Hydrodynamic Model of the Open-Pit Mine Buvač (Republic of Srpska)

Hydrodynamic Model of the Open-Pit Mine Buvač (Republic of Srpska) Dušan Polomčić, Dragoljub Bajić*, Petar Papić, Jana Stojković University of Belgrade, Faculty of Mining and Geology, Department of Hydrogeology e-mail: osljane@orion.rs Cite as: Polomcic, D., Bajic, D., Papic, P., Stojkovic, J., Hydrodynamic model of the open-pit mine Buvač (Republic of Srpska), J. sustain. dev. energy water environ. syst., 1(3), pp 260-271, 2013, DOI: http://dx.doi.org/10.13044/j.sdewes.2013.01.0019 ABSTRACT Projecting of the dewatering system of the open-pit mine Buvač (Republic of Srpska, Bosnia and Herzegovina) is based on the use of hydrodynamic model of groundwater regime. Creating the hydrodynamic model of the open-pit mine Buvač was made in phases, which began by basic interpretation of collected data, along with schematization of the groundwater flow and flow conditions, and finally, forming and calibration of model. Hydrodynamic model was created as multilayer model with eight layers. Calibration of the hydrodynamic model is the starting point for making prognosis calculation in order to create the most optimal system of open-pit mine protection from groundwater. The results of model calibration indicated that the rivers Gomjenica and Bistrica, precipitation and inflow from karstified rocks are the primary sources of recharge of the limonite ore body Buvač. KEYWORDS Limonite ore body, Aquifers, Model calibration, Groundwater balance INTRODUCTION The limonite ore body Buvač is part of the Omarska deposit that also includes Jezero and Mamuze ore bodies SE in Omarska field. The Buvač body is Carboniferous, of 20 metres average thickness and some 3 km 2 in surface area. Fig. 1 shows location of the research area. Extraction of limonite and associated sediments at Buvač deposit is being performed under complex hydrogeological conditions, thus in the final exploitation phase the open pit depth is going to be m. Lowering of mining level will basically disclose all water-bearing sediments, which must be previously drained, in order to provide level stability, as well as equipment and mining staff protection during the deposit exploitation. In mining operations carrying out below the water table, mine operators are potentially faced with two important water-related problems. These are the amount and pressure of groundwater that could flow into an open pit and the effect of pore water pressure on the stability of an open pit high wall. Many analytical solutions for prediction of water inflow into mine excavations can be found in the literature [1-3]. These models often were developed based on some very specific assumptions and boundary conditions that restrict their applicability in many mining situations. The prediction of the amount of water inflow into the open-pit mine is very important for development of a mine dewatering program. Moreover, taking into account that the analytical solutions are not as versatile as numerical methods, which can deal with complex mining situations, so, it is necessary to develop a numerical (hydrodynamic) model that Page 260

PTB1 BU-377 includes all aquifer conditions [4]. Numerical models have not the limitations of analytical solutions and they are suitable for the simulation of all aquifer conditions. Furthermore, numerical models can provide a more realistic representation of the interaction between groundwater systems and mine excavations. Researched area Banja Luka BOSNIA AND HERZEGOVINA Sarajevo Figure 1. Location map of the researched area 0 50 100 200 km Projecting of the dewatering system of the open-pit mine Buvač is based on the use of hydrodynamical model of groundwater regime. The results of numerical simulations would be used, in future, to develop an appropriate water management strategy in order to minimise the operational problems below the water surface and long-term environmental problems and solutions to ensure the stability of an open pit high wall [5, 6]. HYDROGEOLOGY CHARACTERISTICS Characteristics of aquifers, hydraulic and storage properties are deduced from more than 550 cored boreholes (Fig. 2). Hydrogeologic characterization of the Buvač open-pit mine are interpreted from all geological and hydrogeological investigation and laboratory test data. 4971300 4971050 4970800 4970550 4970300 BU-345 4970050 4969800 Krivaja Hgb7 BU-161 BU-425 BU-392 BU-426 Hgb8 Hgb6 BU-404 Om-1 BU-424 W BU-389 Hgb9 BU-393 BU-413 BU-408 BU-423 BU-398 Hgb10 Hgb5 BU-412 BU-405 BU-399 BU-427 BU-400 BU-414 BU-162 BU-403 BU-397 BU-431 BU-407 BU-411 BU-396 Hgb4 BU-395 BU-428 Hgb11 BU-420 BU-416 BU-394 PTB3 BU-391 BU-429 BU-433 BU-401 BU-419 BU-430 BU-163 BU-402 BU-381 BU-432 BU-417 BU-382 BU-434 BU-406 BU-378 BU-505 BU-375 BU-421 BU-377 BU-383 Hgb3 BU-379 BU-376 BU-388 BU-365 BU-369 BU-437 BU-38 Buva č Open-pit BU-673 BU-364 BU-370 BU-673A BU-380 BU-366 BU-506 BU-374 BU-303 BU-305A BU-302 BU-305 BU-358 BU-363 BU-387 BU-181 BU-359 BU-367 BU-436 BU-39 BU-37 BU-301 BU-304 BU-286 BU-306 BU-354 BU-512 BU-368 BU-180 BU-202 BU-104 BU-674 BU-51 BU-168 BU-346 BU-530 BU-373 BU-674A BU-564 BU-236 BU-521 BU-287 BU-170 BU-335 BU-507 BU-347 BU-515 BU-386 BU-500 BU-106 BU-33 BU-52 BU-672A BU-128 BU-111 BU-672 Sbb1 BU-167 BU-362 BU-440 BU-216 BU-249 BU-269 BU-334 BU-333 BU-531 BU-372 Hgb2 BU-422 BU-103 BU-510 BU-45 BU-533 BU-49 BU-172 BU-356 BU-385 BU-300 BU-248 BU-267 BU-516 BU-315 BU-307 BU-504 BU-36 BU-46 BU-534 BU-29 BU-116 BU-514 BU-19 BU-532 BU-361 BU-360 BU-371 BU-217 BU-100 BU-250 BU-509 BU-278 BU-310 BU-355 BU-438 BU-559 BU-53 BU-43 BU-275 BU-159 BU-75 BU-384 BU-83 BU-218 BU-101 BU-268 BU-353 BU-308 BU-439 BU-55 BU-524 BU-30 BU-598A BU- BU-23 BU-165 Go1 BU-110 BU-598 BU-553 BU-279 BU-298 BU-109 BU-316 BU-262 BU-56 BU-560 BU-48 BU-522 BU-98 BU-299 BU-520 BU-77 BU-91 BU-513 Hgb13 BU-124 BU-536 BU-74 BU-239 BU-309 BU-550 BU-31 BU-525 BU-155 BU-22 BU-69 BU-666 BU-169 BU-554 BU-318 BU-284 BU-296 BU-511 BU-220 BU-58 BU-561 BU-173 BU-140 BU-34 Om-5 BU-261 BU-260 BU-70 BU-547 BU-283 BU-297 BU-518 BU-219 BU-240 BU-253 BU-32 BU-273 BU-551 Sbb2 BU-21 BU-152 BU-137 BU-24 BU-166 BU-65 BU-555 BU-285 BU-90 BU-81 BU-80 BU-235 BU-562 BU-132 BU-136 BU-78 BU-15 BU-272 BU-271 BU-289 BU-221 BU-241 BU-252 BU-20 BU-13 BU-149 BU-552 BU-134 BU-201 BU-17 BU-71 Hgb1 BU-317 BU-556 BU-290 BU-282 BU-280 BU-311 BU-337 BU-543 BU-145 BU-147 BU-28 BU-96 BU-503 BU-314 Hgb14 BU-548 BU-264 BU-146 BU- BU-338 BU-12 BU-89 BU-120 BU-184 BU-288 BU-186 BU-291 BU-93 BU-234 BU-251 Go-2 BU-18 BU-200 BU-76 BU-295 BU-73 BU-263 BU-540 BU-243 BU-138 BU-223 BU-185 BU-212 BU-92 BU-293 BU-312 BU-27 BU-148 BU-544 BU-129 BU-99 BU-25 BU-294 BU-245 BU-549 BU-222 BU-281 BU-292BU-16 BU-224 BU-41 BU-259 BU-336 BU-537 BU-123 BU-527 BU-10 BU-79 BU-535 BU-313 BU-72 BU-255 BU-541 BU-242 BU-557 BU-85 BU-523 BU-87 BU-233 BU-171 BU-254 BU-545 BU-7 BU-102 BU-329 BU-244 BU-331 BU-130 BU-26 BU-86 BU-54 BU-50 BU-67 BU-529 BU-571 BU-213 BU-519 BU-210 BU-225 BU-572 BU-330 BU-538 BU-246 BU-528 BU-82 BU-579 BU-11 BU-601 BU-66 BU-256 BU-542 BU-257 BU-558 BU-214 BU-517 BU-276 BU-209 BU-88 BU-227 BU-105 BU-546 BU-247 BU-526 BU-57 BU-593 BU-60 BU-63 BU-68 BU-258 BU-266 BU-208 BU-602 BU-211 BU-226 Hgb15 BU-539 BU-8 BU-580 BU-47 BU-339 BU-215 BU-587 BU-228 BU-351 BU-611 BU-97 BU-341 BU-563 BU-59 BU-94 BU-95 BU-84 BU-265 BU-573 Sbb4BU-342 BU-9 BU-581 BU-14 BU-352 BU-108 BU-44 BU-270 BU-350 BU-61 BU-107 BU-274 Bistrica Gomjenica 6412500 6412750 6413000 6413250 6413500 6413750 6414000 N S E 0 250 500 750 m Figure 2. Test borehole sites in the Buvač limonite deposit area Page 261

4974000 4973000 t 2 4972000 Gomjenica al Bistrica 4971000 B A 4970000 t1 t Pl,Q 1 J M A B t1 Gomjenica 6408000 6409000 6410000 6411000 6412000 6413000 6414000 6400 6400 6417000 Alluvial sand and gravel Legend Intermittent streams First stream terrace of gravel and sand, locally argillically altered Second stream terrace of sand and clay Quaternary and Pliocene sand and clay M J Buva čor body contour Mamuze mine pit contour Jezero open-pit mine contour Perennial streams A A Profile line Figure 3. Hydrogeological map of the Buvač ore body A River Gomjenica BU-318 BU-98 BU-279 BU-101 BU-250 BU-267 BU-533 BU-269 BU-170 Stream Track Bistrica Road BU-375 BU-306 BU-358 BU-364 BU-375 BU-432 BU-433 A Stream Bistrica Embankment 140 120 100 80 60 40 20 0-20 B B 140 120 100 80 60 40 20 0-20 -40-60 Pliocene sand and clay Legend Carboniferous karstified limestone and dolomitic limestone, siderite and ankerite Quaternary clay Carboniferous limonite ore body Carboniferous argillically altered sandstone and compact calcite Carboniferous argillically altered siltstone Figure 4. Hydrogeological sections A-A' and B-B' 0 100 Basic hydrogeological characteristics of the ore body are frequent alternation of permeable and impermeable rocks in the vertical section and four types of aquifers: alluvial aquifer, aquifer in Pliocene sands, aquifer in limonite ore body, and confined karstic aquifer bellow the limonite ore body. The effects of hydraulic communication Page 262

between the aquifers in some areas are almost equal water head in each aquifer and uniform chemical composition of water, therefore marked three-dimensional movement of groundwater within the ore body [7]. Hydrogeological map (Fig. 3) shows the boundaries of Quaternary deposits and hydrogeological sections A-A' and B-B' (Fig. 4) show the relationship of the aquifers. HYDRODYNAMIC MODEL Groundwater flow in an unconfined aquifer may be approximately modeled by the nonlinear Boussinesq equation, assuming Dupuit s hypothesis of horizontal flow applies [8]. This equation in Cartesian coordinate system is shown below: K x x h K x y y h K y z z h W z S s h t (1) where x, y and z are coordinates of Cartesian coordinate system; K x, K y and K z are the hydraulic conductivity along x, y and z coordinates which are assumed to be parallel to the major axes of hydraulic conductivity in m/s; h is hydraulic head in m; W is unit precipitation (precipitation per unit of horizontal spreading of the flow), represents the effective intensity of the vertical recharge in m/s and S s is storage coefficient, -. The code selected to develop the numerical model was MODFLOW-2000; a modular, three-dimensional finite difference groundwater flow model developed by US Geological Survey [9]. The program used in this work is Groundwater Vistas 5.33b (Environmental Simulations International, Ltd.). Discretization of flow field A hydrodynamic model of the Buvač open-pit mine is designed to have eight layers in the vertical section. Each layer corresponds to a real stratum, schematized and delineated on the basis of observations and abundant field investigation data [10]. With respect to the natural strike and dip of the geologic units, the assigned extents of layers in plan and their respective thicknesses are different. Geometrization of the layer contours, their transposition into the coordinate system of the model, is based on abundant data from boreholes all over the research area (Fig. 2). The model layers, downward from the ground surface, are given in Tab. 1. The result of the schematized layers transposition into the model is represented by spatial distribution of the aquifer types in Fig. 5. South North 2 1 3 4 Legend: 1. Alluvial aquifer 2. Pliocene sand aquifer 3. Ore body aquifer 4. Karst aquifer Figure 5. Three-dimensional view (eastern) of the system of aquifers in Buvač ore body Page 263

Basic matrix dimensions for the research area are 2,000 m x 1,750 m (Fig. 8). Cell size of the flow field discretization in plan is 25 m x 25 m, not further reduced with regard to the number and quality of available data [11]. Hydraulic properties of porous rocks Hydraulic parameters for porous rocks are assigned as representative quantities to each discretization cell. Table 1 gives the initial values of the hydraulic parameters in the model [12]. Table 1. Initial values of hydraulic parameters Layer 1 2 3 4 5 6 7 8 Hydrologic function Lithostratigraphic unit Hidraulic conductivity (at x,y-axis) (m/s) Hidraulic conductivity (at z-axis) (m/s) Specific storage (1/m) Specific yield (-) Total porosity (-) Impermeable Clay overburden 8.50x10-7 1.00x10-6 1.00x10-6 0.001 0.037 0.45 and sandy gravel 3.80x10-4 1.00x10-4 2.25x1 0-5 Water-bearing Alluvial gravel 1.40x10-4 - 0.225 0.3 Aquiclude Clay and sandy 6.20x10-6 clay 2.00x10-5 1.00x10-6 5.00x1 0-5 0.06 0.3 Part of ore body 2.30x10-4 6.00x1-0 -5 0.06 0.27 Combines impermeable and permeable Largely aquitard Largely aquitard Combined permeable-im permeable Aquifer in ore body Combined permeable-im permeable Aquifer in carbonate rocks Clay and sandy 5.00x10-6 1.00x10-6 5.00x1 clay 0-5 0.06 0.3 Sand and gravel 1.00x10-4 1.40x10-4 1.00x10-4 2.25x1 0-5 0.23 0.35 Part of ore body 2.3x10-4 1.00x10-4 6.00x1 0-5 0.06 0.27 Clay and sandy 4.0x10-6 1.00x10-6 5.00x1 clay 0-5 0.06 0.3 Part of ore body 2.3x10-4 1.00x10-4 6.00x1 0-5 0.06 0.27 Argillic siltstone 1.00x10-6 1.00x10-6 6.30x1 and sandstone 0-5 0.06 0.25 Part of ore body 2.3x10-4 1.00x10-4 6.00x1 0-5 0.06 0.27 Limonite and fine 5.0x10-5 - limonite 2.40x10-4 1.00x10-6 5.00x1 0-5 0.06 0.3 Argillic siltstone 1.5x10-5 1.00x10-6 6.30x1 and sandstone 0-5 0.06 0.25 Limonite ore body 2.3x10-4 1.00x10-4 6.00x1 0-5 0.06 0.27 Limestone and dolomitic limestone, siderite and ankerite in argillic siltstone and siltstone Argillic siltstone and siltstone 3.52x10-4 - 4.70x10-4 2.50x10-4 6.30x1 0-5 0.06 0.35 1.00x10-6 1.00x10-6 6.30x1 0-5 0.06 0.25 Boundary conditions Boundary conditions used in the hydrodynamic model of the Buvač ore body are the following: river boundary, vertical balance and general head boundary (GHB). River boundary. Surface streams, primarily the rivers Gomjenica and Bistrica, are important for the groundwater flow system. For the influence of surface streams on groundwater flow, the monitoring data for a period of 2.5 years were collected and Page 264

analysed. The analysis used mean monthly levels in the Gomjenica; data for the Bistrica stage were fewer than for the Gomjenica or for precipitation. All this was used as initial data for calibration of the model. A result of the model calibration was quantification of the Bistrica influence on the groundwater flow. Figure 6 compares fluctuations of the measured level in the Gomjenica and water surface in alluvial gravels. Diagrams give monthly amounts of precipitation recorded at the Prijedor rain-gauging station also for 2.5 years. Influence of the Gomjenica on the groundwater flow is seen to exist in the alluvial aquifer, phase-shifted, delayed from the river stage fluctuation. The delay was considerable through the considered period, which suggested the likely high clogging of the riverbed (where it is cut in gravels). In contrast, groundwater in the alluvium reacts sooner to the changes caused by precipitation intensity. 157 KS Prijedor Gomjenica BU-102 BU-145 BU-152 BU-48 BU-101 BU-148 BU-149 BU-58 BU-103 BU-134 BU-140 500 450 400 Piesometric level (m). 155 350 300 250 200 100 50 Precipitation (mm). 153 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Time (month) Figure 6. Fluctuations in the Gomjenica stage, precipitation and water table in alluvium Figure 7 compares fluctuations of the measured level in the Gomjenica and water surface in limonite ore body. It shows fluctuations in the Gomjenica stage and in water tables within the limonite ore body. Groundwater surface oscillates with the fluctuation in the Gomjenica and senses effects of precipitation, which is evidence of the hydraulic communication between this aquifer and alluvial gravel. The flow between Gomjenica nad Bistrica and aquifers was calculated through the Modlow RIVER package, on the basis of the following equation: Q R = C R (h R d) (2) where Q R is the flow between river and the aquifer in m 3 /s. The value of flow is positive if it is directed into aquifer. In above equation, h R is simulated aquifer head along the river in m. River conductance C R is given by equation: K L b C (3) R d where K is hydraulic conductivity of river bottom sediments in m/s, L is length of river cell in m, b is width of river cell in m and d is thickness of river bottom sediments in m. Page 265

Piesometric level (m). 157 157 155 155 KS Prijedor Gomjenica BU-99 BU-90 BU-91 BU-109 BU-105 BU-65 BU-104 500 450 400 350 300 250 200 100 50 Precipitation (mm). 153 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Time (month) Figure 7. Fluctuations in the Gomjenica stage, precipitation and water table in limonite ore body Hydraulic effect of surface streams, the rivers Gomjenica and Bistrica in particular because their influence on groundwater is the greatest within the extent of Buvač ore body, was simulated in the model as the river boundary, a boundary condition depending on the river stage. In view of the riverbed elevations, or that rivers traverse aquifers and the overlying strata, the rivers as a boundary condition are assigned to second model layer the alluvial gravels and sand. The direction of water flow from the river to the aquifer depends on the altitudinal difference between the groundwater surface and the river stage. Either the river feeds the aquifer, when water flows from the river into the aquifer if water in the river is higher than the groundwater surface, or the river drains groundwater from the aquifer to the river. This boundary condition is assigned to the first model layer (Fig. 8). Vertical Balance. Total or vertical balance is an essential element of the groundwater budget. It is the effective, resultant infiltration, the amount of percolated precipitation, evaporation from groundwater surface and evapotranspiration. Besides, also very important are depth to groundwater, moisture and lithology of the overlying strata. Effective percolation is considerable in the study area at the conditions for which data were available, because groundwater levels are at small depths from ground surface. Average percolation is 5 l/s/km 2, or 14.29% of the amount of precipitation. Data used for the mathematical model were precipitation records from the Prijedor rain-gauging station, averaged out to monthly level. Like the river boundary condition, the influence of precipitation and evaporation expressed as effective percolation in real monthly amounts is assigned for the whole groundwater flow simulation period. General Head Boundary (GHB). General Head Boundary is similar in mathematical terms the river boundary. It is used to simulate the impact of external sources of groundwater located outside the flow field covered by mathematical model. Flow rate to or from the observed model s cell is calculated by the formula: QGHB C GHB(H GHB h) (4) 0 Page 266

BU-328 BU-174 BU-111 BU-187 BU-20 Hgb1 BU-171 BU-501 BU-179 BU-175 BU-321 BU- BU-338 BU-323 BU-178 BU-153 PTB1 Hgb2 BU-329 BU-151 BU-110 BU- BU-336 BU-177 BU-144 BU-263 BU-337 BU-157 BU-143 BU-330 BU-117 Hgb3 BU-255 BU-39 BU-6 BU-182 BU-121 BU-83 BU-32 BU-148 BU-300 BU-264 Sbb3 BU-118 BU-125 BU-318 BU-537 BU-256 BU-106 BU-56 BU-62 BU-547 BU-149 BU-540 BU-254 BU-301 BU-218 BU-119 BU-126 BU-550 BU-273 BU-543 BU-245 BU-538 BU-303 BU-339 BU-122 BU-564 BU-36 BU-553 BU-58 BU-272 BU-12 BU-541 BU-105 BU-84 BU-216 BU-559 BU-1 BU-98 BU-551 BU-317 BU-544 BU-244 BU-539 BU-64 BU-180 BU-133 BU-55 BU-554 BU-65 BU-123 BU-542 BU-302 BU-217 BU-560 BU-548 BU-283 BU-552 BU-243 BU-545 BU-127 BU-258 BU-38 BU-103 BU-42 BU-31 BU-555 BU-145 BU-549 BU-7 BU-246 BU-236 BU-279 BU- BU-561 BU-271 BU-89 BU-242 BU-546 BU-139 BU-181 BU-53 BU-152 BU-556 BU-202 BU-248 BU-529 BU-193 BU-524 BU-284 BU-562 BU-146 BU-27 BU-527 BU-257 BU-33 BU-100 BU-48 BU-340 BU-13 BU-138 BU-102 BU-304 BU-9 BU-510 BU-101 BU-525 BU-285 BU-207 BU-147 BU-222 BU-528 Om-1 BU-521 BU-46 BU-173 BU-120 BU-557 BU-305A BU-249 BU-247 BU-299 BU-290 BU-129 BU-331 Hgb4 BU-104 BU-30 BU-132 BU- BU-10 BU-558 BU-305 BU-250 BU-341 BU-522 BU-297 BU-223 BU-130 BU-266 BU-45 BU-155 BU-28 BU-82 BU-287 BU-534 BU-298 Sbb2 BU-289 BU-184 BU-85 BU-526 BU-8 BU-265 BU-37 BU-43 BU-520 BU-99 BU-571 BU-674 BU-500 BU-267 BU-598A BU-21 BU-296 BU-288 BU-214 BU-563 BU-674A BU-52 BU-509 BU-598 BU-77 Hgb5 BU-518 BU-134 BU-185 BU-26 BU-572 BU-286 BU-533 BU-268 BU-44 BU-90 BU-281 BU-579 BU-215 BU-29 BU-109 PTB3 BU-140 BU-18 BU-523 BU-57 BU-573 BU-274 BU-269 BU-275 BU-91 BU-282 BU-673 BU-51 BU-672 BU-516 BU- BU-136 BU-186 BU-580 BU-270 BU-511 BU-276 BU-212 BU-79 BU-673A BU-672A BU-278 BU-513 BU-219 BU-59 BU-201 BU-407 BU-86 BU-593 BU-208 BU-49 Sbb4 BU-581 BU-22 BU-81 BU-96 BU-587 BU-61 BU-221 BU-200 BU-87 BU-342 BU-25 BU-47 BU-291 BU-535 BU-17 BU-92 BU-54 BU-14 BU-292 BU-228 BU-76 BU-280 BU-213 BU-519 BU-210 BU-11 Hgb6 BU-406 Sbb1 BU-170 BU-316 BU-393 BU-401 BU-159 BU-137 BU-128 BU-315 BU-220 BU-400 BU-421 BU-209 BU-124 BU-423 BU-402 BU-306 BU-353 BU-424 BU-80 BU-601 BU-420 BU-380 BU-425 BU-116 BU-34 BU-403 BU-379 BU-334 BU-60 BU-239 BU-412 BU-417 BU-602 BU-168 BU-23 BU-16 Hgb7 BU-389 BU-391 BU-93 BU-88 BU-94 BU-357 BU-536 Om-5 BU-78 BU-426 BU-358 BU-310 BU-240 BU-411 BU-293 BU-50 BU-365 BU-211 BU-352 BU-172 BU-69 BU-503 BU-224 BU-399 BU-351 BU-378 BU-335 BU-262 BU-241 BU-413 BU-514 BU-419 BU-359 BU-24 BU-63 BU-404 BU-395 BU-75 BU-294 BU-225 BU-364 BU-611 BU-350 BU-307 BU-261 BU-234 BU-414 BU-377 BU-422 BU-517 BU-95 BU-167 BU-398 BU-74 BU-381 BU-71 BU-354 BU-226 BU-308 BU-235 BU-416 BU-41 BU-505 BU-366 BU-66 BU-107 BU-392 Hgb8 BU-507 BU-19 BU-70 BU-295 BU-233 BU-397 BU-376 BU-333 BU-309 BU-311 BU-382 BU-97 BU-346 BU-504 BU-165 BU-666 BU-15 BU-251 BU-313 BU-429 BU-363 BU-355 BU-227 BU-253 BU-312 BU-396 BU-369 BU-67 BU-108 BU-512 BU-356 BU-169 BU-314 BU-427 BU-375 BU-506 BU-259 BU-347 BU-532 BU-408 BU-430 BU-252 BU-367 BU-73 BU-531 BU-384 BU-166 BU-68 Hgb9 BU-428 BU-370 BU-530 BU-361 BU-161 BU-162 BU-260 BU-383 BU-72 BU-362 BU-432 BU-368 BU-439 Hgb15 Hgb10 BU-394 BU-374 BU-515 BU-371 BU-431 BU-388 BU-372 BU-405 BU-434 BU-373 BU-433 BU-387 Go1 BU-438 BU-437 BU-385 Hgb14 BU-386 BU-163 BU-436 BU-360 Go-2 BU-440 Hgb13 Hgb11 Hgb12 where Q GHB is flow rate to or from model s cell in m 3 /s; H GHB is hydraulic head in model s cell in m; h is calculated value of hydraulic head in m and C GHB is conductance of model s cell m 2 /s. The influence of karstified limestone beneath the ore body is assigned to the model through this boundary condition, and this only in the eighth layer of karstified rocks in the north where they are actually replenished, and in the second model layer with alluvial gravel and sands in the east (Fig. 8). 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 2 5 10 15 20 25 30 35 40 45 50 55 60 65 1 1 3 70 Figure 8. River boundary and general load boundary conditions of the model Legend: 1. River boundary condition for alluvial gravel and sand; 2. GHB for carbonate rocks; 3. GHB for alluvial gravel and sand MODEL CALIBRATION The model was calibrated under the condition of unsteady flow with the one-month time step through 2.5 years. Groundwater flow was calculated and simulated real flow, under pressure, or free water table, individually in each discretization field, with the model flow conditions simulating the real conditions. For verification of results during the model calibration, mainly the water table records were used. Figure 9 shows sites of the observation wells in which groundwater levels were measured. Initial values of natural rock characteristics (permeability and specific storage or effective porosity) and characteristics of the cell to which the river boundary condition was attributed were modified during the calibration process. Figure 9 shows also the pattern of groundwater levels for the ore body, resulting from the model calibration. The agreement of registered and calibrated groundwater levels was fairly good (±0.2 m). For the purpose of a more detailed analysis, Fig. 10 shows recorded and calculated groundwater levels during the simulation for wells in the alluvial aquifer, and Fig. 11 for aquifer in the limonite ore body. Figure 10 show a generally good accordance of the mathematical simulation results Page 267

with the measured groundwater levels for the investigated area. However, there are a few deviations, in locations of BU-48, BU-58 and BU-148, where the calculated hydrographs are higher than those measured by 30 cm at the most. A better agreement could not be achieved with the given quality of the available data. Any reduction of the discrepancy between the calculated and measured groundwater levels would be associated only with local changes in the hydraulic properties and would not affect the flow pattern in the investigated area. Krivaja BU-104 Bistrica BU-103 BU-101 BU-109 BU- BU-48 BU-91 BU-58 BU-140 BU-152 BU-65 BU-90 BU-81 BU-80 BU-149 BU-134 BU-145 BU-148 BU-89 BU-092 BU-99 BU-102 BU-86 BU-105 Gomjenica.1.0 155.9 155.8 155.7.2 Legend BU-102 Observation wells in alluvial gravel and sand Registered groundwater levels BU-105 Observation wells in limonite ore body Calculated groundwater levels Figure 9. Map of observation wells and groundwater levels in the end of the model calibration period The presented groundwater surface hydrographs in the limonite body (Fig. 11) show good correspondence of the calculated and registered stages. Piesometric Head (m). 152 BU-48 Calc BU-58 Calc BU-80 Calc BU-48 Obs BU-58 Obs BU-80 Obs 0 200 400 600 800 1000 Time (day) Piesometric Head (m). 152 BU-81 Calc BU-86 Calc BU-89 Calc BU-81 Obs BU-86 Obs BU-89 Obs 0 200 400 600 800 1000 Time (day) Piesometric Head (m). 152 BU-103 Calc BU-134 Calc BU-140 Calc BU-103 Obs BU-134 Obs BU-140 Obs 0 200 400 600 800 1000 Time (day) Piesometric Head (m). 152 BU-145 Calc BU-148 Calc BU-149 Calc BU-145 Obs BU-148 Obs BU-149 Obs 0 200 400 600 800 1000 Time (day) Figure 10. Measured and calculated well hydrographs for alluvial aquifer Page 268

Piesometric Head (m). 152 BU-65 Calc BU-90 Calc BU-91 Calc BU-99 Calc BU-65 Obs BU-90 Obs BU-91 Obs BU-99 Obs 0 200 400 600 800 1000 Time (day) Piesometric Head (m). 152 BU-104 Calc BU-105 Calc BU-109 Calc BU-104 Obs BU-105 Obs BU-109 Obs 0 200 400 600 800 1000 Time (day) Figure 11. Measured and calculated well hydrographs for limonite body GROUNDWATER BALANCE A result of model calibration is the quantification of groundwater balance elements (Tab. 2). The groundwater balance in the end of the period used for calibration led to the conclusion that the alluvial aquifer is recharged from the rivers Bistrica and Gomjenica, from precipitation and in part from subsurface inflow. The karstified limestone aquifer is subsurface recharged from the north, and the limonite ore body includes an underground reservoir naturally fed from water in carbonate rocks and partly from alluvial aquifers. Alluvial groundwater partly drains to the rivers and partly through the SW model boundary and seepage to ore body Mamuze. Groundwater in carbonate rocks drains partly into the limonite ore body. Table 2 gives main elements of the groundwater balance in the period considered by calibration. Table 2. Groundwater balance elements for the area of the Buvač mineral deposit Boundary condition Model inflow, l/s Model outflow (drainage), l/s Gomjenica River 8.32 10.52 Bistrica River 6.92 7.58 Subsurface inflow from north 2.26 - (alluvial deposits) Subsurface inflow from north 4.38 - (carbonate rocks) Subsurface inflow from east 2.57 - (alluvial deposits) Subsurface inflow from south 3.22 (alluvial deposits) Subsurface outflow to west - 8.48 (alluvial deposits) Subsurface outflow to south 4.81 in ore body Mamuze Percolation from precipitation 4.22 - Total 31.89 31.39 CONCLUSION The spatial pattern of groundwater flow is related to the vertical lithological stratification and the variable horizontal extent of geologic units. This is particularly true of the ore body and alluvial sediments and of the ore body and limestone contact areas. This was the reason why we developed an eight-layer model. Verification of the model calibration results used well records of the groundwater tables. Mathematical simulation results agreed fairly well with the water table records. Page 269

The results of the model calibration are quantification of the Bistrica and Gomjenica influence on the groundwater flow and determining the groundwater balance elements for the Buvač ore body. As the result of the model calibration, it has been quantified that the karstified limestone aquifer is subsurface recharged from the north (4.38 l/s). The account of percolation from precipitation is 4.22 l/s, water infiltration (summary) of the Bistrica and Gomjenica rivers is 15.24 l/s. It has also been confirmed that aquifers are drained by subsurface outflow from west (8.48 l/s) and south (4.81 l/s). Total model inflow is 31.89 l/s, and model outflow is 31.39 l/s. The hydrodynamic model so calibrated is an initial tool in forecasting groundwater behaviour for an optimum protection of the mine pit. ACKNOWLEDGMENTS The author is very grateful to the Ministry of Education and Science of Republic of Serbia for financing projects 176022, III-43004 and 33039. REFERENCES: 1. Hanna, T.M., Azrag, E.A., Atkinson, L.C., Use of an analytical solution for preliminary estimates of groundwater inflow to a pit. Mining Engineering 46(2), pp.149-152, 1994. 2. Marinelli, F., Niccoli, W.L., Simple analytical equations for estimating ground water inflow to a mine pit. Ground Water 38 (2), pp.311-314, 2000., http://dx.doi.org/10.1111/j.1745-6584.2000.tb00342.x 3. Shevenell, L., Analytical method for predicting filling rates of mining pit lakes: example from the Getchell Mine, Nevada. Mining Engineering 52(3), pp.53-60, 2000. 4. Rapantova, N., Grmela, A., Vojtek, A., Halir, J., Michalek, B., Ground Water Flow Modelling Applications in Mining Hydrogeology. Mine Water and the Environment, Vol. 26, Iss. 4, pp.264 270, 2007., http://dx.doi.org/10.1007/s10230-007-0017-1 5. Daozhong, C., Quingli, Z., Jie, W., Xiaozhi, Z.: Comparative Analysis of Ecological Rucksack Between Open-pit and Underground Coal Mine. Mining Science and Technology, Vol. 20, Iss. 2, pp.266-270, 2010. 6. Adams, R., Younger, P.L., A strategy for modelling ground waterrebound in abandoned deep mine systems. Ground Water 39(2), pp.249 261, 2001., http://dx.doi.org/10.1111/j.1745-6584.2001.tb02306.x 7. Bajic, D., Polomcic, D, 3D Hydrogeological model of the limonite ore body Buvac (Republika Srpska, Bosnia and Herzegovina). IV International Conference Coal 2008, Belgrade, pp.1-9, 2008. 8. Marsily, G., Quantitative Hy-drogeology. Groundwater Hydro-logy for Engineers. Academic Press, Inc.San Diego, CA, USA, 1986. 9. Harbaugh, A.W., Banta, E.R., Hill, M.C. and McDonald, M.G., MODFLOW-2000: The U.S. Geological Survey Modular Ground-Water Model. User Guide to Modularization Concepts and the Ground-Water Flow Process, U.S. Geological Survey Open-File Report 00-92, Reston, VA, USA, pp.121, 2000. 10. Polomcic, D., Schematization types of hydrogeological system for setting up the hydrodynamic model. Proceedings of the XIII Yugoslav symposium of hydrogeology and engineering geology, Herceg Novi, pp.389-396, 2002. 11. Polomcic D., The influence of size of discretization of space and time on the accuracy of results in the conception of the hydrodynamic model, Papers of Geoinstitut, Belgrade, pp.197-209, 2004. Page 270

12. Polomcic, D., Dragisic,, V. and Zivanovic, V., Hydrodynamic model of the groundwater source for water supply of Prijedor town (Republic of Srpska). Int. Conference & Field Seminar - Water Resources & Environmental Problems in Karst, CVIJIĆ, Beograd, pp.539-544, 2005. Paper submitted: 20.02.2013 Paper revised: 03.04.2013 Paper accepted: 18.04.2013 Page 271