuous infiltration of nitrogen from the untreated or purely treated wastewater of settlements and agricultural sources into the Kunda River in 1995 and 2005.
Sampling point   The Purtse River is located mainly in Ida-Viru County in North-East Estonia (Fig. 1).The total length of the Purtse River is 51 km and the catchment covers an area about 816 km 2 .The whole catchment area is located in the western part of the Estonian oil shale deposit and the catchment area is rich in tectonic faults and karst (Rätsep & Liblik, 2001).
Besides the mine water, the Kohtla River (tributary of the Purtse River, Fig. 1) is also polluted with toxic ash-dump waters of the Kohtla-Järve oil shale processing plant, which contain in addition to sulfates (above 600 mg L¯1) and chlorides (up to 500 mg L¯1) also phenols (above 50 mg L¯1) and sulfides (above 200 mg L¯1) (Rätsep & Liblik, 2001).In 1990 the amount of mine water directed into the Purtse River catchment was about 137 million m 3 per year (Rätsep & Liblik, 2000), while the amount of mine water going into the Purtse River decreased to 26 million m 3 per year in 2001 as the Kohtla mine was closed (Rätsep et al., 2002).
Water samples of the Purtse River were sampled and analyzed by different scientists and because there are some problems in comparing different results.The sampling points of the Purtse River are presented in Fig. 1A and results are presented in Tables 6 and 7.In 1995 the concentrations of total, nitrate and nitrite nitrogen (Table 6) in water sampled from the Purtse River were lower than concentrations of respective fractions of nitrogen of water sampled from the Kunda River (Table 2) which can be connected with huge amounts of discharged mine water.In 2000 their concentrations were lowered more than 60%, but in sampling point of Purtse the concentrations were increased more than 30%.Only the concentrations of ammonia were higher in water sampled from the Purtse River in comparison with the results of Kunda River.In 2000 the concentration of ammonia was increased in sampling point of Kohtla (Table 6) and it was lowered in the other sampling points in comparison with results in 1995.In 2005 the concentrations of different forms of nitrogen were significantly increased in water sampled from the Kohtla River.It can be connected with leachate from the ash-dumping heaps.
The concentrations of different forms of phosphorus are presented in Table 7 and in 1995 their values were quite similar to results for the Kunda River (

The Pühajõgi River
The Pühajõgi River is located in Ida-Viru County in North-East Estonia (Figure 1B).The length of the main river is 28 km, and the catchment covers an area of 196 km 2 (Järvekülg, 2001).The whole catchment is located in the western part of the Estonian oil shale deposit area and large volumes of mine water were discharged into the upper reach of the Pühajõgi River (Vaht & Rätsep, 2009).The Pühajõgi River is not so well studied (Selberg et al., 2009, Vaht & Rätsep, 2009), but it was included into the hydrological field works.Water was sampled in different points of the main river and streams in 1995, 2000 and 2005, and the sampling was made in July.Summary of results is presented in Tables 8-12 (Selberg et al., 2009, Selberg, 2010).
The analyzed water samples were mostly weakly alkaline (pH 7.3-8.0)and they had quite high alkalinity (Table 8).The results show that the alkalinity of river water samples decreased on the average to 66% of that of sampling point of Kukruse (6000 mmol m¯3) to Toila-Oru (4200 mmol m¯3) in the Pühajõgi catchment.At the same time the alkalinity of water samples from the sampling points of the Kunda catchment was on an average 4700 mmol m¯3), independent of sampling points (Table 1).Although the concentration of carbonate and bicarbonate anions was quite similar in the water samples of both catchments, the measurement of conductivity showed a significant difference.Conductivity of water samples of the Pühajõgi River was between 800 and 1800 µS cm¯1 which was 1.5-3 times higher than the conductivity of water sample from Mägara sampling point (tributary of the Pühajõgi River, Fig. 1).Conductivity of the water samples of the Kunda River was on the average 500 µS cm¯1, practically independent of the sampling point (Table 1).
The determined concentrations of nitrate nitrogen in water samples of the Pühajõgi catchment (Table 9) had quite different values (variation up to 10 times), and the highest concentration of nitrate nitrogen (252 mmol m¯3) was determined in the sampling point of Toila-Oru in 1995.The concentrations of nitrate nitrogen were very low (up to 1% from the concentration of total nitrogen) in the sampling points of Kukruse and Toila crossroads which means that nitrogen occurred in other forms (ammonia and organic compounds).The concentrations of total nitrogen and nitrate nitrogen were lowered in 2000 and 2005 mainly in the downstream of the Pühajõgi River.
The analysis of water samples showed that variations of concentrations of ammoniacal nitrogen were significant (between 1.1 and 520 mmol m¯3) and the higher concentrations of NH 4 + -N up to 340 mmol m¯3 (up to 99% from N tot ) and 520 mmol m¯3 (up to 81% from N tot ) were determined in the water sampled from Kukruse and Toila crossroads in 1995, respectively.It confirmed that nitrogen occurred in form of ammonia.The concentration of ammonia decreased significantly during the next ten years indicating the reduction of pollution of the Pühajõgi River.
Concentrations of nitrite were quite low (up to 7.8 mmol m¯3) in the water samples from the catchment of the Pühajõgi, whereas variations of concentrations of ammoniacal nitrogen were significant and the highest concentration of NH 4 + -N up to 520 mmol m¯3 was determined in 1995.The concentration of nitrite nitrogen was increased almost in all the sampling points of the Pühajõgi River in 2000, but it decreased during the next five years (Table 9).
Sampling point Calculated fluxes of nitrogen showed that 80% of nitrogen had infiltrated into the Pühajõgi River as NH 4 + -N in 1995.The fluxes of NO 2 ¯-N and NH 4 + -N were reduced in further downstream of the Pühajõgi due to oxidation as the fluxes of nitrate increased (Table 10).In 2005, the fluxes of nitrogen had reduced 3-5 times in comparison with 1995 and the fluxes of organic nitrogen constituted more than 50% of the fluxes of total nitrogen in the Pühajõgi River.
Sampling point N tot , mg s¯1  The results of water analysis showed significantly different concentrations of total phosphorus and phosphate phosphorus in Pühajõgi and Kunda rivers (Tables 4 and 11).The concentrations of total phosphorus were even close to 100% lower during the ten years (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005) which was a good sign for the improvement of water quality in the Pühajõgi River.
The comparison of P tot and [PO 4 3 ¯-P] concentrations showed that the phosphorus occurred mainly as phosphate anion (up to 90%) in the samples of the Pühajõgi River.Water samples from sampling point Mägara (the tributary of the Pühajõgi River) and the Kunda catchment contained a lot of organic phosphorus compounds, and the content of phosphate phosphorus was less than 50% of that of total phosphorus.During the 10-year monitoring period (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005), the concentration of total phosphorus was decreased by 93 to 98% in the water samples of the Pühajõgi River and this is related to the closing of oil shale mines.
In 2005 the situation of the fluxes of phosphorus was much better and PO 4 3 ¯-P formed about 50% from the fluxes of total phosphorus as it was in sampling point of Mägara (the tributary is not directly connected with oil shale mines) in 1995.The fluxes of phosphorus increased further downstream of Pühajõgi River indicating the continuous infiltration of phosphorus into the Pühajõgi River in 1995 and 2005.
The fluxes of PO 4 3 ¯-P formed about 50% from the flux of total phosphorus in water of the Kunda River (

Conclusion: Environmental impact of closing of oil-shale mines on the surface water quality
Four oil shale mines were closed in 1999-2002 and the discharge of mine water into the rivers was reduced.The results of water analysis give information about the restoration of natural conditions in the rivers.The chemical status of the Purtse River is moderate on the basis of data obtained from the supplementary monitoring of 2003 and the national monitoring.For the drainage of mine waters in the Purtse catchment area, a new hydrological scheme is needed to restore gradually the natural hydrochemical conditions of the Purtse River water.However, the inlet of the ash-dump waters into the Kohtla River must be fully stopped.In connection with the closing of mines the impact of mine water on the hydrologic state of the rivers in the Purtse catchment area decreased significantly, whereas its changed balance has set more favorable conditions for development of phytoplankton.
The determined concentrations of nitrate nitrogen in water samples of the Pühajõgi catchment had quite different values (variation up to 10 times), and the highest concentration of nitrate nitrogen (252 mmol m¯3) was determined in 1995.The concentrations of total nitrogen and nitrate nitrogen were lowered in 2000 and 2005 mainly in the downstream of the Pühajõgi River.The analysis of water samples showed that variations of concentrations of ammoniacal nitrogen were significant (between 1.1 and 520 mmol m¯3) and the higher concentrations of NH 4 + -N were determined in 1995.The concentration of ammonia decreased significantly during the next ten years (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005) indicating the reduction of pollution of the Pühajõgi River.During the 10-year monitoring period (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005), the concentration of total phosphorus was decreased by 93 to 98% in the water samples of the Pühajõgi River and this is related to the closing of oil shale mines.
The analysis of water quality showed that the Pühajõgi River could achieve natural conditions after closing oil shale mines, but it needs longer time as additional pollutants can be dissolved from the sediments of rivers.

Table 3 .
The fluxes of dissolved nitrogen in the sampling points of the Kunda RiverThe concentrations of total phosphorus (P tot ) and phosphate phosphorus [PO 4 3 ¯-P] in the water of the Kunda River were quite low in 1995 (Table4) and the concentrations of phosphorus were increased in the most sampling points in 2000.
*-Sampling points of a tributary

Table 4 .
Concentrations of different forms of phosphorus in the water samples from the sampling points of the Kunda catchment in July 1995 and their changes in July 2000 and 2005.In 2005 the concentrations of phosphorus were reduced in all sampling points of the Kunda River.The variations of phosphorus concentrations can be connected with weather condition of sampling time, because rain water can lower the concentration due to dilution.The calculated fluxes of phosphorus were quite similar in the water of the Kunda River in 1995 and 2005 and fluxes of PO 4 3 ¯-P formed about 50% of the flux of total phosphorus (Table5).
*-Sampling points of a tributary

Table 5 .
Fluxes of dissolved phosphorus in the sampling points of Kunda River www.intechopen.com 5.

Table 6 .
Concentrations of different forms of nitrogen in the water samples from the sampling points of the Purtse catchment in July 1995 and their changes in July 2000 and 2005(or 2011).
Table4). is no the clear trend for the concentrations of phosphorus in the water samples of the Purtse River in 2000 and 2005.In the downstream of the Purtse River (sampling point Purtse) the concentration of phosphorus was increased, but it was lowered in 2011.
x -Samples from Kohtla and Lüganuse were analyzed in 2011.

Table 7 .
Concentrations of different forms of phosphorus in the water samples from the sampling points of the Purtse catchment in July 1995 and their changes in July 2000 and 2005 (or 2011).

Table 8 .
pH, conductivity, alkalinity and concentration of total dissolved solid in the water samples from the sampling points of the Pühajõgi catchment in July.

Table 9 .
Concentrations of different forms of nitrogen in the water samples from the sampling points of the Pühajõgi catchment in July 1995 and their changes in July 2000 and 2005.

Table 10 .
The fluxes of dissolved nitrogen in the sampling points of Pühajõgi River Table 12) and similar situation was observed for Pühajõgi River in 2005.It means that the Pühajõgi River can achieve the natural conditions after the closing of oil shale mines.
*-Sampling points of a tributary

Table 11 .
Concentrations of different forms of phosphorus in the water samples from the sampling points of the Pühajõgi catchment in July 1995 and their changes in July 2000 and 2005.

Table 12 .
Fluxes of dissolved phosphorus in the sampling points of Pühajõgi River www.intechopen.com