1 Introduction

Weakly cemented soft rock strata are prevalent in the western region of China. Due to their low strength, poor cementation, susceptibility to mudding, and softening in water, these strata present challenges for supporting the roof and controlling the surrounding rock (Meng et al. 2013; Tan et al. 2021). With the increase of mining depth, especially the poor stability of the roof of the weakly cemented soft rock working face, the development range of the overburden fracture is greatly increased, which leads to the influence of the stability of the aquiclude, and then causes the ecological environment problems such as the loss of the upper aquifer or surface water (Zhang et al. 2018). Hence, it holds paramount significance to investigate the evolution characteristics and developmental patterns of deep weakly cemented overburden fractures and to offer theoretical guidance for the technology of water-preserved mining in deep coal seams.

Currently, research primarily focuses on the theory of ‘transverse three zones’ and ‘vertical three zones’, as well as the height of water-conducting fracture zones and the stability of rock strata in quarry overburden structure division (He and Zhou 2022; Liu 2020; Liu et al. 1981). Regarding indoor mechanical tests, Cook (1965), Kemeny (1991), and Salamon et al. (1968) investigated the dynamic development of coal and rock fractures under stress from the perspective of fracture mechanics. Using regression analysis, Shi et al. (2022) deduced the height of the water-conducting fracture zone in weakly cemented overburden. Zhang et al. (2022a) collected measured data of the height of the water-conducting fracture zone in thick and extra-thick coal seams in China and derived a prediction formula for the height of the fractured zone through regression analysis. Li et al. (2021) determined the fracture development characteristics of high-intensity mining in various overburden types based on drilling flushing fluid consumption and borehole TV observation. They established a mathematical model for the relationship between the height of the caved zone and the layer spacing in the extra-thick coal seam and proposed a comprehensive mining thickness calculation formula. Zhang et al. (2022b) computed the equivalent water-resisting thickness of overlying strata based on the equivalent permeability coefficient, introducing a new method for implementing water-preserved coal mining technology. Xun et al. (2023) developed a prediction model for the height of the water-conducting fracture zone using an adaptive particle swarm optimization least squares support vector machine regression algorithm and UDEC numerical simulation. Xu et al. (2023) employed a simulation test platform to elucidate the relationship between the height of the water-conducting fracture zone and the width-depth ratio and width-thickness ratio of the working face. They obtained a prediction formula for the water-conducting fracture zone based on measured data. Yang et al. (2020) gathered data on the overburden caving zone, water-conducting fracture zone, deflection zone height, fracture-mining ratio, and the layered subsidence pattern of overburden strata at the working face using field test methods including drilling, borehole TV, and geophysical logging. Sun (2019) notes distinct movement disparities between weakly cemented overlying rocks in the western region and those in the central and eastern working faces. Sun’s simulation tests reveal a propensity for ultra-high fracture zone occurrences within weakly cemented overlying rocks. Meanwhile, Zhou Kai et al. (2023) focusing on the linkage instability mechanism of weakly cemented overburden rock, posit that an 'instantaneous and integral' instability failure ensues within the overburden rock layer subsequent to the conglomerate layer's rupture. The empirical formula for the water-conducting fracture zone primarily derives from gathered measured data to establish a correlation between the height of the water-conducting fracture zone and mining thickness. However, variations in stratum structure, overburden composition, lithology, and depth of coal seam burial across different regions and mines lead to discrepancies in the calculated results of the empirical formula (Zhang et al. 2022a). Given the characteristics of overburden fracture evolution zoning in deep, weakly cemented soft rock, additional research and analysis are warranted.

Expanding upon prior research, the author aims to concentrate on fractures within the overlying strata of deep weakly cemented soft rock. The research methods employed encompass theoretical analysis, numerical simulation, and experimental analysis. The author explores the quantitative characterization method for seepage velocities in the lower mining aquicludes of the aquifer. This is combined with the relationship between the seepage velocities of the aquicludes and the recharge velocity of the aquifer under the condition of water-preserved coal mining to derive the method for determining the water resistance of the aquicludes. Using an overlying strata structure model, the author studies and analyzes the characteristics of overlying strata fracture zoning with the ‘mild fracture zone’ as the focal point. The author combines numerical simulation and microseismic monitoring to study the evolution pattern of fracture development in different overburden fracture zones. This research provides insights for the study of morphology of fractures in deep weakly cemented soft rock overlying strata.

2 Fracture zoning characteristics of deep weakly cemented overlying strata

Predominantly occurring within Jurassic and Cretaceous strata in western China, weakly cemented soft rock formations are characterized by their brief diagenesis period, low strength, and susceptibility to water-induced disintegration. These attributes significantly influence the structure, lithology, and distribution of the overlying strata, notably affecting the formation and extent of the water-conducting fracture zone (Sun et al. 2019). Consequently, the investigation into the 'three zones of overlying strata’ and an analysis of fracture structures and water-blocking capacities within these strata serve as a basis for studying and analyzing the characteristics and methodologies employed in determining the zoning of fractures within overlying strata. The development of fractures in overlying strata caused by coal seam mining directly affects the protection of the water body in the aquifer. Based on the structure of overlying strata and the distribution characteristics of rock strata, we study the failure modes and water retention capacity of overlying strata in various layers. Drawing from previous research results on aquiclude seepage, and in combination with the findings of earlier studies on aquiclude seepage, we study and analyze the division characteristics of the overburden caved zone, structural equilibrium zone, severe fracture zone, mild fracture zone, and continuous deformation zone, as well as the characteristics and methods of overburden fracture zones suitable for water-preserved coal mining with the mild fracture zone as the core. The relative distribution of the fracture zones within the overlying strata in the quarry is shown in Fig. 1.

Fig. 1
figure 1

Relative position diagram of overlying strata fracture zone in quarry

Full size image

2.1 Caved zone

Regarding the impact of working face mining on the overlying strata, the attributes of a weakly consolidated roof tend to facilitate easy collapse. This results in large-scale rock mass collapses within the caved zone, forming discrete broken rock blocks that offer support to the overlying strata. The gaps between these rock blocks are substantial, significantly increasing the permeability coefficient of the rock strata, thereby enhancing its conductivity. Empirical calculation formulas based on three rules are employed within the caved zone.

2.2 Structural equilibrium zone

After the collapse of the weakly cemented roof, the structural equilibrium zone undergoes a sequence of damaging processes including overhanging, bending, and fracturing. Under horizontal forces, these zones compress and form stable articulated structure rock layers, typically comprising thick and hard rock layers crucial for bearing the overlying rock. The rock structure's force distribution is depicted in Fig. 2. Within the structural equilibrium zone, the rock strata fracture to create through cracks that facilitate aquifer channels. The load carried by the structural equilibrium zone isn’t entirely transferred to the lower caving rock strata, impeding vertical crack development. This maintenance prevents the complete loss of integrity and continuity within the overlying rock strata, contributing to the structural stability of the stope's overlying rock.

Fig. 2
figure 2

Stress diagram of rock strata in structural equilibrium zone

Full size image

The formation condition of the structural equilibrium zone is met when the thick and hard overlying rock stratum fractures, resulting in the emergence of a stable equilibrium structure within the rock stratum. The load on the lower rock stratum is less than that on the structural equilibrium zone. The ‘key stratum theory’ is employed to identify the equilibrium zone within the overburden structure (Qian et al. 2010).

The load of the k layer to the j layer (k > j) is calculated from the upper part of the caved zone.

$$q_{j\left| k \right.} = \frac{{E_{j} h_{j}^{{3}} \sum\limits_{{i{ = }j}}^{k} {\gamma_{i} h_{i} } }}{{\sum\limits_{{i{ = }j}}^{k} {E_{i} h_{i}^{{3}} } }}$$
(1)

where qj|k is the load of the kth stratum to the jth stratum (KPa); Ej is the elastic modulus of the j layer rock layer (GPa); hj is the layer j rock thickness (m); hi is the layer i rock bulk density (m); γi is the layer i rock bulk density (KN/m3).

The following sliding instability and rotary instability criteria formulas are used to further verify whether the rock layer forms a structurally equilibrium rock layer (Qian et al. 1994). The stress state of the rock block during equilibrium is shown in Fig. 3.

  • 1. The judgment of sliding instability can be expressed by the following formula:

    $$\left\{ \begin{gathered} i \le \tan \varphi + \frac{3}{4}\sin \theta \hfill \\ \sin \theta = \left[ {M - \sum {h\left( {k_{p} - 1} \right)} } \right]/l \hfill \\ i = {h \mathord{\left/ {\vphantom {h {L_{Z} }}} \right. \kern-0pt} {L_{Z} }} = \sqrt {\frac{{\sigma_{t} }}{3q}} \hfill \\ \end{gathered} \right.$$
    (2)

where h is the rock thickness (m); tanφ is the friction coefficient between the broken blocks is generally 0.3; θ is the key rock rotation angle (°); M is the mining height (m); Σh is the falling height (m); kp is the bulking coefficient (generally 1.1 ~ 1.5); LZ is the periodic breaking block length (m). σt is the rock tensile strength (MPa); q is the rock blocks under load (MPa).

Fig. 3
figure 3

Stress state of hinged equilibrium structure rock block

Full size image

  • 2. The judgment of rotary instability can be expressed by the following formula:

    $$i \ge \sin \theta /\left( {1 - \sqrt {\frac{nk\eta }{3}} } \right)$$
    (3)

    where n is the ratio of compressive strength to tensile strength of rock (taken as 10); k is according to the calculation of a simply supported beam (taken as 0.5); and η is the ratio of compressive strength to the compressive strength of rock mass (taken as 0.3).

2.3 Severe and mild fracture zone

The academic approach of virtually considering the aquifuge, characterized by a low degree of development in overlying fractures on the working face, as an equivalent water-resistant layer, is embraced. Utilizing the calculation method of the equivalent seepage coefficient in the vertical direction, associated with the working face length, mining height, subsidence parameters of overlying strata, and water physical parameters of rock strata, the relationship between the seepage velocity of the mining aquiclude and the recharge velocity of the aquifer serves as a quantitative characterization method for the water-blocking capacity of the mining aquiclude. Consequently, the distribution positions of severe and mild fracture zones are determined based on this water-blocking capacity of the mining aquiclude. The fractures within the rock stratum in the severe fracture zone are notably extensive, and they lack water insulating effects. Conversely, the mild fracture zone comprises an aquiclude that retains its capacity to block water even after being affected by mining. In cases where the overlying water-bearing rock group is situated within the bending subsidence zone, it will not induce significant water loss or related issues, as it is shielded by the aquiclude within the mild fracture zone. The range of fracture development is greatly increased during the mining of the weakly cemented soft rock working face. It is crucial to determine the location of the mild fracture zone in order to assess the water-blocking capacity of the lower aquiclude within the aquifer. The water-blocking capacity of the aquiclude is mainly determined by the relationship between its seepage velocity and the recharge rate of the aquifer. The seepage diagram of the rock strata under the aquifer is shown in Fig. 4. Additionally, seepage velocity is suitable for characterizing the flow velocity in the pores of the rock layer but may not apply to rock layers with a significant degree of fracture development in the caved zone and the structural equilibrium zone.

Fig. 4
figure 4

Seepage diagram of rock strata under aquifer

Full size image

3 Determination of water-blocking capacity of severe and mild fracture zones in overlying strata

In situations with complex hydrogeological conditions, empirical formulas have limitations when calculating the height of the water-conducting fracture zone. Preserving water resources within the overlying aquifer in the quarry primarily depends on the water-blocking capacity of the lower aquicludes. The water-blocking capacity of the mining aquicludes beneath the aquifer is the key factor in water-preserved coal mining (Liqiang et al. 2022; Xinyi et al. 2021). Preserving the water resources of the overlying aquifer in the quarry primarily depends on the water-blocking capacity of the lower aquicludes. Nevertheless, the development of fractures in the aquicludes directly affects their water-blocking capacity. Due to mining activities, there is a significant difference in seepage velocity between the fracture-developed aquicludes and their initial state.

3.1 Seepage velocity calculation of mining aquiclude

The water-blocking capacity of the mining aquicludes is primarily assessed through the relationship between seepage velocity and the aquifer’s recharge velocity. Studies have demonstrated that the critical water-retention condition for the aquicludes is when the seepage velocity equals the recharge rate of the aquifer (Zhang 2017). The seepage velocity of the aquicludes is linked to the permeability coefficient. Thus, the permeability coefficient of the aquicludes is employed to calculate both the seepage velocity and the recharge rate of the aquifer. In accordance with the critical water retention condition of the aquicludes, the water-blocking capacity of a particular aquiclude is determined.

The water-resisting layer consisting of micro-cracks in the overburden of the quarry is divided into n small deformation units of equal size. As shown in Fig. 5. All deformation elements are considered uniform media with equal vertical seepage. Consider only the vertical seepage of the aquicludes. According to Darcy’s law (Wang et al. 2013), the seepage velocity v of the aquicludes can be determined by the following formula:

$$v = K_{ij} \frac{{\Delta H_{i} }}{{M_{i} }}$$
(4)

where Kij is the permeability coefficient of deformation unit j in mining aquiclude i; ΔHi is the water head pressure difference of aquiclude i; Mi is the thickness of aquiclude i.

Fig. 5
figure 5

Seepage schematic diagram of single aquiclude division unit

Full size image

According to the relationship between the flow rate of the whole rock layer and the deformation unit, and the functional relationship between the permeability coefficient and the volume strain (Zhang et al. 2022b; Lu and Cheng 2018), the permeability coefficient Kij can be determined by the following formula:

$$\left\{ \begin{gathered} K_{ij} = k_{0} \frac{{(1 + {{\varepsilon_{vij} } \mathord{\left/ {\vphantom {{\varepsilon_{vij} } {\varphi_{0} }}} \right. \kern-0pt} {\varphi_{0} }})^{3} \rho {\text{g}}}}{{(1 + \varepsilon_{vij} )\eta }} \hfill \\ \varepsilon_{vij} = \frac{{{\text{d}}\left[ {\tfrac{{a_{i} }}{{R_{i} }}\int_{{d_{1i} - x}}^{{W - d_{2i} - x}} {S\left( {x^{\prime},z_{i - 1} } \right){\text{e}}^{{ - {\uppi }\left( {\tfrac{{x^{\prime}}}{{R_{i} }}} \right)^{2} }} } {\text{d}}x^{\prime}} \right]}}{{{\text{d}}x}} + \frac{{{\text{d}}\left[ { - 2{\uppi }\tfrac{{qa_{i} }}{{R_{i} h}}\int_{{d_{1i} - x}}^{{W - d_{2i} - x}} {S\left( {x^{\prime},z_{i - 1} } \right)x^{\prime}{\text{e}}^{{ - {\uppi }\left( {\tfrac{{x^{\prime}}}{{R_{i} }}} \right)^{2} }} } {\text{d}}x^{\prime}} \right]}}{{{\text{d}}z}} \hfill \\ \hfill \\ \left( {i \in \left[ {1,2, \cdots ,q} \right],j \in \left[ {1,2, \cdots ,n} \right]} \right),{\text{when }}i = 1,S(x^{\prime},z_{i - 1} ) = m \hfill \\ \end{gathered} \right.$$
(5)

where k0 is the initial permeability; εvij is the volume strain of deformation element j in aquiclude i; φ0 is the initial porosity; η is the dynamic viscosity coefficient; x is the horizontal distance between the surface prediction point of deformation unit j in aquiclude i and the underground mining unit point; S(x',zi-1) is the subsidence value of the upper layer of aquiclude i at x'; ai is the sinking coefficient of aquiclude i; Ri is the main influence radius of aquiclude i; q is the number of strata from the mining coal seam to the target aquiclude; h is the buried depth of the working face; W is the mining length of the coal seam along the direction of the i section of aquiclude in the study area; d1i and d2i are the offsets of inflection points on the left and right sides of aquiclude i (d1i = d2i in near-horizontal strata).

Combining Eqs. (4) and (5), the seepage velocity v1 of the lower mining aquiclude of the aquifer can be expressed by the following formula:

$$\left\{ \begin{gathered} v_{1} = \frac{\Delta H}{{L\sum\limits_{{i = {1}}}^{n} {\left( {{{M_{i} } \mathord{\left/ {\vphantom {{M_{i} } {\sum\limits_{{j = {1}}}^{n} {K_{ij} L_{ij} } }}} \right. \kern-0pt} {\sum\limits_{{j = {1}}}^{n} {K_{ij} L_{ij} } }}} \right)} }} \hfill \\ K_{ij} = k_{0} \frac{{\left( {1 + {{\varepsilon_{vij} } \mathord{\left/ {\vphantom {{\varepsilon_{vij} } {\varphi_{0} }}} \right. \kern-0pt} {\varphi_{0} }}} \right)^{3} \rho {\text{g}}}}{{\left( {1 + \varepsilon_{vij} } \right)\eta }} \hfill \\ \varepsilon_{vij} = \frac{{{\text{d}}\left[ {\tfrac{{a_{i} }}{{R_{i} }}\int_{{d_{1i} - x}}^{{W - d_{2i} - x}} {S\left( {x^{\prime},z_{i - 1} } \right){\text{e}}^{{ - {\uppi }\left( {\tfrac{{x^{\prime}}}{{R_{i} }}} \right)^{2} }} } {\text{d}}x^{\prime}} \right]}}{{{\text{d}}x}} + \frac{{{\text{d}}\left[ { - 2{\uppi }\tfrac{{qa_{i} }}{{R_{i} h}}\int_{{d_{1i} - x}}^{{W - d_{2i} - x}} {S\left( {x^{\prime},z_{i - 1} } \right)x^{\prime}{\text{e}}^{{ - {\uppi }\left( {\tfrac{{x^{\prime}}}{{R_{i} }}} \right)^{2} }} } {\text{d}}x^{\prime}} \right]}}{{{\text{d}}z}} \hfill \\ \left( {i \in \left[ {1,2, \cdots ,q} \right],j \in \left[ {1,2, \cdots ,n} \right]} \right),{\text{when }}i = 1,S(x^{\prime},z_{i - 1} ) = m \hfill \\ \end{gathered} \right.$$
(6)

3.2 Judgment method of water-blocking capacity of mining aquiclude

In the rock strata displaying mining-induced fracture development above the structural equilibrium zone within the deeply weakly consolidated overburden structure, the seepage velocity of the aquifer equals the recharge velocity of the aquifer, serving as the critical condition for assessing the aquifer's water-blocking capacity. The determination of the recharge velocity v of the aquifer aligns with Darcy’s law (Wang et al. 2013), which can be expressed using the following formula.

$$v = \frac{Q}{A}$$
(7)

where Q represents the water inflow of the aquifer; A is the seepage cross-sectional area of the aquicludes in the study area.

According to the critical water-retaining condition of the mining aquicludes, the method for determining the water-blocking capacity of the mining aquicludes is as follows: the seepage velocity of the aquicludes is less than the recharge velocity of the overlying aquifer, i.e., when v1/v ≤ 1, the aquicludes have water-blocking capacity; the seepage velocity of the aquicludes is greater than the recharge rate of the overlying aquifer, i.e., when v1/v > 1, the mining aquicludes do not have water-blocking capacity. Therefore, the determination method of mining aquicludes with water-blocking capacity can be expressed by the following formula:

$$\left\{ \begin{gathered} \frac{{v_{1} }}{v} \le 1 \hfill \\ v = \frac{Q}{A} \hfill \\ v_{1} = \frac{\Delta H}{{L\sum\limits_{{i = {1}}}^{n} {\left( {{{M_{i} } \mathord{\left/ {\vphantom {{M_{i} } {\sum\limits_{{j = {1}}}^{n} {K_{ij} L_{ij} } }}} \right. \kern-0pt} {\sum\limits_{{j = {1}}}^{n} {K_{ij} L_{ij} } }}} \right)} }} \hfill \\ K_{ij} = k_{0} \frac{{\left( {1 + {{\varepsilon_{vij} } \mathord{\left/ {\vphantom {{\varepsilon_{vij} } {\varphi_{0} }}} \right. \kern-0pt} {\varphi_{0} }}} \right)^{3} \rho {\text{g}}}}{{\left( {1 + \varepsilon_{vij} } \right)\eta }} \hfill \\ \varepsilon_{vij} = \frac{{{\text{d}}\left[ {\tfrac{{a_{i} }}{{R_{i} }}\int_{{d_{1i} - x}}^{{W - d_{2i} - x}} {S\left( {x^{\prime},z_{i - 1} } \right){\text{e}}^{{ - {\uppi }\left( {\tfrac{{x^{\prime}}}{{R_{i} }}} \right)^{2} }} } {\text{d}}x^{\prime}} \right]}}{{{\text{d}}x}} + \frac{{{\text{d}}\left[ { - 2{\uppi }\tfrac{{qa_{i} }}{{R_{i} h}}\int_{{d_{1i} - x}}^{{W - d_{2i} - x}} {S\left( {x^{\prime},z_{i - 1} } \right)x^{\prime}{\text{e}}^{{ - {\uppi }\left( {\tfrac{{x^{\prime}}}{{R_{i} }}} \right)^{2} }} } {\text{d}}x^{\prime}} \right]}}{{{\text{d}}z}} \hfill \\ \left( {i \in \left[ {1,2, \cdots ,q} \right],j \in \left[ {1,2, \cdots ,n} \right]} \right),{\text{when }}i = 1,S(x^{\prime},z_{i - 1} ) = m \hfill \\ \end{gathered} \right.$$
(8)

Using the rock (coal) layer distribution in Fig. 6 as an example, the study area resides in western Inner Mongolia, where the prevalence of weakly cemented soft rock is influenced by stratum deposition (Tan et al. 2021). The 3–1 coal seam exhibits a thickness of 5.06 m. The sandy mudstone in the roof displays a compressive strength ranging from 15.3 to 24.3 MPa, falling below 25 MPa. Additionally, its tensile strength ranges from 1.1 to 1.7 MPa, while exhibiting a cohesion of 10.39 MPa and an internal friction angle of 27°. Physical component analysis and X-ray diffraction tests reveal that the primary mineral constituents of the sandy mudstone in the roof comprise quartz, illite, chlorite, and albite. Specifically, illite, chlorite, and other clay minerals constitute 20 to 40% of the composition, indicating a high content of clay minerals characteristic of weakly cemented soft rock (Zhou et al. 2023). The seepage diagram of the rock layer is shown in Fig. 6. Aquiclude #3 is thin, and there are thick aquifers above and below, with the potential for inter-layer overflow. Based on the overburden structure and Eq. (8), the water-blocking capacity of the main aquiclude above the working face is studied. The relevant parameters of sandy mudstone aquiclude in the study area are shown in Table 1. With a working face length of 100 m, the permeability change curve of the mining aquiclude is calculated at varying heights from the working face, as depicted in Fig. 7.

Fig. 6
figure 6

The distribution and seepage diagram of overlying strata of 3–1 coal

Full size image
Table 1 Calculation parameters of sandy mudstone aquiclude
Full size table
Fig. 7
figure 7

Permeability changes of overlying aquiclude at different heights

Full size image

As observed in the curve in Fig. 7, the permeability of the overlying aquiclude is high due to the influence of the boundary coal pillars at both ends of the working face. In the middle of the working face, there is a gradual compaction of the rock strata with settlement cracks, leading to a decrease in the permeability of the aquiclude. Moreover, the permeability variation of the overlying aquiclude is inversely correlated with the height from the roof of the working face. The farther the distance from the working face, the less the water resistance of the aquiclude is affected.

The calculation shows that the recharge rate of the aquifer is 2.1 × 10^-3 m/d, the seepage velocity of overlying Aquiclude #1 is 3.3 × 10^-2 m/d, and the seepage velocity of Aquiclude #2 is 5.2 × 10^-5 m/d. Therefore, under the influence of mining, Aquiclude #2 has the ability to block water, and the overlying Aquiclude #2 is defined as a mild fracture zone. According to the above division method of overlying strata fractures, the division results are shown in Fig. 8.

Fig. 8
figure 8

Overlying strata fracture zoning diagram of the study area

Full size image

4 Fracture evolution law and characteristics of deep weakly cemented overlying strata

4.1 Simulation evolution law of overlying strata fracture under different conditions

4.1.1 Numerical model establishment and related parameters

The discrete element numerical simulation software UDEC was used to simulate the evolution law of overlying strata fractures in 3–1 coal mining. The rock (coal) layer is distributed nearly horizontally, with a buried depth of the coal seam at 495.41 m. The model size is x × y = 400 m × 157 m, and a uniform load of 8.325 MPa is applied to the upper part of the model. The overlying strata in the study area are generalized into layers, combining rock strata with physical and mechanical properties and small thicknesses. The model is depicted in Fig. 9. The physical and mechanical parameters of the rock (coal) are detailed in Table 2 (Zhou et al. 2023).

Fig. 9
figure 9

Numerical simulation model

Full size image
Table 2 Physical and mechanical parameters of coal and rock mass (Zhou et al. 2023)
Full size table

Equation (8) reveals that the permeability coefficient of the aquiclude is dependent on the mining height, length, rock subsidence parameters, and hydrogeological parameters of the working face. This is because the rock mechanics parameters and occurrence of the rock strata in the working face are predetermined. Therefore, a total of eight scenarios were simulated, with the working face length varying between 150 m, 200 m, 250 m, and 300 m, and the mining height at 3 m, 4 m, 5 m, and 6 m, respectively. Considering the characteristics of the overburden fracture zone, an analysis of the evolution law of overburden fractures under varying conditions was conducted.

4.1.2 Overlying strata fracture development law under different working face length

Figure 10 illustrates the development and evolution of overburden fractures at varying working face lengths. At a working face length of 150 m, the middle overburden exhibits a 'trapezoid'-shaped open fracture zone, with the slip fracture range at both ends of the working face symmetrically distributed.

Fig. 10
figure 10

Overlying strata fracture distribution under different working face length

Full size image

The stress distribution of the sandy mudstone aquiclude in the mild fracture zone corresponding to various working face lengths is illustrated in Fig. 11. With the increasing working face length, the pressure relief area in the middle of the overlying Aquiclude on the working face expands. At a working face length of 250 m, a stress recovery zone emerges, and its range expands with the increasing working face length, signifying an increase in the fracture closure range. Consequently, due to the settlement and compaction of overlying strata, designing a longer working face length results in a relatively stable fracture range of overlying strata. Meanwhile, the closed fracture zone in the middle expands, and the degree of fracture in the mild fracture zone diminishes.

Fig. 11
figure 11

Normal stress distribution of mild fracture zone with different working face lengths

Full size image

4.1.3 Fracture development law of overlying strata in different mining height of working face

The development and evolution of overlying rock fissures at varying working face mining heights are depicted in Fig. 12. As the mining height of the working face increases, the open fracture zone expands primarily in the vertical direction, leading to an overall increase in the extent of the slip fracture zone. Conversely, the range of the closed fracture zone in the middle overburden gradually diminishes. At a mining height of 6 m, the open fracture zone extends to the overlying sandy mudstone Aquiclude. As the mining height increases, the extent of the crack range also expands. The mild fracture zone predominantly exists within the relief zone, where the stress relief variation remains marginal despite the heightened mining activity. Consequently, at greater mining heights, the open fracture zone of the overlying strata primarily expands vertically, resulting in an overall enlargement of the slip fracture zone, while the range of the closed fracture zone diminishes (Fig. 13).

Fig. 12
figure 12

Overlying strata fracture distribution under different working face mining height

Full size image
Fig. 13
figure 13

Normal stress distribution of mild fracture zone in different working face mining height

Full size image

4.2 Microseismic monitoring analysis of overlying strata fractured zone

The KJ768 coal mine employs a microseismic monitoring system. GZC10 three-component pickup measuring points are positioned at 150-m intervals along the roadway side from the working face cut, with each crossheading equipped with a KJ959-F microseismic monitoring substation. Figure 14 illustrates the layout of the microseismic monitoring sensors and site. The geological structure in the study area is uncomplicated, with the coal seam occurrence exhibiting general stability, except for minor folds in the working face's local area. Consequently, the impact of geological structure on microseismic events is disregarded. Figure 15 depicts the arrangement of microseismic monitoring. As the working face advances continuously, the monitoring includes tracking the frequency of microseismic events and the energy released throughout the mining period. Figure 16 presents the location and projection of microseismic events along the strike during the mining of the working face. The frequency and energy of microseismic events within the overlying strata indicate the fracture of rock strata across different fracture zones. Figure 17 details the enumeration of frequency, energy, and elevation position of microseismic events in various fracture zones.

Fig. 14
figure 14

Microseismic monitoring sensor and site layout

Full size image
Fig. 15
figure 15

The arrangement of microseismic monitoring points

Full size image
Fig. 16
figure 16

Location and projection of microseismic events in overlying strata

Full size image
Fig. 17
figure 17

Characteristics of daily frequency-energy-elevation position of microseismic events in different fracture zones

Full size image

The analysis of Figs. 16 and 17 reveals 42 occurrences of microseismic events within the caving zone, with an energy release range between 4954.36 and 8238 J. Notably dense microseismic activity was observed in the low hard rock strata adjacent to the stope within the structural equilibrium zone (138 times) and the weak rock strata of the severe fracture zone (258 times). These events were evenly distributed, leading to extensive crack generation and compromised integrity. Conversely, there were few microseismic events detected in the aquiclude of the mild fracture zone, totaling 55 events with a total daily frequency of only 1 to 2 times. This indicates fewer fractures within the rock layer and better structural integrity.

5 Conclusion

  1. 1.

    Considering the traits of low strength, poor cementation, and susceptibility to water softening in weakly cemented soft rock, the proposed fracture zoning characteristics of deep weakly cemented overburden center around the 'mild fracture zone.' Introducing the concept that the key stratum of the overlying strata supports the load, a formula is presented to determine the extent of the structural equilibrium zone. Additionally, the zoning characteristics of severe and mild fracture zones are delineated based on the water-blocking capacity of the aquiclude, offering a novel approach to assess the stability of the upper aquifer during working face mining.

  2. 2.

    Utilizing Darcy's law to establish the relationship between permeability coefficient and volumetric strain, the calculation formula for seepage velocity is derived from the vertical permeability coefficient of the mining aquiclude. This derivation leads to the proposal of a quantitative method for characterizing the water-blocking capacity of the mining aquiclude by correlating the seepage velocity of the aquiclude with the recharge velocity of the aquifer. Analysis of the permeability change curve of the aquiclude reveals a negative correlation between permeability change and distance from the working face. Moreover, the boundary coal pillar significantly influences the permeability of the overlying strata on both sides of the gob, whereas the subsidence-induced compaction results in lower permeability in the central region of the overlying strata.

  3. 3.

    Through combined numerical simulation analysis, it is determined that extending the working face length leads to pronounced overburden fracture development near the gob boundary. Consequently, there is a substantial increase in the permeability of the water-resisting layer within this area, reducing its corresponding water-blocking capacity while fostering conspicuous fracture development. Additionally, there is an increased degree of pressure relief. The overlying rock within the gob’s center settles and compacts, significantly reducing permeability while enhancing the corresponding water-blocking capacity, consequently restoring the pressure relief area to its initial rock stress condition. Moreover, the greater the mining height of the working face, the wider the development range of the overburden fracture, significantly impacting the permeability of the overburden rock. Hence, given a predetermined mining face thickness, designing a longer working face length is intended to augment the compaction and closure range of the overburden fracture within the gob, thereby safeguarding the water resources of the upper aquifer.

  4. 4.

    The structural equilibrium zone exhibits a high density of micro-seismic events within hard rock, while the severe fracture zone presents dense events in the softer rock. These events occur frequently, accompanied by numerous cracks in the rock layer, corresponding to tension and shear failure cracks observed in the UDEC numerical simulation. Conversely, the mild fracture zone records relatively fewer microseismic events, indicative of less rock layer fracturing. Within the numerical simulation, the rock layer primarily experiences shear fractures.