Journal of Cleaner Production 202 (2018) 33e44 Contents lists available at ScienceDirect Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro Goaf water storage and utilization in arid regions of northwest China: A case study of Shennan coal mine district Qiqing Wang a, Wenping Li a, *, Tao Li b, c, **, Xiaoqin Li a, Shiliang Liu a a School of Resources and Geosciences, China University of Mining and Technology, Xuzhou, 221116, China School of Mining & Civil Engineering, Liupanshui Normal University, Liupanshui, 553004, China c Key Laboratory of Mine Geological Hazards Mechanism and Control, Xi’an, 710054, China b a r t i c l e i n f o a b s t r a c t Article history: Received 25 February 2018 Received in revised form 15 July 2018 Accepted 12 August 2018 Available online 13 August 2018 In arid and semiarid regions of northwest China, there are abundant shallow coal reserves throughout the region, yet during the mining process, the shallow water resources available are typically depleted due to the formation of fractures in the shallow subsurface. Water-preserved mining is the only option for coal mining in arid regions of northwest China. Based on the geological and hydrogeological conditions of the study area, this paper investigates the basic conditions and mechanisms of goaf water storage. A time series model for goaf water storage is proposed, along with a method for estimating the quantity of goaf water storage. Finally, hydrochemical characteristics of mine water and water quality characteristics after puriﬁcation in the study area were analyzed. As indicated by the analysis results, after simple puriﬁcation, the goaf water can meet industrial, ecological and agricultural water consumption standards. In the end, combined with current monitoring results of goaf water pressure and the status of goaf water utilization in the study area, this paper describes the quantity and the effect of utilizing goaf water storage. Goaf water storage can effectively alleviate the scarcity of local production and domestic water, and provides the most effective approach of large-scale water-preserved mining of shallow coal seams in the arid regions of northwest China. © 2018 Published by Elsevier Ltd. Keywords: Arid regions Goaf water storage Goaf water utilization Water resources 1. Introduction With the exhaustion of coal resources in the coal mine districts of eastern China, the focus of coal resource development in China has quickly shifted to the environmentally fragile arid and semiarid regions of northwest China. Shaanxi, Inner Mongolia, Ningxia, Gansu and Xinjiang are major coal-producing provinces to be actively built by the state in the future with a total proven reserve of about 776.9 billion tons of coal (as of 2012), roughly accounting for 54% of the total reserves of the country. In 2016, this region’s coal yield reached 1.62 billion tons, accounting for 48% of the total coal yield of the country. However, these ﬁve provinces of northwest China are arid to semiarid regions (Fig. 1), where there is a severe shortage of water resources, and account for only 1.6% of the total * Corresponding author. School of Resources and Geosciences, China University of Mining and Technology, Xuzhou, 221116, China ** Corresponding author. School of Mining & Civil Engineering, Liupanshui Normal University, Liupanshui, 553004, China. E-mail addresses: firstname.lastname@example.org (Q. Wang), email@example.com (W. Li), firstname.lastname@example.org (T. Li). https://doi.org/10.1016/j.jclepro.2018.08.123 0959-6526/© 2018 Published by Elsevier Ltd. water resources of the country. The fragile eco-geological environment has severely restricted and impacted the regional economic and social development in the region (Feng et al., 2000; Ji et al., 2006; Chang et al., 2014). Where regional geographic, geomorphic and stratigraphic conditions are suitable, water in the shallow layer is an extremely precious resource (Qiao et al., 2017). Such valuable water sources include the sand-layer phreatic water within the upper Pleistocene Salawusu Formation (Q3s) distributed extensively in the coal mine districts of northern Shaanxi Province and the bottomland of the Maowusu desert, the surface reservoirs (lakes), and the runoff water formed by atmospheric precipitation in the loess hilly areas (Fig. 2). In recent years, large-scale, highintensity and extensive mining of coal resources in northwest China has caused wide-range water seepage from the shallow layer into the subsurface coalmines, further escalating the shortage of water resources and seriously impacted production and life of local residents. Protection and proper utilization of the water resources in connection with large-scale coal mining poses vital scientiﬁc and technological problems in the arid regions of northwest China. Targeting the above problems, there have been several multidisciplinary studies on “water-preserved mining” from many 34 Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 F1 List of symbols a2 Q4eol Q3s Q2l N2b J2y Qt Q1 Q2 Q3 q1 q2 F a1 Eolian sand of Holocene System Upper Pleistocene Salawusu Formation Middle Pleistocene Lishi Formation Pliocene Baode Formation Middle Jurassic Yan’an Formation The total water inﬂow of goaf water storage, m3/d The water inﬂow of atmospheric precipitation, m3/d The water inﬂow of surface water, m3/d The water inﬂow of groundwater, m3/d The water inﬂow of precipitation within the scope of goaf, m3/d The water inﬂow of precipitation within the peripheral catchment area of the goaf, m3/d The area of the goaf, m2 The precipitation inﬁltration coefﬁcient of the goaf h q3 q4 Q0 S0 S m2 t tm Hm m The peripheral catchment area of the goaf, m2 The precipitation inﬁltration coefﬁcient in the peripheral catchment area of goaf The precipitation intensity of the study area, m/d The ﬂow of surface water into the goaf, m3/d The ﬂow of surface water out of the goaf, m3/d The water inﬂow of the goaf before enclosure, m3/d The drawdown of the phreatic water level before goaf closure, m The drawdown of the phreatic water level in the goaf water storage, m The speciﬁc yield of the goaf caving zone The time of goaf water storage, d The time of full water storage needed by the goaf caving zone, d The development height of caving zone, m The average speciﬁc yield of the goaf ﬁssure zone Fig. 1. Map of the distribution of arid and semiarid regions along with the Jurassic coal basin, China. aspects. Some have explored the impact of coal mining on groundwater (Kim et al., 1997; Karaman et al., 2001). Gu and Zhang (2012) speciﬁed the inﬂuence principle of modern coal mining on the groundwater environment and the trend of self-remediation in the northwest China. Yu et al. (2017) and Wang et al. (2017) stated the development characteristic of water-ﬂowing fractured zones under the inﬂuence of coal mining. Wang et al. (2012) provided a mechanical model, predicting the height of water-ﬂowing fractured zone for a shallow seam covered with thin bedrock and thick windblown sands. Many scholars have developed several waterpreservation mining techniques aimed at protecting impermeable key strata, such as backﬁll mining (Liu et al., 2006), limited-height mining (Miao et al., 2009), room-and-pillar water-preservation mining (Shi and Hou, 2006) and water-preserved mining division (Huang, 2010). In addition, the hydrochemical characteristics of mine water (such as water quality characteristics and evolution characteristics) have been extensively studied (Huisamen and Wolkersdorfer, 2016; Bozau et al., 2017), along with the recycling of mine inﬂow (Zhang, 2010; Gu, 2015). However, the above studies mainly focused on the migration pathways through which coal mining impacts water resources, but have yet systematically studied the water bodies themselves. The quality of mine water and the recycling of mine inﬂow have been investigated to some extent (Li et al., 2013; Gao et al., 2016; Mativenga and Marnewick, 2018), but there is a lack of studies on water resources conﬁned in goafs. Taking the Shennan coal mine district of northern Shaanxi Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 (a) Sand-layer phreatic water 35 (b) Surface runoff (c) Reservoir Fig. 2. Shallow water resources in northwestern arid regions, China. Province as an example, this paper investigates the formation conditions and mechanisms of goaf water storage, establishing a time series model for goaf water storage on this basis, and proposes a time series method for predicting goaf water storage capacity. It also analyzed the water quality characteristics within the goaf water storage and the utilization of mine water in the study area. The research ﬁndings provide a new approach for solving the shortage of water resources within arid coal mining regions, and are of great signiﬁcance for the sustainable development of the environment, economy and society in arid and semi-arid mining areas of northwest China. 2. Study area The Shennan coal mine district encompasses 363.4 km2 and includes a total coal reserve of about 5.5 billion tons with an annual yield well over 50 million tons (Yang et al., 2018). It is located in the contiguous area between the Loess Plateau and the Maowusu Desert in northern Shaanxi Province, and the geomorphological features within this area include desert bottomlands, valleys and loess hill gullies (Fig. 3). This area has an annual precipitation of about 400 mm and an annual evaporation of over 1,700 mm (China Meteorological Data Service Center, 2018). The region is currently undergoing the initial stage for large-scale coal mining, but there has already been an increasing shortage of available water resources (surface water and shallow groundwater). It is necessary to plan the utilization of water resources. The strata from the earth surface to coal seam in the study area are in ascending order (Fig. 4): (1) Sand-layer phreatic aquifer has a general thickness of 10e20 m, and consists of eolian sand of Holocene System (Q4eol) and the upper Pleistocene Salawusu Formation (Q3s). The eolian sand is mainly distributed in the desert bottomland, and constitutes a uniﬁed sand-layer phreatic aquifer with the underlying Salawusu Formation, with a general level depth of 5 m below the surface. The Salawusu Formation is a ﬂuvio-lacustrine deposit dominated by silty-ﬁne sand and rich in groundwater. As the primary aquifer in the coal mine district, it is exposed along the river valleys, and is the target stratum of water-preserved mining. (2) The loess of the middle Pleistocene Lishi Formation (Q2l) is generally 20 m in thickness, and directly distributed on earth surface in loess mountain ridge areas. It is a sandy-clay with a coefﬁcient of permeability of 0.04e0.13 m/d and is the direct aquifuge of the sand-layer phreatic water (Chen et al., 2017). It constitutes a key aquifuge for water-preserved mining together with the underlying Baode laterite. (3) The lateriteof the Pliocene Baode Formation (N2b) is generally 30 m in thickness, and, as a silty-clay with a coefﬁcient of permeability of 0.0016e0.017 m/d, is a relatively strong natural impermeable layer. (4) The weathered bedrock of the middle Jurassic Yan’an Formation (J2y) is generally 25 m in thickness, and consists of mudstone, sandy mudstone and sandstone. It has well developed weathering ﬁssures accompanied by the occurrence of ﬁssure water. (5) The unweathered bedrock of the middle Jurassic Yan’an Formation (J2y) is generally 29e250 m in thickness, and is the coal-bearing strata. The ﬁrst mining coal seam is the No.2-2 with a mining thickness of 6 m and a burial depth of 36 Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 Fig. 3. Study area and examples of typical landform. Fig. 4. Typical proﬁle sketch of the study area. 90e200 m. Coal mining exerts a direct inﬂuence on the overlying key impermeable clay stratum, for which the water within the sand layer experiences varying degrees of leakage. Some coal seams have undergone spontaneous combustion, and formed various forms of burnt rocks, and are accompanied by the occurrence of water resources to some extent. 3. Conditions and mechanism for goaf water storage 3.1. Conditions for goaf water storage Goaf water storage should meet the following basic conditions: (1) The coal-bearing strata are structurally intact and have low permeability. The goaf surrounding rock mass of the coal seam consist of an impermeable or extremely weakly permeable stratum. For instance, the Jurassic stratum in the Shennan coal mine district of northern Shaanxi Province (similar regional conditions to the Jurassic coal seam of northwest China) essentially has no faulting or jointing, and is an overall thick-bedded formation. With a coefﬁcient of permeability of ranging from K ¼ 0.0000244e0.09 m/d, it is an extremely weakly to weakly permeable stratum. (2) The coal seam should be thick, or contain multiple seams and subjected to large-scale coal mining, therefore containing large goaf volume and a large water storage capacity. In the Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 coal mines of northwest China, large-sized coal pillars (generally greater than 50 m in size) remain in place between panels, and panel goafs are stable water storage units with high horizontal impermeability. Adopting panels as basic water storage spaces, multiple panel goafs can be combined for water storage, thus creating large water supply sources for distributed goaf water storage. For instance, one panel of goaf water storage space in the Shennan coal mine district can store millions and millions of cubic meters of water. (3) Recharge path: After coal mining, the overburden waterﬂowing fractured zone becomes a recharge path for goaf water storage, and water converges in the goafs only in places where the mining depth is relatively shallow, bedrock and soil-stratum aquifuges are thin (a thin aquifuge resulting in cross-ﬂow leakage) and the path converges or almost converges with water sources. (4) Sufﬁcient recharge water sources: Available goaf water storage can occur only in regions with sufﬁcient water sources where water can continuously seep into goafs through recharge paths consisting of mining-induced fractures. These water resources mainly include surface reservoirs, gully runoff, the Salawusu sand-layer phreatic water, atmospheric precipitation, etc. (5) Large-sized goaf water supply sources can be formed only in regions with suitable hydrogeological conditions (such as large water storage space, sufﬁcient recharge water sources, high impermeability of the surrounding rocks, stable goafs and so forth). The regions of northwest China have an annual evaporation of 1500e4,000 mm$a1, but their annual precipitation is only 200e400 mm$a1 (China Meteorological Data Service Center, 2018). Through the mining-induced fractures, water from the shallow layer can be transferred underground for storage, thus avoiding the intense evaporation that occurs to surface water and increasing the total amount of water resources in the coal mine district. 3.2. Mechanism of goaf water storage The primary condition for the water storage in a goaf is that water resources can be effectively and safely stored. In fact, a goaf as the lowest point in the region has the lowest water potential energy and water catchment capacity. Before coal mining, three types of water (i.e., surface water, groundwater and atmospheric precipitation) are in constant and orderly circulation, and the hydraulic connections among various types of water are in a relatively balanced state. The sand-layer phreatic water can recharge water within the burnt rocks and surface water; bedrock weathering ﬁssure water can recharge the surface water; in case of a ﬂood during the ﬂood season, surface water can recharge water within the burnt rocks; atmospheric precipitation serves as an ultimate recharge source for all types of surface water and groundwater. However, a fractured zone induced by coal mining may link surface water and groundwater, change the hydraulic connections among various types of water, promote various types of water to directly or indirectly discharge to the lowest goafs through the water-ﬂowing fractured zone, and thus create goaf water storage. The mechanism of the goaf water storage process under shallow coal mining in the arid regions of northwest China is depicted in Fig. 5. The amount of precipitation inﬁltration increased due to mining of shallow coal seams in an arid region. Surface water resources in an area prone to water leakage have also entered the goafs. Although the evaporation amount of an underground water body increases due to ground ﬁssures, relative evaporation of surface water is weak. Therefore, the total amount of evaporative water 37 resources decreases after coal mining; the increase of atmospheric water transformation in a coal mining area is conducive to the positive balance of the regional water system. In the study area, although the recharge of groundwater increased after coal mining, it was found that the groundwater near the goaf is draining into the mine, and the surface water in the mining area is essentially depleted. Without considering the mine water as available water resources, the amount of water resources at the surface of the mining area will continue to decrease year after year. Conversely, water resources in the mining area itself after coal mining are increasing year after year due to the drainage into the mine. 4. Evaluation for goaf water storage The capacity and time of water storage are key to water utilization in goaf. The water quality is another concern for the utilization of water resources available in the goaf. In this study, the dynamic evolution process on goaf water storage was analyzed. The capacity and time of water storage in goaf were evaluated, and the goaf water quality characteristics were tested. The ﬂowchart of evaluation for water storage in goaf is shown in Fig. 6. 4.1. Dynamic evolution analysis on goaf water storage The total water inﬂow Qt of goaf water storage is calculated as follows: Qt ¼ Q1 þ Q2 þ Q3 (1) where Qt is the total water inﬂow of goaf water storage, m3/d; Q1 is the water inﬂow of atmospheric precipitation, m3/d; Q2 is the water inﬂow of surface water, m3/d; and Q3 is the water inﬂow of groundwater, m3/d. After goaf formation, there are obvious differences between the stopped area and the peripheral catchment area of the goaf (the area affected by the cone of depression) in terms of the precipitation inﬁltration coefﬁcient. However, the inﬁltration capacity of atmospheric precipitation does not show any signiﬁcant change with the change of goaf water storage in either case. Thus, the water inﬂow of atmospheric precipitation can be computed by the following formula: Q1 ¼ q1 þ q2 (2) where Q1 is the water inﬂow of atmospheric precipitation, m3/d; q1 is the water inﬂow of precipitation within the scope of the goaf, m3/ d; q2 is the water inﬂow of precipitation within the peripheral catchment area of the goaf, m3/d; q1 and q2 are computed by the following formulae: q1 ¼ F a1 h (3) q2 ¼ F1 a2 h (4) where F is the area of the goaf, m2; a1 is the precipitation inﬁltration coefﬁcient of the goaf; F1 is the peripheral catchment area of the goaf, m2; a2 is the precipitation inﬁltration coefﬁcient in the peripheral catchment area of the goaf; h is the precipitation intensity of the study area, m/d. The amount of water entering the goaf from the surface water does not show any obvious change with the change of goaf water level, so the water inﬂow of surface water can be computed by the following formula: 38 Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 Fig. 5. Water cycle in goaf, illustrating water pathways. Q2 ¼ q3 q4 (5) where q3 is the ﬂow of the surface water into the goaf, m3/d; q4 is the ﬂow of surface water out of the goaf, m3/d. Among the aquifers overlying the coal seams in the study area, the aquifer directly recharging the goaf is mainly the sand-layer phreatic aquifer, whose recharge effect on the goaf is directly related to head difference. Thus, the amount of recharge of the underground aquifer for the goaf is related to goaf water storage level. According to Wei (1991), the groundwater inﬂow is directly proportional to the 1/2 power of the water level decline ratio, and can be computed by Formula (6): sﬃﬃﬃﬃﬃ S Q3 ¼ Q0 S0 (6) fracture zone, providing a spacious underground water storage unit. There is an obvious difference in the speciﬁc yield between the caving zone and ﬁssure zone. Therefore, the water storage process of a goaf includes two stages: the water storage process of caving zone and the ﬁssure zone (Fig. 8). (1) Water storage process computation for the caving zone The time dt of goaf water storage presents the drawdown change dS of phreatic water level, so: sﬃﬃﬃﬃﬃ ! S Q1 þ Q2 þ Q0 dt ¼ F m2 dS S0 (7) where m2 is the speciﬁc yield of the goaf caving zone; assuming 0 ﬃ , Formula (8) can be obtained through K1 ¼ Q1 þ Q2 , K2 ¼ pQﬃﬃﬃ S0 where Q0 is the water inﬂow of the goaf before enclosure, m3/d; S0 is the drawdown of the phreatic water level before goaf closure, m; S is the drawdown of the phreatic water level in the goaf water storage, m; the deﬁnition of the above variables S and S0 are shown in Fig. 7. Coal mining causes overburden failure to form a caving zone and integration: " pﬃﬃﬃﬃﬃ# pﬃﬃﬃﬃﬃ pﬃﬃﬃ 2F m2 K1 þ K2 S0 pﬃﬃﬃ S0 S K1 ln t¼ K2 K22 K1 þ K2 S (8) Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 39 Evaluation of goaf water storage Dynamic evolution on goaf water storage Total water inflow of goaf water storage (Qt) Atmospheric precipitation (Q1=q1+q2) Precipitation within the scope of the goaf (q1) Analysis of goaf water quality The time of goaf water storage (t) Surface water (Q2=q3-q4) Precipitation within the peripheral catchment area of the goaf (q2) Surface water into the goaf (q3) Water storage capacity and time for the caving zone Groundwater (Q3= Q0 S S0 ) Surface water out of the goaf (q4) Proportional to the 1/2 power of the water level decline ratio (Q0 S S0 ) Cations, anions, permanent hardness, pH, ect. Water storage capacity and time for the fissure zone Water storage capacity (Qs) and time (tz) for the goaf Design plan of goaf water utilization Water treatment and utilizationin in goaf Fig. 6. Flowchart of evaluation for water storage in goaf. Fig. 7. Sketch map for water storage variables in goaf. wherein, Formula (8) is a transcendental equation, according to which the approximate solution of S can be computed when time t is given. (2) Water storage process computation for the ﬁssure zone Through substituting S¼S0-Hm into Formula (8), the time tm it takes for full water storage in the caving zone can be solved: tm " pﬃﬃﬃﬃﬃ # pﬃﬃﬃﬃﬃ pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 2F m2 K1 þ K2 S0 pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ S0 S0 Hm K1 ln ¼ K2 K22 K1 þ K2 S0 Hm (9) where tm is the time needed for full water storage within the goaf caving zone, d; Hm is the development height of caving zone, m. When t＞tm, the caving zone is fully ﬁlled with water, and the ﬁssure zone begins to store water. During the time dt taken for the 40 Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 Fig. 8. Diagram of the water storage process stages in a goaf. goaf caving zone to completely ﬁll with water, the phreatic water level changes by dS, so the following formula can be obtained through integration: " pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ# pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ pﬃﬃﬃ 2F m K1 þ K2 S0 Hm pﬃﬃﬃ S0 Hm S K1 ln t ¼ 2 K2 þ tm K2 K1 þ K2 S (10) where m is the average speciﬁc yield of the goaf ﬁssure zone. When S ¼ 0, the time tz taken for the goaf to completely ﬁll with water is: pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ! pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 2F m K1 þ K2 S0 Hm tz ¼ 2 K2 S0 Hm K1 ln þ tm K1 K2 (11) By jointly solving Formulae (8) and (10), the relationship between the time t of goaf water storage and the drawdown S of the phreatic water level can be obtained, and with S we further determine the current goaf water storage Qs: QS ¼ ðS0 SÞF m2 ; S0 Hm S S0 (12) QS ¼ Hm F m2 þ ðS0 Hm SÞF m; 0 S S0 Hm (13) 4.2. Analysis of goaf water quality In the study area, there is plenty of water in the goaf that is also shallow, which is a rare precious water resource in the in an arid and water depleted area. In order to analyze the water quality characteristics of the goaf in the study area, seven sets of water samples were collected from the Shennan coal mine district and its neighbouring mine water, residential well water and loose sandbed spring water. All were tested for constituents, including cat 2 ions (Ca2þþMg2þ, Fe2þþFe3þ, NHþ 4 ), anions (NO3 , NO2 , SO4 , HCO3 ), permanent hardness and pH. The water in the goaf can be treated using a sedimentation tank to control the ﬂow speed, remove most of the suspended solids and chemical oxygen demand (COD) in the water, reduce the chromaticity and adjust the pH value. consists of ﬁve sections with individual goafs within each section interconnected: six goafs on the west side of the north wing (N1201~N1211), four goafs on the east side of the north wing (N1200-1~N1204), two goafs on the west side of the south wing (S1205 and S1207), three goafs on the east side of the south wing (S1217~S1221) and goaf S1210 on the south wing (Fig. 9). Fire dams at relatively low goaf points under the coal mine determined by contour mapping of coal mine district ﬂoor and other reference data were mounted with water pressure monitors, i.e., MB1, MB2, MB3, MB4 and MB5; their speciﬁc positions are shown in Fig. 9. Currently, 95% of the water consumption in the study area comes from goaf water and mine drainage. The production water of the mine consists of three segments. First, that collected from the ﬁre dam of the air return way in goaf S1210 on the south wing, connected with water supply pipeline F219, and supplied after treatment via hole G2 to the south wing for production and ﬁre protection. The second portion is collected from the goafs on the west side of the north wing, connected with water supply pipeline F159, and supplied after treatment to the north wing for ecological restoration. The third portion is collected from the goafs on the south wing and north wing along pipeline G3 of the main haulage roadway pipeline. It ﬁrst arrives at the ground sewage treatment station after going through the south wing sump, north wing sump and central sump. It is then used as supplementary water for the coal preparation plant after treatment. The domestic water of the mine mainly comes from a water well sourced under the mine. The water on the south wing comes from the water well through auxiliary transportation along goaf S1210 at a depth of 4600e5,900 m, and, according to water quality tests, it meets drinking water standards of the state. The water well ﬂows under gravity to the clean water sump along auxiliary transport way S1210 at a depth of 2,700 m. After being pressurized, it is connected with pipeline F159, and supplied via G5 to the surface for use as domestic water. The roadway water well for working face N1206 on the north wing is connected with water supply pipeline F219 via the main haulage roadway, and supplied via G4 to the north wing for use as industrial water by the Shaanxi Tiangong Construction Co., Ltd. 5. Results and discussion 5.1. Estimation of goaf water storage capacity 4.3. Design plan of goaf water utilization The Ningtiaota Coal Mine located to the west of the Shennan coal mine district has an existing goaf area of about 6.97 km2 and Working face N1201 of the Shennan coal mine district, which intersects surface runoff of the Xinmin Gully, was taken as an example. Through combining ﬁeld surveys with the results of Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 41 rection Drainage di n Transportatio N1209 goaf roadway roadway Ventilating 0.5 0 N1211 goaf 1.0 km N er Ground wat g(N1) intake drillin Sump face N1206 coal N1207 goaf N1205 goaf N1204 goaf North area N1203 goaf MP2 N1202 goaf MP1 N1201 goaf N1200 goaf eway Main haulag way ad ro ting Main ventila ting roadway Main ventila af N1200-1 go d Substation an sump of north area t Incline shaf Mine water ants treatment pl MP4 S1210 coalface S1221 goaf Settling pond The annular mp of south su t af Incline sh area S1219 goaf South area S1217 goaf MP3 S1207 goaf S1205 goaf S1203 coalface The central d substation an sump d Substation an h sump of sout area Ground water intake drilling(S1) MP5 Fig. 9. Design plan of goaf water utilization in the Shennan coal mine district. previous studies, parameters Q0, Q1, Q2, F, S0, m2, m and Hm were respectively determined as 15 m3/h, 7.4 m3/h, 22.6 m3/h, 808,300 m2, 65 m, 0.15, 0.11 and 17 m. According to the above calculations, the caving zone of working face N1201 requires 206.1 104 m3 to fully ﬁll the goaf over the course of 5.3 years. The water-ﬂowing fractured zone of the goaf can hold 632.9 104 m3 of water, of which would take 14.1 years to reach full capacity. It should be noted that the time taken for the goaf to completely ﬁll with water is also affected by many uncertainty factors, such climate change or peripheral mining activities that were not taken into account in the calculations. 5.2. The characteristics of goaf water quality The water quality analysis results of the Shennan coal mine district and its peripheral mine water, residential well water and loose sand-bed spring water are in Table 1. Compared with natural water, mine water does not shown any substantial change in any index, except the obvious increase of permanent hardness and SO2 4 . On the whole, the mine water was weakly alkaline, which can be explained by the ultra-low sulfur coal deposited in the Shennan coal mine district of northern Shaanxi Province. The quality of goaf water is extremely complicated, and subjected to many factors, such as water ﬁlling sources, coal quality, surrounding rock characteristics, residence time, hydrodynamic environment, etc. The inﬂuence of coal quality on the initial water quality within the goaf water storage is very signiﬁcant, however, with the long-term circulation of water resources, several harmful elements have shown a signiﬁcant remobilization. Thus the water greatly beneﬁts from puriﬁcation. Before goaf water utilization, the water should undergo simple puriﬁcation treatment to meet relevant water consumption standards. Based on the water quality test on adjacent coalmines (Miao et al., 2009), the water quality characteristics of goaf water storage after puriﬁcation are provided in 42 Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 Table 1 Water quality analysis results for representative samples from Shennan and its neighbouring coal mine districts. Sample sites Cation (mg/L) Springs from sand aquifer Well water Haolaigou opencast coal mine Sunjiacha coal mine Xinyao coal mine Ningtiaota coal mine Hongliulin coal mine Anion (mg/L) Ca2þþMg2þ Fe2þþFe3þ NHþ 4 NO 3 NO 2 SO24 HCO 3 53.7 51.2 120.2 109.2 133.22 78.1 55.6 0 0.02 0.04 not detected not detected 0.08 0.08 0.02 0.06 not detected 0.16 not detected 0.03 0.03 9.0 9.0 not detected 18.0 0.6 6.3 2.5 not detected not detected not detected 0.99 0.16 0.004 0.003 11.1 15.2 120.2 140.7 184.1 138.6 142.5 183.9 168.7 323.3 265.3 353.1 155.6 186.2 Table 2 Water quality characteristics of treated mine water. No. Item Standard Treated 1 2 3 4 5 6 7 8 9 10 11 pH turbidity (mg/L) suspended solids (mg/L) ammonia (mg/L) volatile phenols (mg/L) hexavalent chromium (mg/L) sulﬁde (mg/L) total number of bacteria (a/mL) total hardness (mg/L) COD (mg/L) BOD5 (mg/L) 6.0e9.0 e 100.0 15.0 0.5 0.5 110.0 e 450.0 100.0 30.0 6.9 3.2 25 2.6 not detected 0.03 0.15 63 148.0 31.0 11.0 Table 2. The quality of goaf water can fully meet industrial, ecological and agricultural water consumption standards after the simple puriﬁcation treatment. 5.3. Analysis of goaf water utilization in the Shennan coal mine district Fig. 10 provides the results of goaf water pressure monitoring within various parts of the mine, from which it can be seen that no goaf water pressure experienced any signiﬁcant change on the whole. Some of the water pressures ﬂuctuated over the monitoring period mainly because of typical drainage and utilization of the goaf water throughout the monitoring period. The monitoring results indicate that S1210 (south wing) had the highest goaf water pressure with a maximum water pressure of 202.9 kPa, equivalent to a goaf water head height of 20.29 m (still lower than the height of the caving zone of about 21.0 m). This suggests that the goaf water Permanent hardness (German degree) pH 0 0 1.9 4.4 5.4 3.3 3.2 7.4 7.3 7.4 7.2 7.4 7.8 7.1 storage of the study area is still in stage I, i.e., ﬁlling of the caving zone. Its water quantity can be computed by Formula (12), wherein, m2 was set at 0.15. According to the computation results provided in Table 3, the goaf water quantity of the study area was about 2,061,000 m3 at the time of calculation. Flow meters were used under the mine to monitor the water draw rates of G1, G2, G3, G4 and G5, and data were collected every 5 min. According to the monitoring results in Fig. 11, G1 had a maximum instantaneous ﬂow rate of 84 m3/h, and an average instantaneous ﬂow rate of 1.5 m3/h; G2 had a maximum instantaneous ﬂow rate of 386 m3/h, and an average instantaneous ﬂow rate of 217 m3/h; G3 had a maximum instantaneous ﬂow rate of 111 m3/h, and an average instantaneous ﬂow rate of 5 m3/h. The water supply pipelines had regular rest periods, and were not under continuous 24-h service for production water supply. Production water of the entire mine had an average instantaneous ﬂow rate of 223.5 m3/h. G4 had a maximum instantaneous ﬂow rate of 12 m3/h, and an average instantaneous ﬂow rate of 1.3 m3/h; G5 had a maximum instantaneous ﬂow rate of 190 m3/h, and an average instantaneous ﬂow rate of 99 m3/h. According to monitoring data, the water supply pipelines had normal rest periods, and were not under continuous 24-h service for domestic water supply. The domestic water of the entire mine had an average instantaneous ﬂow rate of 100.3 m3/h. Currently the water consumption and goaf water storage variations within the study area are essentially balanced, suggesting that the goaf water quantity of the study area can fully meet current industrial and agricultural water demands. Because of the entry of surface water bodies and groundwater into the goafs, surface water has a tendency to be completely depleted, and groundwater level drops abruptly. According to ﬁeld 300 MB1 MB3 MB5 250 Water pressure (kPa) 200 MB2 MB4 150 100 50 25 20 15 10 6/1 6/4 6/7 6/10 6/13 6/16 6/19 6/22 6/25 6/28 7/1 Date 7/4 7/7 7/10 7/13 7/16 Fig. 10. Water pressure monitoring results of the goafs in the Shennan coal mine district. 7/19 7/22 7/25 Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 43 Table 3 Current water storage capacity of the goafs in the study area. Goaf Area (1 104m2) Height of water (m) Water storage in goafs (1 104m3) N1201~N1211 N1200-1~N1204 S1205, S1207 S1217~S1221 S1210 Sum 337.5 172.0 112.3 72.6 2.1 2.2 1.7 1.5 1.8 18.6 111.4 43.9 25.3 19.6 5.9 206.1 Flowing rate (m3/h) 400 G1 G3 G5 300 200 G2 G4 100 0 6/1 6/4 6/7 6/10 6/13 6/16 6/19 6/22 6/25 6/28 7/1 Date 7/4 7/7 7/10 7/13 7/16 7/19 7/22 7/25 Fig. 11. Flow monitoring results of water consumption pipelines in the Shennan coal mine district. (a) Unirrigated area (b) Irrigated area utilizing goaf water storage Fig. 12. Ecological environment within the study area. surveys over a 15-year period, after the water level drops and without artiﬁcial irrigation, some tall arbors within the ground surface above the goafs have also been depleted, resulting in the escalation of desertiﬁcation in the region, as shown in Fig. 12 (a). However, in areas where goaf water storage is used for irrigation, various types of vegetation (Table 4) are obviously in better condition than those in unirrigated areas, as shown in Fig. 12 (b). This suggests that it is feasible to use the reclaimed and treated water obtained from goaf water storage for ecological restoration and agriculture. Table 4 Main vegetation types in irrigated and unirrigated areas. Areas Category Names of vegetation Areas Category Names of vegetation Irrigated areas Arbor Pinus sylvestris L. var. mongholica Litv. Populus L. Pinus tabulaeformis Carr. Salix psammophila Caragana korshinskii Kom Artemisia desertorum Spreng. Syst. Veg. Hippophae rhamnoides Linn. Cynanchum hancockianum (Maxim.) Al. Iljinski. Cynanchum thesioides (Freyn) K. Schum Aristida adscensionis Linn Cleistogenes squarrosa (Trin.) Keng. Potentilla chinensis Ser. Agriophyllum squarrosum (L.) Moq. Irrigated areas Herbs Unirrigated areas Arbor Heteropappus altaicus (Willd.) Novopokr Hemerocallis citrina Baroni Medicago sativa L. Poa annua L. Pinus sylvestris L. var. mongholica Litv. Populus L. Salix psammophila Caragana korshinskii Kom Artemisia desertorum Spreng. Syst. Veg. Cynanchum hancockianum (Maxim.) Al. Iljinski. Cynanchum thesioides (Freyn) K. Schum Agriophyllum squarrosum (L.) Moq. Tribulus terrestris L. Shrub Herbs Shrub Herbs 44 Q. Wang et al. / Journal of Cleaner Production 202 (2018) 33e44 6. Conclusions Northwest China consists of typical arid and semiarid regions, where surface water evaporation is far greater than precipitation. In large-scale coal mining, there are extensive goafs throughout the region, and underground goaf water storage can be used to increase the total amount of water resources in the coal mine district. This provides a new approach of effectively achieving “water-preserved mining” and solving the shortage of industrial and agricultural water in these regions. Through analyzing the geological and hydrogeological conditions of the study area, this paper maintains that large-sized goaf water supply sources can be formed only in coal mine districts with suitable hydrogeological conditions (such as large goaf volumes, sufﬁcient recharge water sources, high impermeability of goaf surrounding rocks, stable goafs, and so forth). Beyond proposing that panels can be used as basic units to implement goaf water storage, this paper provides the formulas for calculating goaf water storage capacity and predicting water storage durations. The present situation of water storage in the goaf with in the Shennan mining district was determined to still be within stage I, i.e., ﬁlling of the caving zone, and the goaf water quantity was about 2,061,000 m3. It also analyzed the water quality characteristics of goaf water in northern Shaanxi Province, and concluded that water quality exerts an insigniﬁcant inﬂuence on the utilization of goaf water resources. Through investigating the status of goaf water utilization in the study area, goaf water can be used for industrial purpose and environmental restoration after simple treatment, which further veriﬁes the feasibility of using goaf water storage to solve the shortage of water resources in arid regions. Acknowledgments The authors express their gratitude to everyone that provided assistance for the present study. The study was jointly supported by the National Key Basic Research Program of China (973 Program) (Grant No. 2015CB251601), the State Key Program of National Natural Science Foundation of China (Grant No. 41430643) and the China Postdoctoral Science Foundation (Grant No. 2018M630632). References Bozau, E., Licha, T., Ließmann, W., 2017. Hydrogeochemical characteristics of mine water in the Harz Mountains, Germany. Chem. Erde-Geochem 77 (4), 614e624. Chang, J., Li, W., Li, T., Qiao, W., Wu, F., 2014. Water migration in arid mining area and thought of “water preserved mining”. J. Min. Saf. Eng. 31 (1), 72e77 (in Chinese). Chen, W., Xie, X., Peng, J., Wang, J., Duan, Z., Hong, H., 2017. GIS-based landslide susceptibility modelling: a comparative assessment of kernel logistic regression, Naïve-Bayes tree, and alternating decision tree models. Geomat. Nat. Haz. Risk 8 (2), 950e973. China Meteorological Data Service Cent, 2018. Hourly Data from Surface Meteorological Stations in China. Available online from. http://data.cma.cn/. Feng, Q., Cheng, G.D., Masao, M.K., 2000. Trends of water resource development and utilization in arid north-west China. Environ. Geol. 39 (8), 831e838. Gao, L., Hou, C., Chen, Y., Barrett, D., Mallants, D., Li, W., Liu, R., 2016. Potential for mine water sharing to reduce unregulated discharge. J. Clean. Prod. 131, 133e144. Gu, D., 2015. Theory framework and technological system of coal mine underground reservoir. J. Chin. Coal Soc. 40 (2), 239e246 (in Chinese). Gu, D., Zhang, J., 2012. Modern coal mining affected to underground water deposit environment in West China mining area. Coal Sci. Technol. 40 (12), 114e117 (in Chinese). Huang, Q., 2010. Impermeability of overburden rock shallow buried coal seam and classiﬁcation of water conservation mining. Chin. J. Rock Mech. Eng. 29 (S2), 3622e3627 (in Chinese). Huisamen, A., Wolkersdorfer, C., 2016. Modelling the hydrogeochemical evolution of mine water in a decommissioned opencast coal mine. Int. J. Coal Geol. 164, 3e12. Ji, X., Kang, E., Chen, R., Zhang, W., Zhang, Z., Jin, B., 2006. The impact of the development of water resources on environment in arid inland river basins of Hexi region, Northwestern China. Environ. Geol. 50 (6), 793e801. Karaman, A., Carpenter, P.J., Booth, C.J., 2001. Type-curve analysis of water-level changes induced by a longwall mine. Environ. Geol. 40 (7), 897e901. Kim, J.M., Parizek, R.R., Elsworth, D., 1997. Evaluation of fully-coupled strata deformation and groundwater ﬂow in response to longwall mining. Int. J. Rock Mech. Min. 34 (8), 1187e1199. Li, T., Wang, S., Li, W., Deng, Z., Chang, J., 2013. Conditions and time sequence of water storage by goaf in arid water shortage mining area. Coal Eng. 45 (S1), 94e97 (in Chinese). Liu, Y., Shi, P., Zhang, Z., 2006. Technique parameters analysis of strip mining without destroying water resource in low burying coal seam. Coal Min. Technol. 11 (6), 6e10 (in Chinese). Mativenga, P.T., Marnewick, A., 2018. Water quality in a mining and water-stressed region. J. Clean. Prod. 171, 446e456. Miao, X., Wang, A., Sun, Y., Wang, L., Pu, H., 2009. Research on the basic theory of mining with water resources protection and application to arid and semi-arid mining areas. Chin. J. Rock Mech. Eng. 28 (2), 217e227 (in Chinese). Qiao, W., Li, W., Li, T., Chang, J., Wang, Q., 2017. Effects of coal mining on shallow water resources in semiarid regions: a case study in the Shennan mining area, Shaanxi, China. Mine Water Environ. 36 (1), 104e113. Shi, B., Hou, Z., 2006. Research on coal mining method with water conservation in Yushen Mining Area. Coal Eng. 06 (1), 63e65 (in Chinese). Wang, G., Wu, M., Wang, R., Xu, H., Song, X., 2017. Height of the mining-induced fractured zone above a coal face. Eng. Geol. 216, 140e152. Wang, L., Wang, Z., Huang, J., Zhou, D., 2012. Prediction on the height of waterﬂowing fractured zone for shallow seam covered with thin bedrock and thick windblown sands. J. Min. Saf. Eng. 29 (5), 607e612 (in Chinese). Wei, K., 1991. Mine Hydrogeology. Coal Industry Press, Beijing (in Chinese). Yang, Z., Li, W., Pei, Y., Qiao, W., Wu, Y., 2018. Classiﬁcation of the type of ecogeological environment of a coal mine district: a case study of an ecologically fragile region in Western China. J. Clean. Prod. 174, 1513e1526. Yu, B., Zhao, J., Xiao, H., 2017. Case study on overburden fracturing during longwall top coal caving using microseismic monitoring. Rock Mech. Rock Eng. 50 (2), 507e511. Zhang, J., 2010. Analysis of resource utilization of mine water gushing. Ind. Saf. Environ. Protect. 36 (7), 42e43.