Identification of inrush water recharge sources using hydrochemistry and stable isotopes: A case study of Mindong No. 1 coal mine in north-east Inner Mongolia, China

Abstract

Rapid identification of inrush water sources is vital for the safe operation of a coal mine. Hydrogeochemical (fuzzy comprehensive evaluation method and cluster analysis method) and isotope analyses are applied to identify the inrush water sources of the Mindong No. 1 mine, which is located in northeast Inner Mongolia, China. The clustering analysis and isotope analysis results show that the inrush water sources are from aquifer 1 (A1), aquifer 2 (A2) and Yimin river. However, fuzzy comprehensive evaluation shows that the inrush water sources are from A2, aquifer 3 (A3) and Yimin river. Considering the hydrogeological conditions of the study area, it is concluded that the inrush water sources are A1, A2 and Yimin river, with mixing ratios of 30.8%, 60.6% and 8.6%, respectively. The application of multiple methods makes the conclusion more reliable. Additionally, this study improves the speed and effectiveness of the identification of inrush water sources in coal mines and provides a practical reference for research related to mine water inrush to ensure the safe operation of coal mines.

Hydrochemical differences between river water and groundwater in Suzhou, Northern Anhui Province, China

“…Previous studies have shown that hydrochemistry was a useful tool for identification of the interactions between different water bodies, especially the major ions and stable H-O isotope [4,5,[16][17][18][19]. As mentioned above, the concentrations of most ions in river water are obviously higher than those in groundwater (such as Na + , K + , Mg 2+ , Cl − , and SO 4 2− ), whereas the shallow and deep groundwater show similar chemical concentrations.…”

“…To be a powerful mathematical statistical analysis method, cluster analysis has been widely used in ecology, geochemistry, medicine, and other fields [13][14][15]. In the study of hydro-geochemistry, cluster analysis has often been used for the evaluation of the hydraulic connection between different water bodies [16][17][18][19]. In this study, two types of clustering methods (k-means and Q-type) based on factoextra and igragh packages in RStudio have been chosen for calculation [20].…”

Deep learning model based on big data for water source discrimination in an underground multiaquifer coal mine

“…However, there are limited coal mine water quality analysis indicators, and they generally only include the abovementioned conventional hydrogeochemical variables (Wang et al 2019). In this study, without increasing the testing and calculation workloads, considering the selected discriminant indexes in previous research results (Wu et al 2019a, b;Yan et al 2020a, b;Guan et al 2019;Huang et al 2018), nine variables of the collected water samples, such as TDS, Na + + K + , Ca 2+ , Mg 2+ , Cl − , SO 4 2− , HCO 3 − , CO 3 2− , and pH, were used as model discriminant indexes. To test the correlation between these discriminant indexes and determine the amount of redundant information in the sample data, PASW Statistics 18.0.0 software was used to analyze the correlation among the water sample discriminant indexes (Table 2).…”

“…This provides a basis for the discrimination of mine water sources based on hydrochemical characteristics. There are many ways to distinguish water inrush sources by considering the water chemical characteristics of aquifers, mainly based on water chemical data combined with related mathematical models to realize water source discrimination, such as the Fisher discriminant (Wu et al 2019a, b;Hou et al 2020), Bayes discriminant (Qian et al 2018;Yan et al 2020a, b;Zhang et al 2020a, b), distance discrimination method (Wang et al 2011;Liu et al 2019), gray relational analysis (Qiu et al 2017;Huang et al 2017), and cluster analysis (Guan et al 2019;Zhang et al 2019;Wang et al 2020). In addition, when there are many influencing factors of mine water inrush, scholars, to avoid information overlap between the various variables, have applied a combination of principal component analysis (PCA) and discriminant methods to reduce the model dimensionality, decrease the influence of redundant information, and improve the model effectiveness, e.g., the PCA-Fisher method, PCA-Bayes method, and PCAcluster analysis model (Lu et al 2012;Huang et al 2018;Li et al 2020;Zhang and Yao 2020;Huang et al 2019).…”

Source identification of mine water inrush based on the exponential whitenization function and the grey situation decision model

“…In recent years, some quantitative and semiquantitative methods have been widely used in the identification of water inrush sources. They are generally combined with multivariate statistics, such as the grey correlation analysis method, the grey clustering method, fuzzy comprehensive evaluation, artificial neural network identification technology, fussy grey decision-making, stepwise discriminant analysis, and the support vector machine method (Huang and Wang, 2018;Guan et al, 2019;Lipshutz et al, 2021). Computer technology is also gradually applied to water source identification, such as the Fisher Linear Discriminant (FLD), the Support Vector Machine (SVM), the Bayesian Model, and the Extreme Learning Machine (ELM), overcoming the influence of human factors in the identification process and improving the recognition accuracy (Dong et al, 2019;Zhang and Yao, 2020).…”