A method for identifying and warning karst water layer pollution in coal mine goaf

Abstract

The present invention discloses a method for identifying and warning of karst water layer pollution in coal mine goaf areas, which relates to the technical field of hydrogeology. The method comprises: 1. monitoring the water level of coal-bearing strata, karst water, and goaf areas to determine the pressure head, and measuring the SO4 2‑ , pH, and conductivity are used to identify the pollution trend and cross-layer threshold of karst aquifers; ② determine the hydrodynamic and hydrochemical characteristics of the groundwater system at different times, and determine the chemical composition of the groundwater system discharge; ③ determine the groundwater flow direction, obtain hydrogeological parameters, and judge the hydraulic connection of the aquifer; ④ analyze the activation of fault zones and the generation of fracturing zones, and find potential cross-layer pollution channels; ⑤ carry out cross-layer pollution warning, which effectively reduces the cost of monitoring and warning, and realizes the efficient and comprehensive identification and warning of karst water cross-layer pollution in goaf areas.

Description

Technical Field

[0001] The invention relates to the technical field of hydrogeology, in particular to a method for identifying and warning of karst water layer pollution in coal mine goaf areas. Background Art

[0002] The North China coalfield refers to a combination of Ordovician limestone in the lower part and Carboniferous-Permian sandstone shale coal seams in the upper part. Coal mines will generate wastewater during mining, because our domestic water is taken from the Ordovician limestone in the lower part. Therefore, the wastewater in the upper coal mine goaf can easily infiltrate into the Ordovician limestone, polluting the high-quality groundwater and causing cross-layer pollution.

[0003] It is foreseeable that the cross-layer pollution of groundwater in coal fields will be a major hidden danger that will affect the safety of drinking water in the future. Previous researchers have used the extraction test of borehole water samples to identify groundwater pollution. There are two methods: fixed-depth sampling test in the borehole and intermittent automatic sampling in the borehole. Both methods require drilling first and then sampling in the borehole to identify cross-layer pollution through measured data. Although in-hole sampling can accurately monitor the occurrence of cross-layer pollution, the two methods have low applicability and belong to passive identification. The monitoring and maintenance costs are high, and the occurrence of cross-layer pollution cannot be predicted in time. Summary of the invention

[0004] In response to the above problems, the present invention provides a method for identifying and warning of karst water interlayer pollution in coal mine goaf areas. The method effectively reduces the cost of monitoring and early warning, and realizes efficient and comprehensive identification and early warning of karst water interlayer pollution in goaf areas.

[0005] The technical solution adopted by the present invention to solve the technical problem is:

[0006] A method for identifying and warning karst water layer pollution in coal mine goaf areas includes the following steps:

[0007] First, arrange monitoring points, conduct a unified measurement of the water level in the mining area to be measured, and compile a groundwater flow field map based on the unified measured water level data to determine the approximate flow direction of groundwater in the mining area to be measured;

[0008] Second, tracer tests and pumping tests are conducted to determine groundwater flow direction, hydrogeological parameters and hydraulic connections of aquifers;

[0009] Third, pollution identification and prediction;

[0010] 3.1 Arrange a control sampling point for sampling karst water upstream of the mining area;

[0011] 3.2 Sampling coal-bearing strata, karst water and goaf to obtain SO4 2- , pH, and conductivity background values;

[0012] 3.3 Compare the karst water test results in step 3.2 with the karst water test results of the control sampling points in step 3.1. If the groundwater test results taken in step 3.2 are higher than the karst water test results of the control sampling points in step 3.1, the sampling points are densely arranged and sampling monitoring is carried out;

[0013] 3.4 Based on the background value monitoring results measured in steps 3.2 and 3.3:

[0014] 3.4.1 If the karst water has similar background value characteristics to the mine water, cross-layer contamination may occur. In the numerical simulation process of step 7.8, the SO4 2- The concentration is used as the initial concentration to predict the cross-layer pollution;

[0015] 3.4.2 If the background value of karst water does not detect characteristics similar to mine water, the comprehensive water pollution index combined with the Spearman rank correlation coefficient method is used to analyze the trend of water quality changes;

[0016] Fourth, conduct borehole observations at different depths to obtain the hydrodynamic and hydrochemical characteristics of the groundwater system;

[0017] Fifth, observe the water quality and flow rate at the overflow point to obtain the water chemical composition of the discharge from the groundwater system.

[0018] Sixth, analyze the activation of fault zones and the development height of fracturing zones to find potential pollution channels.

[0019] 6.1 The influencing factors of fault zone activation in the mining area to be tested are determined by using grey correlation analysis.

[0020] 6.2 For the influencing factors of the first 1 / 2 of the grey correlation ranking, the SPSS principal component analysis method is used to analyze the activation law of the fault zone in the tested mining area and the possible activation area of ​​the fault zone in the tested mining area;

[0021] 6.3 Further determine whether the fracture is activated based on the possible activated fracture zone region obtained in step 6.2;

[0022] 6.4 Use empirical formulas and numerical simulation methods to predict the height of the fracture zone in the mining area to be tested;

[0023] Seventh, early warning of layer contamination;

[0024] 7.1 Establish a numerical simulation model based on the regional geological structure and hydrogeological conditions;

[0025] 7.2 Based on the numerical simulation model established in step 7.1, a conceptual model is established in combination with the geological structure and hydrogeological conditions of the area to be tested, and the aquifer and boundary conditions are generalized;

[0026] 7.3 Combine the groundwater measurement data and long-term observation well data in the study area, consider the annual and interannual changes of groundwater, and determine the model identification period;

[0027] 7.4 Based on the principle of parameter partitioning and indirect parameter calculation, the model parameters are debugged using the historical sequence data within the model identification period;

[0028] 7.5 The model verification period shall be determined based on the distribution of water level observation points in the simulation area and the continuity of observation data, and the model verification period shall not overlap with the model identification period;

[0029] 7.6 Using the model obtained in step 7.4, simulate the water level dynamic process curve within the model verification period to obtain the water level dynamic process simulation curve within the model verification period;

[0030] 7.7 Fit the water level dynamic process simulation curve within the model verification period obtained in step 7.6 with the water level dynamic process curve actually measured within the model verification period, thereby verifying the numerical simulation model obtained in step 7.4;

[0031] 7.8 Use the numerical simulation model obtained in step 7.7 to establish a solute transport model;

[0032] 7.9 Use the solute transport model obtained in step 7.8 to simulate sulfate ions, and use the simulation data of the observation well and the historical sequence data of the model identification period in step 7.3 to debug the parameters of the solute transport model;

[0033] 7.10 Use the solute transport model obtained in step 7.9 to simulate sulfate ions, and compare the simulated data of the observation well with the historical sequence data of the model verification period in step 7.5 to verify the solute transport model obtained in step 7.9;

[0034] 7.11 Simulate and predict the changes in groundwater flow field in the test area based on existing rainfall and mining data.

[0035] Further, the second step includes the following steps,

[0036] 2.1 Conduct tracer tests to determine the direction of groundwater flow;

[0037] 2.2 Calculate the conductivity T, permeability K and storage coefficient S of each aquifer;

[0038] 2.3 Determine the hydraulic connection between aquifers;

[0039] 2.3.1 According to the groundwater flow direction determined in step 2.1, arrange pumping wells in the pumping layer and arrange observation holes in the target aquifer;

[0040] 2.3.2 Conduct a steady flow pumping test to obtain the groundwater level drawdown S0 of the target aquifer observation hole;

[0041] 2.3.3 According to the plane position of the observation hole of the target aquifer, calculate the groundwater level drop S of the pumping layer at that position p ;

[0042] 2.3.4 Substitute the groundwater level drop S0 of the observation hole of the target aquifer obtained in step 2.3.2 and the groundwater level drop S of the pumping layer at this location obtained in step 2.3.3 p The division results in a ratio, through which the hydraulic connection between the pumping layer and the target aquifer can be quantitatively evaluated.

[0043] Furthermore, in step 2.3, when multiple pumping tests are conducted, the groundwater drawdown S of the pumping layer should be calculated using the data at the maximum drawdown of the pumping well. p .

[0044] Furthermore, when arranging pumping wells, the pumping wells should be away from the permeable and impermeable boundaries of the aquifer, and should be arranged as far as possible in places that are representative of the water conduction and storage properties, recharge conditions, thickness and lithology conditions of the aquifer, and the drainage conditions near the pumping wells should be good.

[0045] Furthermore, in the first step, the number of monitoring points is increased in karst water mines, areas prone to karst collapse, areas with groundwater funnel distribution, areas with large groundwater dynamic changes and complex recharge conditions.

[0046] Furthermore, in the first step, when conducting stratified monitoring, the aquifers to be detected should cover coal-bearing strata, goafs and karst water.

[0047] Further, the fourth step includes the following steps,

[0048] 4.1 Observation points are arranged in the direction of groundwater flow field and on both sides of the boundary line between the goaf and the strata exposed on the surface, and in-hole observations are carried out to draw the curve of conductivity variation with depth;

[0049] 4.2 According to the monitoring results of step 4.1, analyze the depth of the point where the conductivity changes suddenly, and collect water samples at the specified depth according to the conductivity change point to detect pH, SO4 2- Concentration and TDS, combined with the Piper three-line diagram, are used to study the groundwater chemical type and regional characteristics of the mining area.

[0050] Furthermore, in step 3.2, when arranging sampling points in the goaf, the goafs in the same hydrogeological unit are sampled at a rate of 10 km. 2The sampling points were arranged at a density of 4-6, located in the goaf of different hydrogeological units, and each unit was sampled at a rate of 10 km 2 The sampling points are arranged at a density of 3-4 sampling points, among which 3-5 sampling points are arranged at the mine drainage outlet according to the drainage volume, and 2-6 sampling points are arranged at the groundwater overflow points in each goaf;

[0051] When arranging sampling points for coal-bearing strata, the groundwater aquifer system within the mining area is calculated at a rate of 10 km 2 The sampling points are arranged at a density of 2-6 sampling points, of which 2-6 sampling points are arranged in the mining area and on both sides of the water-conducting fracture;

[0052] When arranging karst water sampling points, 2-6 sampling points are arranged where the groundwater and surface water are hydraulically closely connected. Along the flow direction of groundwater, at least one sampling point is arranged downstream and on both sides of the goaf.

[0053] The beneficial effects of the present invention are:

[0054] 1. Groundwater pollution control and restoration are difficult. Some people even proposed that groundwater pollution restoration is inaccessible, that is, some pollution cannot be repaired within our life cycle. Therefore, early warning of pollution is far more urgent than passive identification of pollution. This method can not only actively identify layer pollution, but also warn of the development trend of future pollution, providing a scientific basis for accurate remediation.

[0055] 2. This method effectively reduces the cost of monitoring and early warning, and can achieve efficient and comprehensive identification and early warning of karst water layer pollution in goaf areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is the technical flow chart of this method;

[0057] Figure 2 Map of monitoring points layout for mining areas in the central and southern Shandong basin;

[0058] Figure 3 This is a schematic diagram of the hydraulic connection channel;

[0059] Figure 4 This is a three-dimensional map of the simulated area of ​​the Central and Southern Shandong Basin;

[0060] Figure 5 This is a cross-section of the simulation area in the central and southern Shandong Basin;

[0061] Figure 6 This is the boundary condition map of the simulation area of ​​the Central and Southern Shandong Basin;

[0062] Figure 7 This is the karst water flow field map of the simulation area after 5 years;

[0063] Figure 8 The SO4 in the karst water of the simulated area after 5 years 2- Ion concentration distribution contours;

[0064] Fig. 9 The SO4 in the karst water of the simulated area after 5 years 2- Ion concentration distribution profile;

[0065] Fig.10 This is the karst water flow field diagram of the simulation area after 20 years;

[0066] Fig.11 The SO4 in the karst water of the simulated area after 20 years 2- Ion concentration distribution contours;

[0067] Fig.12 The SO4 of karst water in the simulated area after 20 years 2- Ion concentration distribution profile contour plot. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the technical solution of the present invention is further described in detail below by taking a basin in south central Shandong as an example.

[0069] A method for identifying and warning karst water layer pollution in coal mine goaf areas comprises the following steps:

[0070] First, arrange monitoring points to measure the water level in the mining area to be tested, and compile a groundwater flow field map based on the measured water level data. By analyzing the flow field, the approximate flow direction of groundwater in the mining area to be tested can be determined.

[0071] As a specific implementation method, in this embodiment, 2-6 water level monitoring points are arranged at the places where the groundwater and surface water are closely connected hydraulically; 2-6 water level monitoring points are arranged at the groundwater overflow points in each goaf; and the groundwater multi-aquifer system within the mining area is monitored every 10km 2 Arrange 2-6 water level monitoring points; arrange 2-6 water level monitoring points in the mining area and on both sides of the water-conducting fracture.

[0072] Furthermore, monitoring points should be appropriately increased in karst water mines, areas prone to karst collapse, areas with groundwater funnel distribution, areas with large groundwater dynamic changes and complex recharge conditions.

[0073] Among the above monitoring points, the monitoring frequency of the monitoring points located upstream and at the boundary of the mining area is 12 times / year, the monitoring frequency of the monitoring points located downstream of the mining area is 24 times / year, and the monitoring frequency of the monitoring points located inside the mining area is 36 times / year. According to the monitoring data, automatic monitoring is adopted for the monitoring points with large water level fluctuations. The rest are monitored manually, and the time interval between the two monitorings on both sides is the same. During the water level monitoring process, the measurement accuracy of the water level shall not be less than 1mm.

[0074] Furthermore, when conducting stratified monitoring, the aquifers to be tested should include coal-bearing strata, goaf and karst water. That is, during the drilling process, by placing the filter pipe at the target layer, the water levels of coal-bearing strata, goaf and karst water are monitored respectively, thus achieving stratified monitoring of the water level.

[0075] Through the above method, the water level of the mining area to be tested is measured in a unified way, which can not only provide data for the calculation of the mathematical model of the mining area to be tested, but also compile the groundwater flow field map within a certain period of time according to the water level data of the unified measurement, and judge the general flow direction of the groundwater in the mining area to be tested by analyzing the flow field, so as to provide a basis for the selection of the tracer test points in the next step. At the same time, the formation and position of the groundwater drop funnel can be analyzed according to the flow field, and the occurrence and position of the cross-layer pollution can be preliminarily judged, so as to reversely test the reliability of the prediction results of the model of the mining area to be tested.

[0076] Taking a basin in central and southern Shandong as an example, there are currently many mining areas in the basin, the largest of which covers an area of ​​about 10km 2 According to the above requirements and the actual situation of the basin in central and southern Shandong, monitoring points are set up as follows: Figure 2 As shown, some of the water level monitoring points in the figure can also be used as water quality monitoring points.

[0077] Second, tracer tests and pumping tests are conducted to determine groundwater flow direction, hydrogeological parameters and hydraulic connections of aquifers.

[0078] 2.1 Conduct tracer tests to determine the direction of groundwater flow.

[0079] According to the possible flow direction of groundwater obtained in the first step, the source point and monitoring point are selected. When monitoring the monitoring points, the sampling order is from near to far, and the monitoring frequency can be appropriately increased according to the monitoring results.

[0080] Here, the following must be considered when selecting tracers: ① The tracer does not have much impact on groundwater quality; ② It is easy to detect and the detection sensitivity must be high; ③ It is stable within a certain period of time and not easily adsorbed and filtered by rocks.

[0081] Furthermore, in order to determine the flow direction of groundwater in the mining area to be tested, this embodiment adopts the instantaneous injection method for source injection, that is, a known amount of tracer is injected into the source water in the shortest time. The concentration-time penetration curve is established through continuous observation or continuous water sample test analysis, and the structure of the water flow channel is reflected according to the shape of the curve: a single peak indicates the existence of a single pipeline; a multi-peak curve indicates that there may be multiple water flow channels. The tracer directly enters the storage position when passing through the pipeline. The storage part is a dead-end cavity or a series cavity. The tracers stored in these positions are driven by the changing water flow, resulting in the sudden disappearance or natural appearance of the tracer. According to the above situation, the flow direction of groundwater in the mining area to be tested is further finely portrayed.

[0082] 2.2 Calculate the conductivity T, permeability K and storage coefficient S of each aquifer.

[0083] 2.2.1 According to the groundwater flow direction determined in step 2.1, arrange pumping wells and observation holes in a certain aquifer. When arranging observation holes, for homogeneous aquifers, arrange 1-3 observation holes perpendicular to the groundwater flow direction and upstream and downstream; for heterogeneous aquifers, arrange 1-3 observation holes along the directions of different water storage and water conduction properties; for fissures or karst fissures, arrange 1-3 observation holes in the water-rich section, upstream and downstream of the groundwater, and in the section with weak water-richness.

[0084] 2.2.2 Pump water with a constant flow rate and observe until the water level recovers. Obtain the depth drop of the water level observation hole and the corresponding time data. By simplifying the well function, the following formula can be obtained:

[0085] Y=ts

[0086]

[0087]

[0088]

[0089] 2.2.3 Draw the straight line Y=AX+B with Y as the ordinate and X as the abscissa, where A is the slope and B is the intercept. Based on this, we can get the values ​​of A and B:

[0090]

[0091]

[0092]

[0093] Where: T is the hydraulic conductivity, (L 2 T -1 );

[0094] s is the water level drawdown in the observation well, (L);

[0095] r is the distance from the observation hole to the pumping well, (L);

[0096] t is the time corresponding to the drop in water level in the observation well, (T);

[0097] M is the effective aquifer thickness, (L);

[0098] Q is the pumping flow rate, (L 3 T -1 );

[0099] t p is the duration of pumping before stopping pumping, (T);

[0100] S is the water storage coefficient;

[0101] K is the permeability coefficient, (LT -1 ).

[0102] According to the above method, the water conductivity T, water storage coefficient S and permeability coefficient K of a certain aquifer in the mining area to be tested can be solved.

[0103] 2.2.4 Repeat steps 2.2.1-2.2.3 to determine the water conductivity T, water storage coefficient S and permeability K of each aquifer in the mining area to be tested, providing a data source for early warning of cross-layer pollution in coal mine goaf areas.

[0104] 2.3 Determine the hydraulic connection between aquifers.

[0105] 2.3.1 According to the groundwater flow direction determined in step 2.1, pumping wells are arranged in the pumping layer, and observation holes for the target aquifer are arranged in the target aquifer. When arranging the observation holes for the target aquifer, for homogeneous aquifers, 1-3 observation holes for the target aquifer are arranged perpendicular to the groundwater flow direction and upstream and downstream; for heterogeneous aquifers, 1-3 observation holes for the target aquifer are arranged along the directions of different water storage and water conductivity properties; for fissures or karst fissures, 1-3 observation holes for the target aquifer are arranged in the water-rich section, upstream and downstream of the groundwater, and in the section with weak water-richness.

[0106] 2.3.2 Conduct steady flow pumping test, such as Figure 3 As shown, the groundwater level drop S0 of the target aquifer observation hole is obtained.

[0107] 2.3.3 According to the plane position of the observation hole of the target aquifer, calculate the groundwater level drop S of the pumping layer at that position p , the calculation formula is:

[0108]

[0109] Where: Q pumping well water volume, (L 3 T-1 );

[0110] M is the thickness of the pumping layer, (L);

[0111] S p is the drawdown of the groundwater level in the pumping layer at that location, (L);

[0112] r1 is the radius of the pumping well, (L);

[0113] r w is the plane distance between the pumping well and the observation hole, (L);

[0114] K is the permeability coefficient of the pumping layer, (L); this parameter has been obtained in step 2.2.

[0115] 2.3.4 Substitute the groundwater level drop S0 of the observation hole of the target aquifer obtained in step 2.3.2 and the groundwater level drop S of the pumping layer at this location obtained in step 2.3.3 p The division results in a ratio, through which the hydraulic connection between the pumping layer and the target aquifer can be quantitatively evaluated. The hydraulic connection coefficient calculation formula is as follows:

[0116]

[0117] Where: S0 is the water level drawdown of the observation hole of the target aquifer, (L);

[0118] S p is the water level drop of the pumping layer at that location, (L).

[0119] Here, the pumping layer and the target aquifer can be the same aquifer or different aquifers. When they are the same aquifer, the hydraulic connection coefficient can reflect the hydraulic connection between boreholes in different directions and at different distances in the aquifer; when they are different aquifers, the hydraulic connection coefficient can reflect the vertical hydraulic connection between the pumping layer and the target aquifer. In addition, when multiple drawdown pumping tests are conducted, the data at the maximum drawdown of the pumping well should be used to calculate the groundwater drawdown S of the pumping layer at that location. p .

[0120] The hydraulic connection coefficient C is calculated according to the above formula, where the closer the C value is to 0, the weaker the hydraulic connection between the aquifers, and the closer the C value is to 1, the stronger the hydraulic connection between the aquifers. The above method can effectively judge the vertical hydraulic connection between the goaf, coal-bearing strata, and karst water. Water quality exchange may occur between two aquifers with hydraulic connection, thus causing cross-layer pollution.

[0121] Furthermore, in step 2.2.1 and step 2.3.1, when arranging the pumping wells, the pumping wells should be far away from the permeable and water-proof boundaries of the aquifer, and should be arranged as far as possible in places where the water conduction and water storage properties, recharge conditions, thickness and lithology conditions of the aquifer are representative. Here, the existing observation wells can be used as pumping wells, and the drainage conditions near the pumping wells should be good, that is, the pumped water can be discharged outside the influence radius of the pumping well without leakage.

[0122] Furthermore, before conducting a pumping test, the corresponding water output at the maximum water level drop should be estimated in order to select a suitable water pump. The maximum water output can be estimated based on the water output of existing wells in the same aquifer, or calculated based on the empirical permeability coefficient of the aquifer and the designed water level drop, or determined based on the water volume during well washing.

[0123] Furthermore, in step 2.2.2, the water level of the pumping well, the water output and the water level of each observation hole should be observed at the same time after the pumping test begins. The water level observation time is generally once at the 1st minute, 3 minutes, 5 minutes, 7 minutes, 10 minutes, 15 minutes, 20 minutes, 30 minutes, 45 minutes, 60 minutes and 90 minutes after the start of pumping, and then once every 30 minutes. After 5 hours, observe once every 1 hour, and the water level reading is accurate to centimeters. From the moment of stopping pumping, water level recovery observations are carried out at the time intervals of the above-mentioned pumping process until the water level returns to the natural water level. In step 2.3.2, the water level of the observation hole of the target aquifer is generally observed once every 30 minutes, and can be appropriately increased according to its water level response time.

[0124] Third, pollution identification and prediction.

[0125] 3.1 A control sampling point for sampling karst water is arranged upstream of the mining area. As a specific implementation method, in this embodiment, 2-4 water quality background value monitoring points are arranged upstream of each mining area.

[0126] 3.2 Sampling coal-bearing strata, karst water and goaf to obtain SO4 2- , pH, and conductivity background values.

[0127] Arrangement of sampling points in goaf areas: goaf areas in the same hydrogeological unit are sampled every 10km. 2 The sampling points are arranged at a density of 4-6 sampling points; they are located in the goaf of different hydrogeological units, and each unit is sampled at a density of 10km 2 The sampling points are arranged at a density of 3-4 sampling points. Among them, 3-5 sampling points are arranged at the mine drainage outlet according to the drainage volume; 2-6 sampling points are arranged at the groundwater overflow points in each goaf.

[0128] Arrangement of sampling points for coal-bearing strata: The groundwater aquifer system within the mining area is divided into several layers every 10km. 2 The sampling points are arranged at a density of 2-6 sampling points. Among them, 2-6 sampling points are arranged in the mining area and on both sides of the water-conducting fracture.

[0129] Arrangement of karst water sampling points: 2-6 sampling points are arranged where the groundwater and surface water are hydraulically closely connected; along the flow direction of groundwater, at least one sampling point is arranged downstream and on both sides of the goaf.

[0130] 3.3 Compare the karst water test results in step 3.2 with the karst water test results of the control sampling points in step 3.1. If the groundwater test results taken in step 3.2 are higher than the karst water test results of the control sampling points in step 3.1, the sampling points are densely arranged and sampling monitoring is carried out.

[0131] 3.4 Speculate on the possibility of cross-layer contamination.

[0132] The mine water in the goaf shows high conductivity, slight acidity and high SO4 2- Concentration characteristics, based on the background value monitoring results measured in steps 3.2 and 3.3:

[0133] 3.4.1 If the karst water has a background value characteristic similar to that of the mine water, then cross-layer contamination occurs. During the numerical simulation in step 7.8, the SO4 2- The concentration is used as the initial concentration for cross-layer pollution prediction.

[0134] 3.4.2 If the background value of karst water does not detect characteristics similar to mine water, the Spearman rank correlation coefficient method is used to predict the trend of water quality changes.

[0135] Monitor the sampling points, the monitoring sequence is Y1, Y2, ... Y n , the corresponding SO4 2- The concentration values ​​are C1, C2, ... C n , the sequence number X is sorted in order of concentration i , the rank correlation coefficient r of this set of values ​​is calculated by the following formula s :

[0136]

[0137] d i =X i -Y i

[0138] Where, d i For variable X i and Y i The difference between

[0139] X i is the sequence number of periods i to N arranged from small to large in terms of concentration;

[0140] Y i is the time-ordered sequence number,

[0141] N is the time period.

[0142] The water quality pollution comprehensive index combined with the Spearman rank correlation coefficient method is used to analyze the water quality change trend. The calculation formula of the water quality pollution comprehensive index is:

[0143]

[0144] In the formula, C i is the measured content of water quality factor, (ML -3 );

[0145] C oi The maximum permissible standard for groundwater drinking water (ML -3 ).

[0146] The rank correlation coefficient is used to analyze the trend of the comprehensive index of groundwater pollution, so as to determine the trend of groundwater quality change. s The absolute value of is the same as the critical value w in the Spearman rank correlation coefficient statistics table p In comparison.

[0147] If |r s |≥w p (95% confidence level), and r s If it is a negative value, it means that the change trend is significant, pollution is decreasing, and water quality is improving;

[0148] If |r s |≥w p (95% confidence level), and r s It is a positive value, indicating that the trend of change is significant, pollution is on the rise, and water quality is on the deterioration trend;

[0149] If |r s | <w p , indicating that the water quality has no obvious changes and remains basically stable.

[0150] If the prediction results show that the karst water quality is deteriorating, that is, SO4 2- The concentration change shows the same trend as that of mine water, which can also be used as a basis for judging the occurrence of cross-layer pollution. The reliability of the model in step 7.8 can be tested based on the above prediction results, and it can also be used as a basis for early prevention and control of cross-layer pollution.

[0151] Fourth, conduct in-hole observations at different depths to obtain the hydrodynamic and hydrochemical characteristics of the groundwater system.

[0152] 4.1 Observation points were arranged in the goaf, coal-bearing strata and Ordovician strata along the direction of groundwater flow field, and in-hole observations were carried out to draw the curve of conductivity variation with depth.

[0153] The monitoring frequency of the observation points located in the goaf is once every five days, and the monitoring frequency of the other observation points is once every ten days.

[0154] Furthermore, the monitoring frequency should be appropriately increased during the flood season, dry season and rainy season.

[0155] During observation, the instrument cable is marked every one meter, and the instrument is lowered accordingly. Every five meters, the data is recorded after one minute of stabilization. At the observation points on both sides of the stratum boundary and in the goaf, the monitoring is intensified, and data is recorded every two meters.

[0156] 4.2 According to the monitoring results of step 4.1, analyze the depth of the point where the conductivity changes suddenly, and collect water samples at the specified depth according to the conductivity change point to detect pH, SO4 2- Concentration and TDS, combined with the Piper three-line diagram, are used to study the groundwater chemical type and regional characteristics of the mining area.

[0157] Fifth, observe the water quality and flow rate at the overflow point to obtain the water chemical composition of the discharge from the groundwater system.

[0158] Overflow points in mining areas are mainly divided into mine water overflow and karst water overflow. Through field surveys, we find the overflow points of abandoned mines, take samples and monitor the overflow points, and obtain samples for simple analysis to determine the chemical composition of the discharge volume of the groundwater system. The analysis items include: K + 、Na + , Ca 2+ Mg 2+ NH4 + , Cl - 、SO4 2- 、HCO3 3- 、CO3 2- OH - 、F - 、NO2 - 、NO3 - , total hardness, dissolved solids, pH, and the water chemistry type is determined by the Shukarev classification method. The detection accuracy of anions, cations and total dissolved solids is 0.01 mg / L, and the detection accuracy of total hardness is 0.01 mg / L.

[0159] Among them, for some overflow points with large water output, the monitoring frequency is once a day, and observation is once every two hours if necessary.

[0160] Based on the test results of water samples from the overflow point, monitoring of overflow points with large changes in water quality will be intensified, with a frequency of once every two days.

[0161] The monitoring frequency of other overflow points is once every 10 days, and monitoring is intensified during the flood season, dry season and rainy season.

[0162] Sixth, analyze the activation of fault zones and the development height of fracturing zones to find potential pollution channels.

[0163] 6.1 The influencing factors of fault zone activation in the mining area to be tested are determined by using grey correlation analysis.

[0164] To obtain the correlation result, the correlation coefficient must be calculated first:

[0165]

[0166] Where, L oi (k) is the correlation coefficient;

[0167] Δ oi (k) is the absolute difference between the two sequences at time k;

[0168] Δ min is the minimum absolute difference;

[0169] Δ max is the maximum value;

[0170] ρ is the resolution coefficient.

[0171] The correlation coefficient reflects the closeness of two sequences at a certain moment, ranging from 0 to 1, and then the grey correlation degree is calculated. The formula is as follows:

[0172]

[0173] In the formula, r oi is the grey relational degree, ranging from 0 to 1;

[0174] N is the number of sequences.

[0175] where r oi The closer it is to 1, the higher the correlation between the two is, and finally the grey correlation of each influencing factor is obtained.

[0176] 6.2 For the influencing factors in the first 1 / 2 of the grey correlation ranking, the SPSS principal component analysis method is used. The factors whose variance and correlation coefficient are closer to 1 have a higher influence on the activation of the fault zone. Based on this, the activation law of the fault zone in the tested mining area is analyzed, and the areas where the fault zone in the tested mining area may be activated are analyzed.

[0177] 6.3 Based on the possible activated fracture zone area obtained in step 6.2, further determine whether the fracture is activated.

[0178] The specific confirmation method for fault activation requires observation within the tunnel. For tunnels that pass through the fault zone, holes are drilled inside the tunnels on both sides of the fault zone to measure the drainage volume and water quality changes. The monitoring frequency is once every five days. Based on the drainage volume and water quality changes in the tunnels in the tested mining area, it is determined whether water inrush has occurred in the tunnel, and whether the fault zone is activated is further confirmed.

[0179] 6.4 The height of the fracture zone is not only related to the mining thickness, but also to many factors such as the mining area, overburden lithology, coal seam inclination, coal mining method, etc. Based on these factors, the empirical formula summarized by predecessors is used to predict the height of the fracture zone. The applicable empirical formulas in different scenarios are as follows:

[0180] Coal seam inclination 0-54°, ​​rock saturated uniaxial compressive strength 40-80Mpa:

[0181]

[0182] Coal seam inclination 0-54°, ​​rock saturated uniaxial compressive strength 20-40Mpa:

[0183]

[0184] Coal seam inclination 0-54°, ​​rock saturated uniaxial compressive strength 10-20Mpa:

[0185]

[0186] Coal seam inclination 0-54°, ​​rock saturated uniaxial compressive strength less than 10Mpa:

[0187]

[0188] Coal seam inclination 55-90°, rock saturated uniaxial compressive strength 10-80Mpa:

[0189] H m =(0.4-0.5)H li

[0190] where ∑ M Cumulative mining thickness (L); the single layer mining thickness is 1-3m, and the cumulative mining thickness does not exceed 15m.

[0191] The above method can effectively determine the activation of fault zones and the generation of fracturing zones. For the mine water to cross-layer and pollute the karst water, there must be a certain channel to connect the two, forming the so-called pollution channel, in which the pollution channel includes the activated fault zone and fracturing zone. Through the analysis of the first, fourth and fifth steps above, if the karst water has been polluted or the mine water level is low and the recharge and discharge conditions have no obvious impact on the karst water, then the fault zone and fracturing zone are determined to be potential cross-layer pollution channels. Identifying pollution channels can serve as the basis for identifying the numerical model of cross-layer pollution. At the same time, grouting can be used to control cross-layer pollution according to the potential pollution channels.

[0192] Seventh, early warning of cross-layer pollution in different scenarios

[0193] 7.1 Establish a numerical simulation model based on the regional geological structure and hydrogeological conditions.

[0194] (1) When the aquifer in the test area is a homogeneous aquifer of equal thickness, the boundary is an impermeable boundary or the water head is known, and there is no vertical recharge and discharge, the one-dimensional steady flow formula can be used. The analytical solution is used to test and compare with the numerical solution, and the analytical solution is used to fit the observed data to obtain the hydrodynamic coefficient. The one-dimensional groundwater flow equation is:

[0195]

[0196] In the formula, K x is the permeability coefficient in the x direction, (LT -1 );

[0197] S is the water storage coefficient;

[0198] H is the water head height, (L).

[0199] The one-dimensional diffusion equation is:

[0200]

[0201] Where c(x,t) is the concentration of dissolved solute, (ML -3 );

[0202] D x is the longitudinal diffusion coefficient, D x =α L v x ,(L 2 T -1 );

[0203] α L is the longitudinal dispersivity, (L);

[0204] V x is the seepage velocity, (LT -1 );

[0205] t is the total simulation time, (T).

[0206] The analytical solution of the one-dimensional diffusion equation is:

[0207]

[0208] (2) When the groundwater flow direction in the test area is mainly horizontal flow, the aquifer can be generalized as a heterogeneous anisotropic confined aquifer with a water-proof boundary at the bottom, and the vertical water exchange is ignored, the two-dimensional unsteady flow formula can be used to solve it using the numerical method. The two-dimensional groundwater flow equation is:

[0209]

[0210] Where: T is the water conductivity in the main direction, (L 2 T -1 );

[0211] S is the water storage coefficient;

[0212] H is the water head height, (L).

[0213] The two-dimensional diffusion equation is:

[0214]

[0215] Where: C is the solute concentration in the solution, (ML -3 );

[0216] t is the time variable, (T);

[0217] D xx , D xy , D yy is the diffusion coefficient in each direction, (L 2 T -1 );

[0218] V x 、V y is the actual flow velocity in the x and y directions, (LT -1 );

[0219] f is the source and sink term, (L 3 T -1 ).

[0220] (3) When the area to be measured can be generalized as a multi-layer vertical heterogeneous aquifer, the hydrogeological conditions are relatively complex and contain multiple types of boundary conditions and source and sink terms, the three-dimensional unsteady flow formula can be used and solved using the finite element or finite difference method. The three-dimensional groundwater flow equation is:

[0221]

[0222] Where: K xx , K yy , K zz are the main permeability coefficients of the anisotropic medium, (LT -1 );

[0223] S S is the water storage coefficient;

[0224] H is the water head height, (L);

[0225] W is the source and sink term, (L 3 T -1 ).

[0226] The three-dimensional diffusion equation is:

[0227]

[0228] Where: C k is the concentration of solute k in groundwater, (ML -3 );

[0229] n is the porosity;

[0230] D ij is the diffusion coefficient, (L 2 T -1 );

[0231] ∑R n is the source and sink term, (L 3 T -1 );

[0232] C ks is the concentration of the source and sink term of substance k, (ML -3 );

[0233] q s is the flow rate per unit volume of fluid in the aquifer, (L 3 T -1 );

[0234] v i is the seepage rate, (LT -1 ).

[0235] As a specific implementation method, in this embodiment, the coal mining history in the basin is long, there are multiple goafs and multiple water-bearing rock groups, and the water level geological conditions are complex, including: Quaternary loose rock porous water-bearing rock group, coal-bearing strata overlying clastic rock porous fissure water-bearing rock group, multiple Carboniferous-Permian fissure water-bearing rock group, multiple coal seams in the old hole and tunnel water-filled rock group is the coal-bearing stratum water-bearing rock group, Cambrian-Ordovician carbonate rock fissure karst water-bearing rock group. Therefore, a three-dimensional numerical model is used for calculation, and the mathematical model of this area is given as follows:

[0236] Three-dimensional unsteady flow model:

[0237]

[0238] Where: Ω is the simulation area;

[0239] Γ2 is the flow boundary of the simulation area.

[0240] K xx , K yy , K zz are the main permeability coefficients of the anisotropic medium, (LT -1 );

[0241] H is the water head height, (L).

[0242] Solute transport model:

[0243]

[0244] Where: C k is the concentration of solute k in groundwater, (ML -3 );

[0245] n is the porosity;

[0246] D ij is the diffusion coefficient, (L 2 T -1 );

[0247] ∑R n is the source and sink term, (L 3 T -1 );

[0248] C ks is the concentration of the source and sink term of substance k (ML -3 );

[0249] q s is the flow rate per unit volume of fluid in the aquifer, (L 3 T -1 );

[0250] v i is the seepage rate, (LT -1 ).

[0251] C0 is the concentration of the source and sink term, (ML -3 );

[0252] Γ2 simulation area boundary;

[0253] C2(x, y, z) is the known concentration on the second type boundary, (ML -3 ).

[0254] 7.2 Based on the numerical simulation model established in step 7.1, a conceptual model is established in combination with the geological structure and hydrogeological conditions of the area to be tested, and the aquifer and boundary conditions are generalized.

[0255] As a specific implementation method, since the numerical simulation model of Taozao Basin in step 7.1 is calculated using a three-dimensional unsteady flow model, the aquifer generalization and boundary condition generalization results are as follows: Figure 4 , Figure 5 and Figure 6 shown.

[0256] 7.3 Combine the groundwater measurement data and long-term observation well data in the study area, consider the annual and interannual changes of groundwater, and determine the model identification period.

[0257] As a specific implementation method, the model identification period determined in this embodiment is from 2010 / 6 / 1 to 2017 / 12 / 31, a total of 2770 days. Each month is a stress period, a total of 91 stress periods. The groundwater system recharge and discharge items remain unchanged in the same stress period, and each stress period has 1 time step.

[0258] 7.4 Based on the principle of parameter partitioning and indirect parameter calculation, the model parameters are debugged using the historical sequence data within the model identification period.

[0259] That is, according to the observation values ​​of each observation well during the model identification period, the parameters (including permeability coefficient, water supply degree, water storage coefficient, etc.) are repeatedly modified and the source and sink items are adjusted for fitting, so as to realize the inversion of hydrogeological parameters and some source and sink items.

[0260] 7.5 The model verification period shall be determined based on the distribution of water level observation points in the simulation area and the continuity of the observation data, and the model verification period shall not overlap with the model identification period.

[0261] As a specific implementation, the model verification period determined in this embodiment is from January 1, 2018 to December 31, 2020.

[0262] 7.6 Using the model obtained in step 7.4, simulate the dynamic process curve of the water level during the model verification period to obtain the simulated dynamic process curve of the water level during the model verification period.

[0263] 7.7 The simulated curve of the water level dynamic process during the model verification period obtained in step 7.6 is fitted with the measured water level dynamic process curve during the model verification period, so as to verify the numerical simulation model obtained in step 7.4.

[0264] If the degree of fit between the two is high, it means that the numerical simulation model obtained in step 7.4 is accurate; if the degree of fit between the two is low, repeat steps 7.3-7.7 until an accurate numerical simulation model is obtained.

[0265] 7.8 Use the numerical simulation model obtained in step 7.7 to establish a solute transport model.

[0266] In this example, the pollutants in the mine water in the basin in the central and southern part of Shandong Province are mainly sulfate, so the solute transport simulation is carried out with sulfate ions. The pollution source location, pollutant sulfate ion concentration, and cross-layer pollution location are determined in the simulation area, and different initial concentrations, background values, total porosity, fixed porosity, and diffusion coefficients are assigned respectively.

[0267] 7.9 Identification of Solute Transport Models

[0268] The solute transport model obtained in step 7.8 is used to simulate sulfate ions, and the parameters of the solute transport model are debugged using the simulated data of the observation well and the historical sequence data of the model identification period in step 7.3.

[0269] 7.10 Verification of the Solute Transport Model

[0270] The solute transport model obtained in step 7.9 is used to simulate sulfate ions, and the simulated data of the observation wells are compared with the historical sequence data of the model verification period in step 7.5 to verify the solute transport model obtained in step 7.9.

[0271] If the simulation data of the observation well is close to the historical sequence data of the model verification period, it means that the solute transport model obtained in step 7.9 is accurate; if the simulation data of the observation well is not close to the historical sequence data of the model verification period, repeat steps 7.9-7.10 until the fitting results of the solute transport model in steps 7.9 and 7.10 meet the requirements, that is, the final solute transport model is obtained.

[0272] 7.11 Simulate and predict the changes in groundwater flow field in the test area based on existing rainfall and mining data.

[0273] As a specific implementation, this example simulates and predicts the changes in groundwater flow field and sulfate ion concentration field in the basin in central and southern Shandong after 5 and 20 years. Based on the established numerical simulation model, the sulfate ion concentration in the test area is simulated and predicted for 5 and 20 years.

[0274] (1) Changes in groundwater flow field and sulfate concentration in the simulation area after 5 years ( Figure 7 , Figure 8 , Fig. 9), the karst water of Dongcheng’s domestic water source is affected by the overlying coal mine goaf area’s mine water layer, and the sulfate ion concentration in the groundwater gradually increases to 127 mg / L. The maximum sulfate ion concentration in the northwest Bei-Liu area is 754 mg / L.

[0275] (2) Changes in groundwater flow field and sulfate concentration in the simulation area after 20 years ( Fig.10 , Fig.11 , Fig.12 ), the karst water of Dongcheng’s domestic water source is affected by the overlying coal mine goaf old pit water layer, the sulfate ion concentration in the groundwater gradually increases to 146 mg / L, and the maximum sulfate ion concentration in the northwest Bei-Liu area is 1152 mg / L.

Claims

1. A method for identifying and warning karst water layer pollution in coal mine goaf areas, characterized by: The following steps are included: First, arrange monitoring points, conduct a unified measurement of the water level in the mining area to be measured, and compile a groundwater flow field map based on the unified measured water level data to determine the approximate flow direction of groundwater in the mining area to be measured; Increase the number of monitoring points in karst water mines, areas prone to karst collapse, areas with groundwater funnel distribution, areas with large groundwater dynamic changes and complex recharge conditions; When conducting stratified monitoring, the aquifers to be tested should include coal-bearing strata, goafs and karst water; Second, tracer experiments and pumping tests were conducted to determine groundwater flow direction, hydrogeological parameters and hydraulic connections of aquifers; Third, pollution identification and prediction; 3.1 Arrange a control sampling point for sampling karst water upstream of the mining area; 3.2 Sampling coal-bearing strata, karst water and goaf to obtain SO4 2- , pH, and conductivity background values; 3.3 Compare the karst water test results in step 3.2 with the karst water test results of the control sampling points in step 3.

1. If the groundwater test results taken in step 3.2 are higher than the karst water test results of the control sampling points in step 3.1, the sampling points are densely arranged and sampling monitoring is carried out; 3.4 Based on the background value monitoring results measured in steps 3.2 and 3.3: 3.4.1 If the karst water has similar background value characteristics to the mine water, cross-layer contamination may occur. In the numerical simulation process of step 7.8, the SO4 2- The concentration is used as the initial concentration to predict the cross-layer pollution; 3.4.2 If the background value of karst water does not detect characteristics similar to mine water, the comprehensive water pollution index combined with the Spearman rank correlation coefficient method is used to analyze the trend of water quality changes; Fourth, conduct borehole observations at different depths to obtain the hydrodynamic and hydrochemical characteristics of the groundwater system; Fifth, observe the water quality and flow rate at the overflow point to obtain the water chemical composition of the discharge from the groundwater system; Sixth, analyze the activation of fault zones and the height of the fracturing zones to find potential contamination channels; 6.1 Determine the influencing factors of fault zone activation in the mining area to be tested by using grey correlation analysis; 6.2 For the influencing factors of the first 1 / 2 of the grey correlation ranking, the SPSS principal component analysis method is used to analyze the activation law of the fault zone in the tested mining area and the possible activation area of ​​the fault zone in the tested mining area; 6.3 Further determine whether the fracture is activated based on the possible activated fracture zone region obtained in step 6.2; 6.4 Use empirical formulas and numerical simulation methods to predict the height of the fracture zone in the mining area to be tested; Seventh, early warning of layer contamination; 7.1 Establish a numerical simulation model based on the regional geological structure and hydrogeological conditions; 7.2 Based on the numerical simulation model established in step 7.1, a conceptual model is established in combination with the geological structure and hydrogeological conditions of the area to be tested, and the aquifer and boundary conditions are generalized; 7.3 Combine the groundwater measurement data and long-term observation well data in the study area, consider the annual and interannual changes of groundwater, and determine the model identification period; 7.4 Based on the principle of parameter partitioning and indirect parameter calculation, the model parameters are debugged using the historical sequence data within the model identification period; 7.5 The model verification period shall be determined based on the distribution of water level observation points in the simulation area and the continuity of observation data, and the model verification period shall not overlap with the model identification period; 7.6 Using the model obtained in step 7.4, simulate the water level dynamic process curve within the model verification period to obtain the water level dynamic process simulation curve within the model verification period; 7.7 Fit the water level dynamic process simulation curve within the model verification period obtained in step 7.6 with the water level dynamic process curve actually measured within the model verification period, thereby verifying the numerical simulation model obtained in step 7.4; 7.8 Use the numerical simulation model obtained in step 7.7 to establish a solute transport model; 7.9 Use the solute transport model obtained in step 7.8 to simulate sulfate ions, and use the simulation data of the observation well and the historical sequence data of the model identification period in step 7.3 to debug the parameters of the solute transport model; 7.10 Use the solute transport model obtained in step 7.9 to simulate sulfate ions, and compare the simulated data of the observation well with the historical sequence data of the model verification period in step 7.5 to verify the solute transport model obtained in step 7.9; 7.11 Simulate and predict the changes in groundwater flow field in the test area based on existing rainfall and mining data.

2. A method for identifying and warning karst water layer pollution in coal mine goaf areas according to claim 1, characterized in that: The second step includes the following steps, 2.1 Conduct tracer tests to determine the direction of groundwater flow; 2.2 Calculate the water conductivity T, permeability K and storage coefficient S of each aquifer; 2.3 Determine the hydraulic connection between aquifers; 2.3.1 According to the groundwater flow direction determined in step 2.1, arrange pumping wells in the pumping layer and arrange observation holes in the target aquifer; 2.3.2 Conduct a steady flow pumping test to obtain the groundwater level drawdown S0 of the target aquifer observation hole; 2.3.3 According to the plane position of the observation hole of the target aquifer, calculate the groundwater level drop S of the pumping layer at that position p ; 2.3.4 Substitute the groundwater level drop S0 of the observation hole of the target aquifer obtained in step 2.3.2 and the groundwater level drop S0 of the pumping layer at the location obtained in step 2.3.

3. p The division results in a ratio, through which the hydraulic connection between the pumping layer and the target aquifer can be quantitatively evaluated.

3. A method for identifying and warning karst water layer pollution in coal mine goaf areas according to claim 2, characterized in that: In step 2.3, when multiple pumping tests are conducted, the groundwater drawdown S of the pumping layer should be calculated using the data at the maximum drawdown of the pumping well. p .

4. A method for identifying and warning karst water layer pollution in coal mine goaf areas according to claim 2, characterized in that: When arranging pumping wells, the pumping wells should be away from the permeable and impermeable boundaries of the aquifer, and should be arranged as far as possible in places that are representative of the water conduction and storage properties, recharge conditions, thickness and lithology conditions of the aquifer, and the drainage conditions near the pumping wells should be good, that is, the pumped water can be discharged without leakage to outside the influence radius of the pumping wells.

5. The method for identifying and warning karst water layer pollution in coal mine goaf according to claim 1 is characterized by: The fourth step includes the following steps, 4.1 Observation points are arranged on both sides of the boundary line between the goaf and the strata exposed on the surface in the direction of groundwater flow field, and in-hole observations are carried out to draw the curve of conductivity variation with depth; 4.2 According to the monitoring results of step 4.1, analyze the depth of the point where the conductivity changes suddenly, and collect water samples at the specified depth according to the conductivity change point to detect pH, SO4 2- Concentration and TDS, combined with the Piper three-line diagram, are used to study the groundwater chemical type and regional characteristics of the mining area.

6. The method for identifying and warning karst water layer pollution in coal mine goaf according to claim 1 is characterized by: In step 3.2, when arranging sampling points in the goaf, the goafs in the same hydrogeological unit are sampled at a rate of every 10 km. 2 The sampling points were arranged at a density of 4-6, located in the goaf of different hydrogeological units, and each unit was sampled at a rate of 10 km 2 The sampling points are arranged at a density of 3-4 sampling points, among which 3-5 sampling points are arranged at the mine drainage outlet according to the drainage volume, and 2-6 sampling points are arranged at the groundwater overflow points in each goaf; When arranging sampling points for coal-bearing strata, the groundwater aquifer system within the mining area is calculated at a rate of 10 km 2 The sampling points are arranged at a density of 2-6 sampling points, of which 2-6 sampling points are arranged in the mining area and on both sides of the water-conducting fracture; When arranging karst water sampling points, 2-6 sampling points are arranged where the groundwater and surface water are hydraulically closely connected. Along the flow direction of groundwater, at least one sampling point is arranged downstream and on both sides of the goaf.

Images