Introduction

Water inrush from confined aquifers in coal seam floor is a common geological disaster in the process of coal mining, which poses a great threat to the safety of mining1,2,3,4. The aquiclude in coal seam floor is of great significance to prevent the occurrence of water inrush, and through analysis of the water-resisting ability of the aquiclude, the potential risk areas of water inrush can be effectively identified, which can provide scientific references for preventing and controlling water inrush disasters. As early as the beginning of the 20th century, some European scholars noticed the role of the aquiclude in coal seam floor during coal mining, and realized that the frequency of water inrush, the water inflow and the thickness of the aquiclude were closely related through analysis of water inrush statistics5. In the 1940s, the Hungarian scholar Wegfrens first proposed the concept of the relative aquiclude in the floor. He believed that the water inrush from the coal seam floor was related to the thickness of aquiclude and water pressure, and established a quantitative relationship between the three factors6,7. In the 1980s, geological researchers of Shandong Institute of Mining and Jingxing Mining Bureau first proposed the concept of “lower three-zones”, which can be explained as, after the panel moves forward, the aquiclude in coal seam floor under the goaf can be divided into three zones. And they are “floor failure zone”, “floor water-resisting zone” and “original lifting zone” specifically8. Dusan Kuscer used the water pressure drop gradient P and the specific thickness r (r = 1/P and P is the water pressure value of the lower part of the water-resistant rock layer, MPa) to study the safe thickness of aquiclude under constant water pressure9. Santos and Bieniawski (1989) used the Hoek-Brown rock mass strength criterion, introduced the concept of critical energy release point, and analyzed the anti-damage ability of the floor aquiclude10. Zhang Jincai and Liu Tianquan pointed out that the floor strata under the coal mining panel was composed of “water-conducting fractured zone” and “relative aquiclude zone”11. And they combined the Coulomb-Mohr and Grifith strength theories to obtain the maximum failure depth of the floor strata during mining. Wang Zuoyu and Liu Hongquan proposed the theory of “original position tension crack and the zero position failure”12. According to the theory, it is believed that when the mining panel moves forward, the floor releases elastic strain energy in the form of brittle failure, forming zero position failure of the floor rock mass and the confined water rises along the zero position failure zone and the original position failure zone.

Recent advances also contribute to the understanding of the water-resisting ability of aquicludes. For instance, Li et al. investigated rock burst monitoring through integrated microseismic and electromagnetic radiation methods, offering insights into the dynamic behaviors of rock masses under stress, which is essential for understanding water inrush phenomena in coal mines13,14. Additionally, experimental studies on the effect of liquid nitrogen cold soaking on coal pore structure and fractal characteristics have provided important data on the mechanical properties of coal under extreme conditions, which can be linked to the integrity of the aquiclude15. Furthermore, Li et al. conducted an experimental study on compressive behavior and failure characteristics of imitation steel fiber concrete, which could inform the understanding of structural failure mechanisms in aquicludes16. In the process of confined water moving upwards in the aquiclude in the coal seam floor, the formation of residual water pressure is closely related to the water-resisting ability of the strata. Therefore, objective and accurate evaluation of the water-resisting ability of the aquiclude plays an important role in preventing water inrush disasters in coal mines. The water-resisting ability of the aquiclude in the coal seam floor is not only related to its thickness but also affected by factors such as the mechanical properties and integrity of the strata. The formation mechanism of its ability is complex, and currently, there are few related studies.

In this study, Longfeng coal mine was taken as an example. The mining area is seriously threatened by the bottom plate karst water from the Maokou Formation aquifer, the aquifer is highly variable, and it is difficult to accurately describe it by traditional methods. Based on the availability of data and the representativeness of factors, the evaluation factors for the water-resisting ability of aquiclude in coal seam floor were determined. Besides, the theory of multi-source integrated information was applied. And the geographic information system with spatial data analysis and processing function as well as the non-linear mathematical method-entropy weight method was adopted to evaluate and analyze the water-resisting ability of aquiclude in coal seam floor. The evaluation factors and methods proposed in this study can provide certain references for research in related fields.

An overview of the research area

Geography location

With a designed production capacity of 1.2 million tons/year, Longfeng Coal Mine is situated in the southwest of Jinsha County, Guizhou Province, China, and it is about 14.5 km away from the county in a straight-line distance, geographical coordinates: 106°07′30″~106°15′00″, 27°14′45″~27°24′45″, as shown in Fig. 1. Longfeng Coal Mine belongs to the plateau-low mountain-hilly landform, and the highest elevation is + 1628.3 m, the lowest + 980.2 m, so the relative maximum elevation difference reaches 648.1 m.

Fig. 1
Fig. 1
Full size image

Geography location.

Hydrogeological conditions

Generally, the mining area is supplied by atmospheric rainfall, which mainly occurs in summer, with the maximum annual precipitation of 1346.2 mm, the minimum annual precipitation of 721.90 mm, and the annual average precipitation of 1048 mm. Groundwater outcrops on the surface in the form of springs or spring groups through flows like karst fissures, karst pipelines, disappearance of river, etc., characterized by strong alternation, shallow circulation, short runoff and concentrated discharge. The aquifers and aquicludes in the mining area, from old to new, are as follows: Maokou Formation (P2m) Aquifer of Permian Middle System, Longtan Formation (P3l) Aquiclude of Permian Upper System, Changxing Formation (P3c) Aquifer of Permian Upper System, Jiujitan beach section (T1y3 Aquiclude of Yelang Formation, Yulongshan section (T1y2 Aquifer of Yelang Formation, Shabaowan section (T1y1) Aquiclude of Yelang Formation and loose pored aquifer of Quaternary System (Q).

In the mining area, the coal bearing strata is Longtan Formation (P3l) of Permian Upper system, and it is a coal-bearing series composed of marine-continental facies and polycyle sedimentation, and it contains mudstone, silty mudstone, fine sandstone, siliceous limestone, bioclastic limestone, carbonaceous mudstone and coal seam. The aquifer which seriously threatens seam mining in the mining area is the corrosion fissure aquifer of Maokou Formation in the Permian Middle System, with a sound water yield and pressure bearing capacity, and the water yield is highly uneven due to strong karstification. The mining of the lower group coal in Longfeng Coal Mine will inevitably damage the aquiclude of the seam floor. Once the damage depth reaches the karst aquifer of Maokou Formation, a water inrush disaster will occur readily under the action of water pressure. Therefore, the evaluation of water-resisting ability of the aquiclude of the coal seam floor will help to identify areas with weak water-resisting ability and provide basis for the prevention and control of water disasters. The aquiclude evaluated in this paper is the one between the 11th coal seam of the lower group coal and the aquifer of Maokou Formation in the southwest mining area of Longfeng Coal Mine (Fig. 2).

Fig. 2
Fig. 2
Full size image

Structural relationship between coal seams and aquifers.

Evaluation factors of water-resisting ability of aquiclude and making of thematic maps

Evaluation factors of water-resisting ability of aquiclude

The water-resisting ability of aquiclude mainly includes resistance to seepage failure and mining failure, and it is closely related to the thickness of the rock stratum, lithologic combination, and the integrity of the rock mass. In this paper, based on the consideration of the availability of data collection and the representativeness of factors, and combined with the geological conditions of the research area, the evaluation factors of water-resisting ability of the aquiclude of the coal seam floor are determined. The selection of these specific factors is justified by their direct relevance to the mechanical and hydrological behavior of the floor aquiclude, as supported by previous studies. For instance, factors such as rock mechanical strength parameters (e.g., uniaxial compressive strength) and in-situ stress were considered but excluded due to data limitations in the study area; however, the chosen factors effectively capture key aspects of aquiclude performance, as explained below.

  1. 1.

    Thickness of effective aquiclude: It is yielded by subtracting the depth of the fracture zone caused by underground pressure from the total thickness of the aquiclude. This factor directly influences the mechanical resistance to water pressure, as a thicker effective aquiclude reduces the risk of hydraulic failure by providing a longer seepage path and greater structural integrity. Relevant literature highlights that effective thickness is a critical parameter in assessing the stability of floor strata under mining-induced stress17. In this paper, the depth of the fracture zone caused by underground pressure of the coal seam floor is determined according to the empirical formula 118.

    $${\text{h}}\,=\,0.00{\text{85H}}\,+\,0.{\text{1665a}}\,+\,0.{\text{1}}0{\text{79L}} - {\text{4}}.{\text{3579}}$$
    (1)

    Where h is fracture zone depth under underground pressure; H is mining depth; a is dip angle of coal seam; L is dip length of working surface.

  2. 2.

    Brittle rock thickness of aquiclude: The brittle rock mass is hard, with great mechanical and compressive properties, so it can effectively resist the ground pressure in the mining process. This factor is linked to the anti-mining failure ability, as brittle rocks (e.g., sandstone and limestone) enhance the overall strength of the aquiclude by resisting deformation and fracture propagation14,19.

  3. 3.

    Rock qualitydesignation of aquiclude: Rock quality designation (RQD) is defined as the ratio of the sum of the length of the core section with a length of more than 10 cm obtained in the core to the drilling footage during the drilling round when using a 75 mm diameter diamond bit and a double-layer core tube to drill in the rock20. The RQD reflects the integrity of rock mass. The larger the RQD value, the more integrated the rock mass and the stronger the water-resisting ability. RQD is directly correlated with the hydrological behavior of the aquiclude, as it indicates the degree of fracturing and permeability; higher RQD values imply lower conductivity and better seepage resistance, which is crucial for preventing water inrush21,22. In this study, the RQD of aquiclude is calculated by the sum of each strata’s RQD value multiplied by its thickness, divided by the thickness of aquiclude.

  4. 4.

    Consumption of drilling fluid: The consumption and the properties variation of drilling fluid can often reflect the permeability of rock stratum and demonstrate the fragmentation degree of rock mass and the development condition of fissure joints. The greater the degree of the rock mass, the poorer the integrity and the weaker the water-resisting ability of the aquiclude. This factor serves as a proxy for hydrological behavior, as increased fluid consumption indicates higher permeability and fracture density, which are directly associated with reduced water-resisting capacity23.

  5. 5.

    Mudstone ratio of aquiclude: The water-resisting ability of aquiclude is closely related to its lithologic association. Generally speaking, mudstone has good plasticity and weak permeability, and it has greater water-resisting ability than brittle rock, while brittle rock can maintain the stability of aquiclude. The mudstone ratio is directly tied to the mechanical and hydrological properties of the aquiclude; mudstone layers act as barriers to water flow due to their low permeability and plasticity, which enhance the overall sealing efficiency24,25. In this paper, the ratio of mudstone thickness to the total thickness of aquiclude is employed to reflect the water-resisting ability of mudstone.

  6. 6.

    Fault influenced zone and fault fracture zone: Fault destroys the integrity of the rock strata, causing weak zones to develop in the water-resisting strata, which weakens the water-resisting ability of the aquiclude26. Therefore, the influence of fault must be considered when determining the key factors of the water-resisting ability of the aquiclude of the coal seam floor water. Fault zones directly impact the mechanical behavior by creating discontinuities that reduce shear strength and increase permeability, leading to higher water inrush risks. In this paper, the fault fracture zone and fault influenced zone are considered in the quantification process of fault.

Thematic map of influencing factors of aquiclude

The data of geological drilling and geological structure of Longfeng Coal Mine were collected systematically and analyzed statistically. On this basis, thematic maps (Figs. 3, 4, 5, 6, 7 and 8) on influencing factors in the research area were made with the GIS technology. Specifically, the discrete borehole data were interpolated into continuous spatial surfaces using the Inverse Distance Weighting (IDW) method in ArcGIS 10.8 software. IDW was selected due to its simplicity and effectiveness in handling spatially distributed borehole data, where closer points have more influence on the interpolated values, which is suitable for the relatively uniform distribution of boreholes in the study area. This method is justified by the geological context of the aquiclude, where factors such as thickness and rock quality exhibit gradual spatial variations without abrupt changes, as supported by the borehole data distribution. The interpolation process ensured that the thematic maps accurately represent the spatial characteristics of each influencing factor, enhancing the reproducibility of the GIS analysis. Through the thematic map of each influencing factor, the spatial distribution characteristics of the index value of each influencing factor can be known easily.

Fig. 3
Fig. 3
Full size image

Thematic map of effective aquiclude thickness.

Fig. 4
Fig. 4
Full size image

Thematic map of brittle rock thickness.

Fig. 5
Fig. 5
Full size image

Thematic map of rock quality designation.

Fig. 6
Fig. 6
Full size image

Thematic map of drilling fluid consumption.

Fig. 7
Fig. 7
Full size image

Thematic map of aquiclude mudstone ratio.

Fig. 8
Fig. 8
Full size image

Thematic map of fault influenced and fault fracture.

Data normalization

The influencing factors are normalized, aiming to eliminate the influence of different dimensions of each factor on the evaluation result, so that the data has statistical significance and comparability27,28. For the effective thickness of the aquiclude, the brittle rock thickness of the aquifer, and the mudstone ratio of the aquifer, the water-resisting ability increases with the increase of the above factors, which is a positive factor. The normalization formula of the index value is:

$${S_i}=v+\frac{{(u - v) \times ({x_i} - \text{min} ({x_i}))}}{{\text{max} ({x_i}) - \text{min} ({x_i})}}$$
(2)

Where min(xi) is the minimum quantization value; max(xi) is the maximum quantization value; u is the upper limit of the normalization range, which is 1; v is the lower limit of the normalization range, which is 0; Si is the normalized data .

For RQD, consumption of drilling fluid, fault influenced zone and fault fracture zone, the water-resisting ability decreases as the values of the above factors rise, which is a negative factor. The normalization formula of the index value is:

$${S_i}=1-{\text{ }}(v+\frac{{(u - v) \times ({x_i} - \text{min} ({x_i}))}}{{\text{max} ({x_i}) - \text{min} ({x_i})}})$$
(3)

After normalizing the data of each influencing factor by formula 2 and formula 3, the corresponding attribute database is established.

Determination of factor weight based on entropy weight method

An overview of the entropy weight method

The entropy weight method is an objective weighting method. Compared with subjective weighting methods such as the Delphi method and the analytic hierarchy process, the objective weighting has the original information which takes root in objective facts, so the influence of decision makers on factor weights can be avoided, with results more objective29,30. This method is basically to determine the weight based on the difference of the evaluation factors. Generally speaking, if a certain factor has a smaller entropy value, it’s indicated that this factor has a greater variation of its index value, it offers more information, it plays a bigger role in in comprehensive evaluation, and it has a larger weight. On the contrary, if a certain factor has a greater entropy value, it’s indicated that this factor has a smaller variation of its index value, it offers less information, it plays a smaller role in in comprehensive evaluation, and it has a smaller weight31. The calculation process of employing the entropy weight method to solve weight is as follows:

  1. 1.

    Calculate the proportion Pij of the i-th standard value of the j-th evaluation factor:

    $$P_{ij}=\frac{{U_{ij}}}{{\sum\limits_{{i=1}}^{m} {U_{ij}} }}$$
    (4)
  2. 2.

    Calculate the entropy ej(0 ≤ ej≤1)of the j-th factor:

    $${e_j} = - k\sum\limits_{i = 1}^m {\left( {{p_{ij}}\ln {p_{ij}}} \right)}$$
    (5)
  3. 3.

    Calculate the utility value dj of the j-th factor. The greater dj, the greater the value of the factor and the greater its weight.

    $${d_j} = 1 - {e_j}$$
    (6)
  4. 4.

    Calculate the entropy weight Wj of the j-th factor:

    $${W_j} = \frac{{{d_j}}}{{\sum\nolimits_{j = 1}^n {{d_j}} }},\;{\text{and}}\;{\text{satisfy}}\;\sum\limits_{j = 1}^n {{W_j} = 1}$$
    (7)

    Where xij and Uij are the value and standard value of the j-th evaluation factor of the i-th evaluation unit respectively; K is the adjustment coefficient, and K = 1/lnm (m is the number of evaluation unit in the evaluation system).

Factor weight solution

According to formulas 2 and 3, the data of each influencing factor is normalized to obtain the standard value of each factor, see Table 1.

Table 1 Standard values of evaluation factors for water-resisting ability.
Full size table

In this study, the evaluation units are defined as the drill points (e.g., zk01 to zk21) listed in Table 1. Each drill point represents a specific location in the study area, and the data from these points are used to calculate the weights using the entropy weight method. For the GIS-based composite evaluation in Sect. 5, the evaluation units are grid cells for spatial analysis, but the initial weight determination is derived exclusively from the drill point data.

A brief example using drill hole zk01 from Table 1 is provided below. The normalized values for zk01 are: Thickness of effective aquiclude = 0.445, Rock quality designation = 0.801, Consumption of drilling fluid = 0.119, Mudstone ratio of aquiclude = 0.377, Brittle rock thickness = 1.000, Fault influenced and fault fracture = 1.0.

Step 1: Calculate the proportion Pij for each factor using Eq. (4). For example, for the Thickness of effective aquiclude factor, Pij for zk01 is calculated as Pij = Uij / ΣUij, where Uij is the normalized value for zk01 (0.445), and ΣUij is the sum of normalized values for all drill points for this factor. From Table 1, the sum for Thickness of effective aquiclude is approximately 6.823 (calculated from the data: 0.445 + 0.000 + 0.046 + … + 0.413). Thus, Pij for zk01 is 0.445 / 6.823 ≈ 0.0652. Similarly, Pij values are computed for all factors and all drill points.

Step 2: Calculate the entropy ej for each factor using Eq. (5). For Thickness of effective aquiclude, ej = -K ΣPij ln Pij, where K = 1/ln(m) and m is the number of drill points (21, so K ≈ 0.334). The sum is over all drill points. Using the Pij values from Step 1, ej is computed. For instance, if the sum of Pij ln Pij for Thickness of effective aquiclude is -2.100, then ej = -0.334 × (-2.100) ≈ 0.7014. However, the actual calculation in this study yielded ej = 0.9343 for this factor, as shown in Table 2.

Step 3: Calculate the utility value dj for each factor using Eq. (6): dj = 1 - ej. For Thickness of effective aquiclude, dj = 1–0.9343 = 0.0657.

Step 4: Calculate the weight Wj for each factor using Eq. (7): Wj = dj / Σ dj, where the sum is over all factors. The sum of dj values from Table 2 is 0.0657 + 0.0456 + 0.0512 + 0.0508 + 0.0528 + 0.0228 = 0.289. Thus, Wj for Thickness of effective aquiclude is 0.0657 / 0.289 ≈ 0.2272.

Based on the standard value of each factor, the proportion Pij of the influencing factors of each borehole is calculated with formula 4. The entropy value ej and utility value dj of each factor are calculated with formulas 5 and 6, respectively (Table 2). If a factor has a smaller entropy value, it’s indicated that the factor has more information, it plays a greater role in the analysis of water-resisting ability, and it has a greater weight. Finally, the weight value of each influencing factor is calculated with formula 7 (Table 2).

Table 2 Weight of evaluation factors for water-resisting ability.
Full size table

Comprehensive evaluation of water-resisting ability of aquiclude

Comprehensive evaluation of water-resisting ability of the aquiclude

In the evaluation process of the water-resisting ability of the aquiclude of the coal seam floor, the comprehensive effect of various factors has to be considered, so a composite layer needs to be established, which contains the data of all factors35,36,37,38. Therefore, this paper uses the geographic information technology to superimpose the thematic maps of various influencing factors, establish composite layers containing the data of each influencing factor, and introduce the comprehensive evaluation coefficients for water-resisting ability to evaluate the water-resisting ability of the seam floor aquiclude, see the formula 8. The comprehensive evaluation coefficient for water-resisting ability of the aquiclude is the sum of the superimposed influences of various influencing factors on a certain geographic grid on the water-resisting ability of the aquiclude. Calculate the comprehensive evaluation coefficient for water-resisting ability of each geographic grid with Formula 8, and then use the Natural Breaks classification method in the GIS software to partition the comprehensive evaluation coefficients of all geographic grids, and finally form a comprehensive evaluation zone map (Fig. 9) for water-resisting ability that integrates multiple influencing factors.

$$RC=\sum\nolimits_{{i=1}}^{n} {{W_i}} \times {f_i}(x,y)$$
(8)

Where RC is the comprehensive evaluation coefficient for water-resisting ability, Wi is the weight value, fi(x, y) is the index value after normalization, x, y is the coordinate, and n is the number of factors.

Fig. 9
Fig. 9
Full size image

Comprehensive evaluation map of water-resisting ability of the 11th coal seam floor aquiclude.

From Fig. 9 it can be seen that the red and orange areas are where with weak and weaker water-resisting ability, mainly distributed in the middle and northwest of the study area, and where the effective aquiclude and the brittle rock have smaller thickness, the core adoption rate is poor, with the RQD value of 49%-57%, and the mudstone ratio is low; The green and light green areas are where with great and greater water-resisting ability, distributed extensively around the study area, and where the rock integrity is universally great, the core adoption rate is high, with the RQD value of 67%-93%, and the effective aquiclude and the brittle rock have greater thickness, with a medium to high mudstone ratio. The yellow area is the transition area between the area with weaker water-resisting ability and the area with greater water-resisting ability; the fault influenced zone and the fault fracture zone in the evaluation map are mostly areas with weaker and weaker water-resisting ability.

Model consistency check

Due to the lack of an independent dataset for validation, the same boreholes (e.g., zk02, zk03, etc.) that were used to calculate the factor weights are employed to check the internal consistency of the model. This approach is not intended to demonstrate predictive power but to ensure that the model is consistent with the data used in its construction. The consistency is assessed using Eq. 9, where a high consistency rate indicates that the model parameters are well-calibrated to the available data.

$$VA=\frac{{DF}}{{DS}} \times 100\%$$
(9)

Where VA is the accuracy rate, DF is the number of points where the evaluation results are consistent with reality; DS is the total number of verification points.

For instance, the borehole zk03 has an effective aquiclude thickness of 1.67 m, which is thin, an RQD value of 0.64 indicating poor rock integrity, and a low mudstone ratio of 0.19. The model correctly identifies this as having weak water-resisting ability, demonstrating internal consistency. Similarly, other boreholes such as zk02, zk04, zk05, zk10, zk17, and zk18 show a consistency rate of 100%. This high consistency rate suggests that the model is well-tuned to the available data, although it does not validate predictive performance. The distribution of these points is shown in Fig. 10. It is important to note that this consistency check is a limitation of the study, and future work should include independent data for proper validation.

Fig. 10
Fig. 10
Full size image

Verify drill hole distribution map.

Discussion

Comparative analysis with existing studies

The application of the entropy weight method in evaluating the water-resisting ability of coal seam floor aquiclude has been explored in previous studies, but this research integrates it with GIS technology to provide a spatially explicit assessment, which represents a significant advancement. For instance, Zhang et al. utilized the entropy weight method in northern China coalfields to assess groundwater inrush vulnerability, focusing on factors like aquifer thickness and hydraulic conductivity32. Similarly, Qiu et al. applied entropy weights in shale gas areas to evaluate karst aquifer water abundance, emphasizing data-driven objectivity33. In contrast, our study selects six specific factors (e.g., effective aquiclude thickness, RQD, and fault zones) tailored to the hydrogeological conditions of Longfeng Coal Mine, leveraging GIS for spatial superposition and visualization. This approach not only enhances the objectivity of weight determination by minimizing subjective bias but also enables intuitive risk zoning, as demonstrated in the comprehensive evaluation map (Fig. 9). Compared to traditional methods that rely on expert judgment (e.g., analytic hierarchy process), our GIS-entropy weight fusion offers a more reproducible and scalable framework for similar karst mining environments, such as those in Guizhou Province. However, it is important to note that while this method effectively handles multi-factor interactions, its reliance on data quality and availability may limit applicability in data-scarce regions, a common challenge in coal mine hydrogeology.

Limitations and future research directions

Despite its advantages, this study has limitations that warrant further investigation. The entropy weight method, though objective, is sensitive to data variability and may not fully capture expert knowledge, potentially overlooking site-specific nuances. For example, in cases where factors like rock mechanical properties are underrepresented due to data constraints, as in Longfeng Mine, the weights might not reflect true geological complexities. Additionally, while our evaluation factors were chosen based on representativeness and data availability, they may not encompass all influential parameters, such as in-situ stress or long-term mining-induced changes, which have been highlighted in studies like Chen. (2023) on fault zones34. Future research could integrate subjective methods (e.g., fuzzy logic) with entropy weights to balance objectivity and expertise, or expand data sources through advanced monitoring techniques. Moreover, applying this model to other mines with similar conditions, such as those in karst areas, would validate its generalizability and refine the factor selection process.

Conclusions

  1. 1.

    A full understanding of the water-resisting ability of the coal seam floor aquiclude is the key to prevent water inrush from the coal seam floor. In this paper, based on considerations of the availability of data and the representativeness of factors, the effective aquiclude thickness, the brittle rock thickness of aquiclude, the RQD of aquiclude, the consumption of drilling fluid, the mudstone ratio of aquiclude and the fault influenced zone and fault fracture zone are used as the evaluation factors of the water-resisting ability of the coal seam floor aquiclude, which can provide a certain theoretical basis for the research on the water-resisting ability of the coal seam floor aquiclude.

  2. 2.

    Combined with the non-linear mathematical method—entropy weight method, a comprehensive evaluation model for the water-resisting ability of the coal seam floor qauiclude is established. Compared with the subjective weighting method, the entropy weight method is more objective and can better interpret the results. With the GIS technology, the collected data on evaluation factors for the water-resisting ability of the aquiclude was processed, and the thematic map of the spatial distribution of the index values of each evaluation factor was made, and the thematic map was compositely superposed and evaluated, thereby presenting the evaluation area map on water-resisting ability of the coal seam floor aquiclude of Longfeng Coal Mine.

  3. 3.

    The multi-source information fusion evaluation method based on the GIS-entropy weight method for the water-resisting ability of the coal seam floor aquiclude can comprehensively and objectively reflect that the water-resisting ability of the aquiclude is controlled by multiple factors and has a very complex formation mechanism. Meanwhile, the evaluation results are displayed clearly and visually, which can better solve the problem of water-resisting ability evaluation of aquiclude and provide a scientific basis for the prevention and control of water inrush disasters from the coal floor.

  4. 4.

    The applicability and limitations of the proposed method should be considered for broader implementation. In this study, the method was tailored to the specific hydrogeological conditions of Longfeng Coal Mine, such as karst aquifer characteristics and fault distributions. Its universality may vary in other mining areas with different geological settings, e.g., regions with distinct rock types, aquifer pressures, or tectonic activities. The method’s reliability depends on data availability for the selected factors; in data-scarce regions, alternative factors or adjustments might be necessary. To enhance generalizability, future research should focus on testing this approach in diverse coal mines, optimizing factor selection through sensitivity analysis, and integrating it with emerging technologies like machine learning for improved adaptability and risk prediction. This direction will help validate the method’s robustness and promote its application in global mining water hazard prevention.