Environ Earth Sci DOI 10.1007/s12665-015-4132-1

ORIGINAL ARTICLE

Quantitative evaluation and prediction of water inrush vulnerability from aquifers overlying coal seams in Donghuantuo Coal Mine, China Qiang Wu • Yuanzhang Liu • Lihong Luo Shouqiang Liu • Wenjie Sun • Yifan Zeng

•

Received: 16 June 2014 / Accepted: 30 January 2015 Ó Springer-Verlag Berlin Heidelberg 2015

Abstract The water inrush vulnerability from overlying aquifers is evaluated for a coal seam in the Donghuantuo Coal Mine, China. The water inrush vulnerability is based on superposition of two zoning maps—water abundance map and connectivity map. The water abundance zoning map was derived from a comprehensive analysis of six types of geoscience data, including (1) lithological changes of the overlying aquifer (2) geological structure, (3) pumping test results, (4) seepage field of water inrush event, (5) geochemistry, and (6) loss of drilling mud. The connectivity map was constructed based on comparing the numerically calculated height of the induced fracture zone above mining areas with the thickness of formation between the coal seam and the overlying aquifer. Visual Modflow was used to predict the mine discharges at working face 2,284 under both natural and mining conditions. Based on the water inrush vulnerability map and the modeling results, advanced dewatering from the overlying

aquifer is proposed as the main prevention and control measure against the overlying-aquifer water inrush. Keywords Overlying-aquifer water inrush  Mining  Vulnerability  China

Study purpose and significances With the increase of mining depth and the continuing extension to deeper parts of coal mines in China, the hydrogeological conditions will become increasingly complicated. The threat of water hazards to mining is further aggravated due to water inrush from overlying aquifers and is a major safety problem that threatens the production of coal (Wu and Jin 1995; CGBC 2000; Wu 2001; SACMS 2009). Safe production of coal and long-term development planning of the mine require a scientific evaluation of the risk of overlying-aquifer water inrush potential and a reasonably accurate forecast of the expected groundwater discharge.

Introduction to study area Q. Wu  S. Liu  W. Sun  Y. Zeng China University of Mining and Technology, Beijing 100083, China e-mail: [email protected] Y. Liu (&) Beijing Institute of Hydrogeology and Engineering Geology, Beijing 100195, China e-mail: [email protected] L. Luo Tianjin Institute of Geo-Environment Monitoring, Tianjin 300191, China e-mail: [email protected]

Donghuantuo Coal Mine is located in Tangshan of Hebei Province, China. It is approximately 180 km from Beijing and the area of the mine field encompasses 52 km2. The mining area is on the fluvial plain and is covered with Quaternary sediments. The area has a temperate continental climate with a cold winter and hot summer, and the monthly average temperature ranges from 17.1 to 23.3 °C. The average annual precipitation is approximately 614.7 mm, and the average annual evaporation is 1321.1 mm. There is no surface water system within the coalfield.

123

Environ Earth Sci

Geological conditions

1.

The coal-bearing strata are Permian and the mine belongs to North-China-type coalfields, and the strata of the mine are the result of a normative sedimentary sequence of the North-China-type coalfields (Fig. 1). Donghuantuo Coal Mine is on the Chezhoushan syncline with an axial direction of NE 608 (Fig. 1). The syncline is long, narrow and unsymmetrical and it dips toward the northwest. The stratigraphic attitudes of the syncline’s two wings vary greatly. The stratum of the southeast wing is flat, while the stratum of the northwest wing is steeply inclined. Faults associated with the syncline are well developed and strike approximately in the same direction as the axial direction of the syncline, as shown in Fig. 2. No magmatic intrusion and karst collapse columns were observed within the mine area. Hydrogeological conditions Aquifers There are two major aquifer systems that are related to the roof water inrush in Donghuantuo Coal Mine: the lower Quaternary alluvium porous medium aquifer and Permocarboniferous fractured sandstone aquifer overlying coal seam #5. The former is the indirect recharge source, and the later is the direct recharge source of the overlyingaquifer water inrush for coal seam #8. There is a layer of aquitard of bauxite mudstone between the two aquifers.

Fig. 1 Structures of northern mining area of Donghuantuo Coal Mine

123

2.

Quaternary alluvium aquifer The Quaternary loose alluvium covers the entire region and overlies unconformably on the Paleozoic strata. It thins gradually from south to north. The thickness in the north part is less than 185 m, whereas the thickness is approximately 650 m in the south part. The loose alluvium can be divided into two sections according to the deposition characteristics. The upper section is 120–180 m thick and is composed of gravel layers with different sizes interbedded with sand and clay layers. The sand and gravel layers are water-bearing zones, while the clay layers are good aquitards. The lower part is a thick layer of gravel and pebble mixed with sand and sandy gravel layers in some parts. The pebbles are poorly sorted with diameters ranging from 20 to 40 mm. The layer is mainly recharged by the precipitation and has strong water abundance with the unit inflow varying from 0.275 to 2.258 l/s m. The present water level in the aquifer is between -15 and 18.5 m above mean sea level. Fractured rock aquifer in the Permo-carboniferous sandstone There are thick layers of medium and coarse sandstone with argillaceous or siliceous cementation in the Permo-carboniferous strata. Because of the different lithology, degrees of development of fracture structure in the sandstone are different. The sandstone layer is interbedded with coal seam, claystone, tuffite and silty mudstone, in which the fracture structure is poorly developed.

Environ Earth Sci Fig. 2 Schematic diagram of geological profile A–A0

Aquitards 1.

2.

Claystone There are between 7 and 13 layers of claystone in the claystone of upper Shihezi Formation (P2). Their thickness varies from 2 to 8 m with interlayer spacing of between 4 and 20 m. Pumping tests indicate that the claystone layer has very low permeability. Tuffite, mudstone and clayey sandstone The tuffite, mudstone and clayey sandstone, located between coal seam #5 and coal seam # (P1), tend to swell when in contact with water. As a result, any cracks can be closed easily. They are aquitards with very weak permeability and thickness ranges from 2 to 28 m. Together with the coal seams they form the aquitard group. The interbedding structure of the aquifers and aquitards exhibits a blocking effect for the vertical flow and leads to a layering feature of the aquifers.

The water flow in the Permo-carboniferous fractured sandstone aquifer is mainly controlled by two factors: form of the syncline in the coalfield, and topography of bedrock. Water abundance zoning map of aquifer overlying coal seam #8 This study selected coal seam #8 in North 2 mining area of Donghuantuo Coal Mine as the evaluation and analysis objective. The western boundary of North 2 mining area is the axis of Chezhoushan syncline and the southern boundary fault F2. The rest of the mining area is bounded by the subcrops. Evaluation of the water abundance relies on various data. All the exploration data and observation data were collected. Then multi-source geo-information is then integrated with the function of superposition of geographic information system (GIS) (Chen 1999; Tanabe et al. 2007; Morillas et al. 2007).

Characteristics of recharge, runoff and discharge in the coalfield

Gravel and pebble aquifer in the lower part of quaternary sediments

The rainfall recharges directly the upper Quaternary porous medium aquifer of sand and gravel layers, which subsequently recharges the lower Quaternary gravel and pebble aquifer and the fractured sandstone aquifer. The rainfall could also directly recharge the karst and fractured rock aquifer at outcrops of the Ordovician limestone. Long-term observations show that the water discharge at the Donghuantuo Mine is insensitive to local rainfalls. Therefore, the recharge from the lower gravel and pebble aquifer to the karst and fissure aquifer is relatively stable, and the seasonal changes are not obvious.

The lower quaternary porous medium aquifer is the indirect recharging source of the roof water inrush to coal seam #8. The aquifer is at the bottom of the quaternary alluvium, and its thickness becomes gradually smaller from northeast to southwest in the North 2 mining area. Based on the pumping test results in the aquifer, the hydraulic conductivity decreases from northwest to southeast from the synclinal axis section to southeast. The feature is related to the changing environments of the alluvial deposits. In general, the water yield of the aquifer increases from northeast to southwest and reaches to the maximum

123

Environ Earth Sci

2.

3.

Fig. 3 Distribution of specific capacity (L/s m) of lower quaternary aquifer

amount in the southwestern region of the study area (Fig. 3). The above analysis indicates that both the hydraulic conductivity and water yield of the aquifer reach to the maximum amount in the southwest part of the area, which means that water abundance of the aquifer is the biggest in this part of the study area.

4.

value, which is defined as the thickness ratio of brittle rocks and plastic rocks in the roof layers, decreases gradually roughly from north to south. The m value in the study area is between 10 and 90. Geologic structures: the stratum of the southeast wing is flat with bedding angles of 12°–25°, while the stratum of the northwest wing is steeply inclined with bedding angles of 65°–80°. The structure is more complicated in the northwest wing. Underground construction, drilling, and seismic exploration revealed 25 faults of different sizes in the study area. The strike direction is mainly NE and NW directions. This means the main direction of the hydraulic conductivity tensor is NE and NW directions. Pumping test: single-well pumping test data revealed that the specific capacity of the fractured sandstone increases from northeast to southwest, which suggests that the water abundance in the southwest is stronger than that in the northeast. The hydraulic conductivity increases gradually from north to south in the study area. The zones with stronger water abundance and stronger seepage are in the southern part of the study area, where the fractured sandstone aquifer overlying coal seam #5 is buried (Fig. 4). Features of seepage field of water inrush events and data analysis of long-term monitoring wells: according to systematic data analysis of three water inrush events from working face 2,088, upper 2,182 and lower 2,182,

Fractured sandstone aquifer overlying coal seam #5 The fractured sandstone aquifer is the direct recharge source to the overlying-aquifer water inrush of coal seam #8. The characteristics of the aquifer can be described from six aspects consisting of (1) lithologic changes, (2) geologic structure, (3) pumping test, (4) seepage field of water inrush events, (5) geochemistry, and (6) loss of drilling mud. The water abundance zoning map of the aquifer overlying coal seam #5 can be constructed from analysis of the six factors (Wu et al. 2007, 2008). 1.

Feature of lithologic facies changes: lithologic facies changes include three aspects, aquifer thickness and proportions of the brittle rock (mainly siltstone and sandstone) and plastic rock (mainly mud and shale). The thickness of the aquifer increases gradually from northeast to southwest. The fractured sandstone aquifer overlying coal seam #5 is mainly composed of sandstone, siltstone and shale. The thickness of the brittle rock is much larger than that of the plastic rock. In the study area, the m

123

Fig. 4 Distribution of specific capacity (L/s m) in fractured sandstone aquifer

Environ Earth Sci

5.

there exists a separate water system for the fractured sandstone aquifer in the North 2 mining area. Figure 5 shows the groundwater level contours of the fractured sandstone aquifer overlying coal seam #5 based on the observations in September 2007. There is a difference of 170 m in the groundwater level between two monitoring wells, DG27 in the North 2 mining area and DG21 near F2. The two monitoring wells are very close to each other. The level difference forms a clear ‘‘scarp’’ in the contours map. The scarp has been stable during the mining course. This shows that the strata have been split into two different hydrogeological units by F2 fault. In south of F2, dewatering has been carried out in each aquifer of the footwall, while in the hanging wall north to the F2 the aquifer still maintains a natural state with high groundwater level. Geochemistry field: the water quality analysis could help to divide different groundwater flow systems. The results indicate that there is an obvious difference in the chemical characteristics of the water from the

6.

aquifer in the North 2 mining area, and it may be taken as a separate flow system. Loss of drilling mud: based on the exploration drilling data, significant loss of drilling mud occurred when the drilling reached the aquifer for almost all of the boreholes in the area. The loss amount increases from northwest to southeast gradually.

The comprehensive analysis of the geoscience information of each physical field and the mutual comparison of the multi-source geo-information led to production of the thematic map for each main controlling factor. Complex superposition of the thematic maps was then applied with the use of GIS function of spatial overlay operation. The zoning map of the water abundance of the aquifer was then constructed based on the thematic maps (Zhang et al. 1994; Shang et al. 2006); (Fig. 6).

Connectivity zoning map for coal seam #8 According to the theory of ‘‘the upper three zones’’, the fracture zone induced by mining is the basis of the roof water inrush (Liu 1981). In most of the coal mines of China, the height of the mining-induced fracture zone is usually estimated with the empirical formula introduced in Codes of Mine Hydrogeology (for trial implementation) (MCIC 1984). However, only the compressive strength is considered in the empirical formula, while some other important mechanical properties of the rocks, such as cohesion, Poisson’s ratio, internal friction angle, and tensile strength are not taken into consideration (Wu et al. 2002, 2011). Also the lithologic association and the spatial distribution of the overlying strata cannot be well processed by the empirical formula, and there would be large errors in the practical application. Numerical simulation can provide a more realistic result of the height of the breaking belt. Simulation of all of the study area is unrealistic because the size of the mining area and the amount of subdivision meshes are too large. So the paper predicted the height of the fractured water suture zone with numerical simulations at sites with more data, and compared the results with those from the empirical formula. This comparison resulted in a proportional coefficient, which was then used to revise the empirical formula. Then the revised empirical formula can be applied to the entire mining area. Calculation of height of the fractured water suture zone

Fig. 5 Groundwater level contours in fractured sandstone aquifer (m)

Simulation results show that the height of the roof breaking belt caused by mining of coal seam #8 is about 58 m (Nukala et al. 2007; Coulthard 1999). The height is 73.6 m from the traditional empirical formula. The proportional coefficient between these two results is approximately

123

Environ Earth Sci Fig. 6 Zoning map of water abundance of the fractured sandstone aquifer overlying coal seam #5

1.269. The empirical formula is then revised with the coefficient, and the revised empirical formula is as follows: Hfm ¼

100M ; 1:269 ð5:1n þ 5:2Þ

where M is the cumulative mining thickness and n is the number of the coal layers. Connectivity map of roof breaking belt for coal seam #8 If the height of roof breaking belt is less than the thickness of the overlying strata between coal seams #8 and #5, the overlying fractured water of coal seam #5 cannot flow into the mining roadway. On the contrary, a water inrush occurs. The boundary of the safe and unsafe zone can be determined by this criterion. Figure 7 shows the zoning map of roof breaking belt for coal seam #8. Because the heights of roof breaking belt of coal seam #8 are all greater than the thickness of the overlying strata between coal seams #8 and #5 in the North 2 mining area, all of the mining area of coal seam #8 is in the unsafe zone.

Comprehensive vulnerability zoning map of overlyingaquifer water inrush for coal seam #8 There are two necessary and sufficient conditions for the water inrush: one is the connectivity to the overlying

123

Fig. 7 Connectivity map of roof breaking belt for coal seam #8

aquifer by mining and the other is that the water abundance of the roof aquifer at the site of the mining face is stronger. The comprehensive vulnerability map of overlyingaquifer water inrush for coal seam #8 was obtained by complex superimposition of the water abundance zoning

Environ Earth Sci

map and the connectivity map of breaking belt. This comprehensive vulnerability map has integrated all of the geological information. The study area is divided into six zones and they are safe zone, less safe zone, transition zone, less dangerous zone, dangerous zone, and the most dangerous zone, as shown in Fig. 8. According to the preliminary design of the layout plan of working faces, more than half of the North 2 mining area is in the transition zone, less dangerous zone, and dangerous zone. If the full-height-mining method is used as designed, the mine-induced fracture zone can develop to the overlying fractured sandstone aquifer. Also these areas are characterized with stronger water abundance, bigger loss of the drilling mud (approximately 3 m3/h) and greater specific capacity of 0.25 m3/min. There is no impermeable layer between the buried fractured sandstone aquifer and the overlying lower quaternary porous medium aquifer of gravel and pebble. A hydraulic connection between these aquifers is likely. Part of working face 2,282 in the North 2 mining area is located in the affected zone of the porous medium aquifer. Prediction of mine discharge into working faces for coal seam #8 Prediction of water inflow under natural state The hydrogeological conceptual model for prediction of mine water discharge of coal seam #8 can be established based on the information from the physical fields. The

conceptual model includes two aquifers and one aquitard. The two aquifers are the lower Quaternary alluvium porous medium aquifer and the fractured sandstone aquifer overlying coal seam #5. The aquitard consists of the claystone, tuffite, mudstone and clayey sandstone layers with poor permeability. Eastern and western boundaries of the model are recharge and drainage boundary. Fault F2 in the southern boundary is a water-blocking fault, so it is taken as a zero-flow boundary. Northern boundary is recharge boundary. Permeability coefficient and initial storage rate of grids are assigned according to pumping test results of hydrological holes. Three-dimensional numerical simulation model software Visual Modflow (Guiguer 2004) was used to predict the groundwater discharge. The model was calibrated with the water level data from January 2005 to December 2006 of the DG27 and DG39 monitoring wells. Working face 2,284 in North 2 mining area was selected as the target of simulation because this working face is in the predicted dangerous zone and the breaking belt caused by mining is predicted to reach the overlying fractured sandstone aquifer. Based on the law of the stress cycle of adjacent working face, dynamic prediction of the mine discharge with the advancing of working face 2,284 (take the periodic weighting step as the unit) was performed. The process is as follows: 1.

Layout pumping wells evenly in the first weighting step, and pump till the level reaches to the floor of the roof sandstone layer of coal seam #5;

Fig. 8 Comprehensive vulnerability map of overlyingaquifer water inrush for coal seam #8

123

Environ Earth Sci Fig. 9 Prediction of water inflow into working face 2,284 under natural state

Fig. 10 Prediction of mine discharge into working face 2,284 under mining conditions

2.

3.

Set the balanced field with the Zone Budge module, then run the Visual Modflow and get the mine discharge of the first weighting step; The discharge of the second step is the roof water inrush amount when the working face advances to the second step. At this time, the roof plate of the first step has been damaged, so set pumping wells evenly in the first and second step then the mine discharge of the second step is predicted. The prediction results are shown in Fig. 9.

Prediction of mine discharge under mining conditions Under the mining condition, dewatering holes will be set at -500 m level near F2. Based on the previous prediction results, the mine water discharge under mining conditions can be made. The pumping rate is set at 325 m3/h. The simulation results show that there is an obvious decrease in the water inflow (Fig. 10). Comparison of the two predictions indicates that there are increasing trends for both prediction discharges. There

123

is a significant decrease for the mining discharge than that in natural state. For example, the discharge amount decreases from 6.16 to 1.31 m3/min in the first weighting step; it decreases from 8.60 to 1.58 m3/min in the 15th weighting step, and from 11.34 to 2.37 m3/min in the 30th weighting step. Therefore, the simulation results suggest that dewatering should be carried out at -500 m level before mining. Dewatering of the overlying aquifer in North 2 mining area of Donghuantuo Coal Mine makes coal seam #8 minable.

Conclusions 1.

Based on the comprehensive analysis of hydrogeological conditions of Donghuantuo Mine, the hydrogeological conceptual model for prediction of mine discharge of coal seam #8 is established. The fractured sandstone aquifer is the direct recharging source of water inrush when mining coal seam #8, and the lower quaternary porous medium aquifer of gravel and

Environ Earth Sci

2.

3.

4.

5.

6.

pebble is the indirect recharging source of the overlying water inrush. Based on the comprehensive analysis of the geoscience information of six physical fields, together with the mutual comparison of the multi-source geo-information, the thematic map of each main controlling factor is obtained. The zoning map of water abundance of the direct recharging aquifer is constructed with complex superposition of the thematic maps with the use of GIS function of spatial overlay operation. The connectivity map of breaking belt caused by mining of coal seam #8 is constructed of the analysis and calculation of the height of the breaking belt. Superposition of the zoning map of the water abundance and connectivity map of breaking belt leads to the comprehensive vulnerability map of overlyingaquifer water inrush for coal seam #8. The study area is divided into six zones and they are safe zone, less safe zone, transition zone, less dangerous zone, dangerous zone, and the most dangerous zone. The discharges of mining face 2,284 under both natural and mining conditions are predicted with Visual Modflow, which is calibrated with relevant data. Based on above analysis and conclusions, advanced dewatering from the overlying aquifer is proposed as the main prevention and control measure against the overlying-aquifer water inrush.

References Chen S (1999) Introduction of geography information system. Science Press, Beijing, pp 101–163 (in Chinese) Coalfield Geological Central Bureau of China (2000) Coalfield hydrogeology of China. Coal Industry Publishing House of China, Beijing, pp 15–59 (in Chinese) Coulthard MA (1999) Applications of numerical modelling in underground mining and construction. Geotech Geol Eng 17(3–4):373–385 Guiguer N (2004) Waterloo hydrogeologic visual Modflow user’s manual. Waterloo Hydrogeologic, lnc, Ontario

Liu T (1981) The law of surface movement of coalmine and damage of overlying rocks and its application. Coal Industry Publishing House of China, Beijing, pp 10–29 (in Chinese) Ministry of Coal Industry of PRC (1984) Codes of mine hydrogeology coal industry. Publishing House of China, Beijing, pp 45–51 (in Chinese) Morillas S, Gregori V, Riquelme J, Defez B, Peris-Fajarne´s G (2007) Fuzzy directional-distance vector filter. Lect Notes Comput Sci 4578(2007):355–361 Nukala PKVV, Sˇimunovic´ S, Zapperi S, Alava MJ (2007) Fracture in three-dimensional random fuse model: recent advances through high-performance computing. J Comput Aided Mater Des 14(1):25–35 Shang D, Wu Q, Zhao S (2006) Guidance for the prevention and control of mine water disaster. State Administration of Coal Mine Safety Supervision Bureau of China, Beijing, pp 26–48 (in Chinese) State Administration of Coal Mine Safety of China (2009) Interpretation of the regulations of mine water disaster prevention. China University of Mining and Technology Press, Xuzhou, pp 227–236 (in Chinese) Tanabe A, Fukumizu K, Oba S, Takenouchi T, Ishii S (2007) Parameter estimation for von Mises–Fisher distributions. Comput Stat 22(1):145–157 Wu Q (2001) Supervision of coal mine safety (mine water disaster and investigation). China University of Mining and Technology Press, Xuzhou, pp 23–38 (in Chinese) Wu Q, Jin Y (1995) Decision system of water control in mines of North-China-type coal fields. Coal Industry Publishing House of China, Beijing, pp 45–51 (in Chinese) Wu Q, Wei X, Zhang H, Dong D, Yang M, Zhang K (2002) Visual numeric simulation of the height of fractured zone in North 2 working area in Donghuantuo Mine, Kailuan. Coal Geol Explor 30(5):41–44 (in Chinese) Wu Q, Zhang Z, Ma J (2007) The new practical evaluation method of floor water inrush I: the establishment of the main controlling index system. J China Coal Soc 32(1):42–47 (in Chinese) Wu Q, Xu H, Pang W (2008) GIS and ANN coupling model: an innovative approach to evaluate vulnerability of karst water inrush in coalmines of north China. Environ Geol 54(5):937–943 Wu Q, Liu Y, Yang L (2011) Using the vulnerable index method to assess the likelihood of a water inrush through the floor of a multi-seam coal mine in China. Mine Water Environ 30(1):54–61 Zhang D, Zheng S, Sun Y, Ji J (1994) Geographic information system technology and the application in prediction of coalmine water disaster. China University of Mining and Technology, Beijing, pp 45–123 (in Chinese)

123