FORECAST OF INFLUENCE OF PLANNED SAND OUTPUT FROM UNDER WATER FROM “OBORA” SANDPIT ON LEVEL AND QUALITY OF UNDERGROUND WATER

Mieczyslaw Chalfen, Wieslaw Szulczewski

ABSTRACT

The analysis of the influence of planned increase of sand output from under water in “Obora” sandpit on water relations of neighboring areas is conducted in this paper. The bases for the performed simulations were mathematical models based on Boussinesq equations and hydrodynamic dispersion equations. The model includes all most important factors occurring in a given area influencing the hydroisohipses system and lines of chemical pollutant concentrations. Apart from typical hydrogeological elements such as water-courses, wells, and supplying from the aeration zone, the terrain’s depression caused by copper ore exploitation was also considered. As results from the carried out multi-variant simulation computations, the planned sand output will not cause the formation of depression hopper and will not significantly deteriorate the quality of underground water in the area surrounding the sandpit.

Key words: Boussinesq equation, hydrodynamic dispersion equation, mathematical model, filtration, contaminant migration.

INTRODUCTION

A considerable increase of sand output from under water in “Obora” sandpit is planned for the next 10 years, accompanied by successive enlargement of the open reservoir area from present 5 to planned 35ha. Unfortunately, it may have a significant impact on the arrangement of piezometric pressure around the sandpit. The fact that the water intake of Owczary – SI and SII wells is located in the close neighborhood of the sandpit (Figure 1) calls for particular attention. Besides, the potential formation of enlarged depression hopper might result in increased inflow of contaminated water from “Gilow” reservoir of post-flotation waste located 3km away. A precise answer to these basic questions requires the application of advanced simulation technologies taking into account all the most important factors influencing the level variation of ground water, and contaminant migration.

MATHEMATICAL MODEL

Mathematical models of water and contaminant flow in the saturation zone are based on Boussinesq equation for water movement and the hydrodynamic dispersion equation for contaminant migration. In both cases, two-dimensional models of horizontal nature were applied including space variables x, y,

• Boussinesq equation

• equation of hydrodynamic dispersion

The application of dynamic models, i.e. taking into account time, allows not only for obtaining the stationary state picture, but also for evaluating the dynamics of phenomena occurring in the saturation zone. The accepted equations permit to take into account the variation in time of interior boundary conditions, the intensity of water input in wells, and also the variation in time of the level of contaminant sources. Both equations are solved simultaneously by means of the finite elements’ method based on triangular elements with linear base functions. The computations were carried out by the use of FIZ author’s program [1].

FILTRATION AREA

The computations were conducted for the area of the shape close to a polygon determined by geodetic points 351, 354, 369, 371, 353, 360, 356 357, 355 according to the Hydrogeological Map of Poland [2]. The description of geodetic points mentioned above is included in Table 1. The coordinates of points in kilometers in the system of topographic coordinates are provided in column 2 and 3 of this table. Column 4 contains ordinates of piezometric pressure for the initial condition accepted on the basis of the Hydrogeological Map [2] of year 1967, measured in meters from the sea level. These points were selected in such a way that the distance from the area edge to the centrally located “Obora” sandpit was not shorter than 4-5 km. With this distance, the impact of boundary conditions on potentially enlarged depression hopper is hardly significant, considering the coefficients of water-permeability and effective porosity present in this area. Further calculations confirmed the legitimacy of th is initial assumption. Also, the defined area comprises all significant elements influencing the form of piezometric pressure in the past, at present and in future, i.e.:

• Obora sandpit,
• water intake of Szklary-Jedrzychowice,
• Owczary water intake,
• Zielenica and Zimnica rivers,
• post-flotation waste reservoir Gilow.

The assigned area was divided into 276 triangles by means of 151 nods, the division being congested in the region of Obora sandpit, in the area of the planned open reservoir as well as in the forearea of Gilow reservoir. The division was less congested in the borders of the region. Important hydrological points of the area were selected as network nod points.

HYDROGEOLOGICAL DATA

The level of the impermeable layer deposition of quaternary layer and the terrain’s surface was accepted on the basis of data procured from [5]. Piezometric ordinates from the map elaborated by Downarowicz for the year 1967 [4] were accepted as the initial condition of the simulation (figure 1). In the paper, the ordinates of impermeable layer, terrain and piezometric pressure were considered in relation to the sea level in all figures. The coefficient of effective porosity was accepted as [5] equal to =0.14. The coefficient of water permeability for a selected region is provided in available research papers in the range from 2m/d [7], through 18m/d [5], up to 20m/d [6]. As these measurements demonstrated a non-univocal character in further computations, originally a uniform and estimated coefficient of water-permeability was accepted for the whole area, equal to k=15 m/d. In the phase of the model taring, this coeffici ent was modified in selected regions of the area.

EXTERNAL BOUNDARY CONDITIONS

Along nearly the whole border of the region, boundary conditions of Dirichlet type were assumed, providing the ordinates of piezometric pressure in boundary nods complying with the hydroisohipses arrangement in the year 1967. It was accepted that these ordinates did not change in time and were valid throughout the simulation period. Only in the north-western section of the border, between the geodetic points of 354 and 369, the Neuman condition equaling zero was accepted, since the region’s border in this area overlaps with the border of Zielenica river basin.

INTERNAL BOUNDARY CONDITIONS AND STRUCTURES CAUSING CHANGES IN WATER LEVEL

There are two rivers in the discussed area: Zielenica and Zimnica. In water nods, through which the two rivers flow, Dirichlet type conditions were assigned complying with mean water levels in water courses [5,8]. The water intakes of Owczary-Jedrzychowice were marked with appropriate symbols SI, SII and S4, S7, S10, S12 and S14 in all figures according to source maps. In the network nods locating the water intakes mentioned above, the source function was assigned with the input rate complying with the output table in the years 1967-1996, and with the output forecast up to the year 2005 in m³/year [9] – table 2.

The region of Obora sandpit was treated in an analogous way. The sandpit was marked in the central part of the filtration area in all figures. The planned and enlarged water region will be located in the northern part of the sandpit. The sand output will take place only in the area of the water region. In the network nods located in the area of the planned water region a source function was assumed complying with the table of sand output from under water comprising the period of 1992-1997 [10] – table 3. In the years 1998 – 2007 the output was assumed according to the project of the order of 350000 m³ per year. Apart from increasing the sand output, the enlargement of open water region of up to 35 ha is planned in the forecasted period. In the model, the water volume filling the water region is assumed to be equal to the volume of put out sand. Thus, it was assumed that the values of the source function describing the output are equal to the values provided in table 3, while the output in particular operation years is spread on larger and larger area according to the planned water region enlargement [3].

Gilow reservoir of post-flotation waste was treated as a dry reservoir in the model, having no influence on the arrangement of equipotential lines. It results from the fact that the reservoir’s operation ceased in 1980, which led to its nearly complete drying-up after several years. Therefore, its impact was taken into consideration only in the part of the model concerning the quality of underground water.

CONTACT WITH AERATION ZONE

The distributions of index spatial evaporation, current evapotranspiration and precipitation deficiencies in lowland Poland in the period of 1951-1990 are included in papers [11,12,14]. As for the data related to mathematical modeling of retention capacity of small lowland river basins of regions located tens of kilometers northwards from the discussed area, they are presented in paper [13]. The analysis of evaporation, evapotranspiration and precipitation was made for each day of simulation in three variants, for dry, average and wet years. An example of the evapotranspiration value is presented in table 4. The characteristic of precipitation was accepted on the basis of measurements from the precipitation post of Lubin comprising the period of 1961-1995 (table 5). For the subsequent years, after 1995, the characteristics of precipitation was accepted as average values from the whole period of 1961-1995, preserving the division into dry, average and wet years. In t he regions of open reservoirs, the values of index evaporation were accepted instead of the values of evapotranspiration. The analysis elaborated in this way was introduced into the model as a source function.

GROUND SETTLEMENT

Non-uniform ground settlement can be observed in the analyzed region for the reason of copper ore output. The value of settlement oscillates between 0.5 and 2.5 meter in various points of the region [17] (Figure 2). A map of forecasted ground settlement of up to the year 2013 is also available in elaboration [15] (Figure 3). On the basis of these data, average-yearly rate of ground subsidence in particular network nods was calculated, separately to the year 1994, and separately in the forecasted period to the year 2007. The ordinates of impermeable layer and terrain were lowered in the simulation period at the beginning of each computation year, in accordance with maps presented in figure 2 and figure 3.

CONTAMINANT MIGRATION

The main contamination source in the discussed area is Gilow reservoir of post-flotation waste, which operated in the period of 1967-1980 [17]. When simulating the migration process, it was presumed that Gilow reservoir was a superficial contaminant source in this period, realized in the model by Dirichlet condition with the contaminant concentration equal to 100% (Figure 4). After the year 1980, the reservoir was no longer exploited, which caused its drying-up nearly on its entire surface. Only in the southern part of the area there is a tiny water region of contaminated surface water. It was accepted in the model that beginning with the year 1980 the reservoir in Gilow should be described in the non-dried area, by means of Dirichlet condition constant in time with the concentration equal to 100%. As for the dry area, it is not a source of contaminant emission in this period. On the border of the area, the condition of the third type was assumed for the contaminant flow, that is a conditi on of free outflow/discharge. The constant of the longitudinal dispersion was accepted to be [18] equal 8m. The constant of the crosswise dispersion was defined, in compliance with theoretical recommendations, as several-fold lower; in this case it is 2 m. The process of contaminant adsorption through the ground medium was presumed not to take place in the model.

MODEL TARING

The empirical material comprising the period of 1967-1997 was used for the model taring. As a result of correction of the water-permeability coefficient accepted in the beginning, it was possible to obtain a realistic piezometric picture (Figure 5). The river basins of Zielenica and Zimnica with the watershed going through the area of the sandpit can be clearly visible in figure 5. The central region is supplied with water from the south, and from the north from the area of Gilow reservoir. The discharges take place eastwards and westwards, in the direction that both rivers flow. Despite intensive sand output from under water in the years of 1992-1997, a significant depression hopper did not appear in Obora sandpit. The ordinates of piezometric pressure stay within the range of 152-153 m in this region. The influence of dry and wet years on the variation of underground water level was also examined in this phase of the model taring. On the basis of simulation computatio ns one can state that one dry year causes the lowering of piezometric pressure of approx. 1m in relation to average years, while a wet year results in the level rising of over a meter.

The basic direction of contaminant migration coming from Gilow is the same as the direction of advective movement determined by the main directions of water flow (Figure 6). The lines of 50% and 10% of contaminant concentration cross the borders of Zielenica-Zimnica rivers and reach the region of the sandpit. After the year 1980, a significant reduction of contaminated water region has been observed, particularly northwards from the reservoir. It is a result of the main direction of pure water discharge from the north and the washing out of contaminated water in the direction of Zielenica and Zimnica rivers.

PROGNOSTIC COMPUTATIONS UP TO THE YEAR 2007

In this phase of the research we attempted to answer three basic questions.

1. Depression hopper – Will the increased operation in Obora sandpit cause the formation of significant depression hopper?

2. Water intake endangerment – Is there a real endangerment for the water of Obora and Owczary intake from the part of contaminated water flowing from the northern side from Gilow reservoir?

3. Ecological disaster – What would be the directions of contaminant migration in case of a potential ecological disaster in the region of Obora sandpit?

Depression hopper

When investigating the problem of potential enlargement of the depression hopper around Obora sandpit, a comparison of computed piezometric pressure was made with the planned output of 350 000 m³, and then with the output at the level present so far, and with no output in particular years from 1998 to 2007. The comparison was drawn in three computation variants: for dry, average, and wet years. An example of the hydroisohipses map for evapotranspiration and precipitation characteristic of average years, at the end of the year 2006, is demonstrated in figure 7. For the average years, the differences of ordinates of piezometric pressure between extreme computation variants, i.e. between the planned output of the order of 350 000m³, and the results in case of complete resignation from output are not higher than 0.3m (Figure 8). Figure 8 indicates the izolines of differences of piezometric pressure bet ween extreme computation variants. The formed enlarged depression hopper has a diameter smaller than 2km and, in principle, does not go beyond the sandpit limits. In the medium computation variant, with the future output at the level kept so far, the differences of piezometric pressure will be even smaller. The conducted simulation computations show that the two subsequent dry years, with the maximal sand output, cause the lowering of the water level of maximum 2 meters in relation to average years. It gives an estimated rate of lowering of underground water of 1 meter a year. As a result, the piezometric ordinates on the terrain of the sandpit lowered from the average 152 –153m to approx.150m.

Figure 7.

Figure 8.

The lowering does not cause any considerable changes in the izoline arrangement of contaminant concentration in relation to average years. The maximal concentration difference was not more than several per cent. A similar analysis was performed for two subsequent wet years. In this case, an increase of piezometric pressure was observed, maximally of 2.5m in relation to average years. One can infer that the annual rate of rising approx. amounts to 1.2m. Similarly as for dry years, a wet year does not cause any significant changes in the izoline arrangement of contaminant concentration (figure 9). Summarizing this part of the analysis one can state that supplying from the aeration zone has a more vital importance for the hydroisohipses system in the region of the sandpit than the quantity of output from the sandpit.

Water intake endangerment

The computations of contaminant concentration were carried out for all three calculation variants, i.e. for the planned output, for the one applied so far and for the case of giving up the operation of Obora sandpit. The comparison of calculated concentrations prove that neither the quantity of output in the sandpit nor supplying from the aeration zone have a considerable influence on the range of the contaminated water front. The process of reducing the area of contaminated underground water was not inhibited in any computation variant. It is worth highlighting that the simulation computations of up to 2007 were conducted with the assumption that the region of contaminated surface water in the southern part of Gilow would be present. However, a potential sooner drying-up of the region may have a significant impact on reducing the area of contaminated water in the pre-area of Gilow.

Ecological disaster

It was assumed in this variant that the entire region of open water in Obora sandpit is the momentary contamination source. The contamination concentration equal to 100 % was assumed in all the network nods forming the sandpit in the model. In order to isolate the influence of Gilow reservoir on the arrangement of concentration lines being formed around the sandpit, the computations in this variant were realized with the assumption that the contaminants from Gilow reservoir did not migrate. Using such an assumption, the main direction of contamination migration would be a westward direction towards SI and SII wells. After one year, the contaminant concentration in the nod corresponding to the location of the wells mentioned above amounted to 0.2%, and after two years of simulation – over 1% (Figure 10). The slow course of the migration process results mainly from small gradients of underground water flow in the area of the sandpit. A potential ecological dis aster in the area of the sandpit is believed to have a minimal importance for the quality of underground water, in comparison to the effect of operation of Gilow reservoir.

CONCLUSIONS

The multi-variant simulation computations conducted in the region surrounding Obora sandpit allow stating that there is no risk of considerable enlarging or deepening of the depression hopper, with the assumed, for the nearest 10 years, increased sand output from under water of up to 350 000 m³ per year, and successive enlarging of the open water region of up to 35ha. The output should not reduce the discharges of surrounding water intakes in Owczary, either. Besides, there is no danger of inhibiting the positive tendency of regression of the contaminant front migrating from Gilow reservoir. As a result, the water quality in the planned open reservoir and Owczary intake will not deteriorate. Also, a potential momentary ecological disaster in the area of the sandpit will not cause the water endangerment in Owczary intake, though the contaminant concentration in the wells will increase in the course of time.

REFERENCES

1. M.Chalfen, Matematyczny model nieustalonego ruchu wod podziemnych z uwzglednieniem obiektow melioracyjnych oraz ujec wody, ZNAR Melioracje XXXVI, nr 192, 1990,
   (M.Chalfen, Mathematical model of non-steady movement of ground water with regard to land improvement structures and water intakes)

2. D.Jasiniak, J.Wojciechowski, Mapa Hydrogeologiczna Polski – Arkusz Leszno, Wyd. geologiczne, W-wa 1990,
   (D.Jasiniak, J Wojciechowski, Hydrogeological Map of Poland – Publisher’s sheet of Leszno, Geological Publishing House)

3. Projekt eksploatacji spod wody w piaskowni “Obora”. GeoSoft , Wroclaw. 1998r.
   (Design of exploitation from under water in “Obora” sandpit)

4. S.Downarowicz, Mapa hydroizohips poziomu czwartorzedowego rejonu zbiornika szlamow poflotacyjnych “Gilow”, Lubin 1971
   (S.Downarowicz, Hydroisohipses map of Quartenary layer of the region of “Gilow” reservoir of post-flotation sludge)

5. Ocena podstaw merytorycznych i formalno-prawnych nieodplatnej dostawy wody pitnej dla mieszkancow wsi Szklary Gorne i Szklary Dolne przez KGHM “Polska Miedz” S.A. Oddzial ZG “Lubin”, HydroGeoMetal s.c., Lubin 1996,
   (Assessment of substantial and formal-legal bases of free delivery of fresh water to the inhabitants of Szklary Gorne and Szklary Dolne villages, conducted by KGHM)

6. Ocena wplywu eksploatacji piaskowni “Obora” ponizej zwierciadla wody gruntowej i okreslenie niezbednych do spelnienia warunkow dla utrzymania rownowagi ekologicznej srodowiska, Okregowy Osrodek Rzeczoznawstwa i Doradztwa Rolniczego, Wroclaw 1987,
   (Assessment of the influence of “Obora” sandpit exploitation below the ground water level, and specification of necessary conditions to be met to preserve the ecological equilibrium of the environment, Regional Center of Expertise and Agricultural Consulting)

7. Dokumentacja geologiczna zloza materialu podsadzkowego Obora, Przedsiebiorstwo Geologiczne Wroclaw, 1973 r.
   (Geological documentation of the deposit of Obora backfilling material, Geological Company)

8. Studium kompleksowego uregulowania stosunkow wodnych zlewni potoku Zielenica, Przedsiebiorstwo Innowacyjno-Wdrozeniowe “GeoSoft”, Wroclaw 1996 r,
   (Study of complex control of water ratios of Zielenica stream basin, “GeoSoft” Innovative-Implementing Company)

9. Bilans wody wypompowanej ze studni z ujecia wody w Szklarach, KGHM Polska Miedz S.A., Lubin 1998 r.
   (Balance of water pumped out of the well in Szklary water intake, KGHM Polish Copper Stock Company)

10. Sprawozdania z obserwacji hydrogeologicznych na zlozu “Obora”, KGHM Polska Miedz S.A., Oddzial Zaklad Hydrotechniczny w Rudnej
    (Report on hydrogeological observations of “Obora” deposit, KGHM Polish Copper Stock Company)

11. M.Rojek, T.Wiercioch, Rozklad przestrzenny parowania wskaznikowego na terenie Polski w okresie 1951-1990, ZNAR 243, In.Srod. VI, 1994, str.23-35,
    (M.Rojek, T.Wiercioch, Spatial distribution of index evaporation in the area of Poland in the period of 1951-1990)

12. M.Rojek, T.Wiercioch, Indicatory Evaporation of Summer Half-Year in Various Long-Term Periods, Zesz Problemowe Postepow Nauk Rolniczych 1994, str.155-161,

13. A.Miler, Modelowanie Matematyczne Zdolnosci Retencyjnych Malych Zlewni Nizinnych, Zeszyt 258, Roczniki Akademii Rolniczej w Poznaniu, 1994
    (Mathematical Modeling of Retention Capacities of Small Lowland River Basins, Paper 258, Annual Publications of Agricultural University in Poznan, 1994)

14. M.Rojek, T.Wiercioch, Zmiennosc czasowa i przestrzenna parowania wskaznikowego, ewapotranspiracji aktualnej i niedoborow opadowych w Polsce nizinnej w okresie 1951-1990, ZNAR Wroclaw, nr 238, monografie VI, 1995,
    (M.Rojek, T.Wiercioch, Time and space variability of index evaporation, current evapotranspiration, and precipitation deficiencies in lowland Poland in the period of 1951-1990)

15. Mapa deformacji terenu gorniczego na 03.1996 i prognoza do konca eksploatacji – 2013, Przedsiebiorstwo Innowacyjno-Wdrozeniowe “GeoSoft”, 1996 r.
    (Map of mining ground deformation for 03.1996 and forecast till the end of exploitation – 2013, “GeoSoft” Innovative-Implementing Company, 1996)

16. Badania modelowe dynamiki i propagacji zanieczyszczen wod podziemnych i powierzchniowych wraz z projektami ochrony zlewni i ujec wod “Szklary” i “Rynarcice”, Przedsiebiorstwo Konsultingowe HydrGeoMetal, Lubin, 1995,
    (Model tests of dynamics and propagation of ground and surface water, and designs of protection of river basin and water intakes of “Szklary” and “Rynarcice”, HydrGeoMetal Consulting Company)

17. Analiza mozliwosci zwiekszenia zakresu eksploatacji piasku podsadzkowego spod wody w piaskowni “Obora”, Przedsiebiorstwo Konsultingowe HydrGeoMetal, Lubin, 1995,
    (Analysis of the possibility of increasing the range of backfilling sand exploitation from under water in “Obora” sandpit, HydrGeoMetal Consulting Company)

18. Ochrona wod podziemnych, Instytut Geologiczny, Warszawa 1984
    (Protection of ground water, Geological Institute)

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed ‘Discussions’ in each series and hyperlinked to the article.