Machine learning is one of the widely used techniques to pattern recognition. Use of the machine learning tools is becoming a more accessible approach for predictive model development in preventing engineering disaster. The objective of the research is to for estimation of water source using the machine learning tools. Random forest classification is a popular machine learning method for developing prediction models in many research settings. The type of mine water in the Pingdingshan coalfield is classified into surface water, Quaternary pore water, Carboniferous limestone karst water, Permian sandstone water, and Cambrian limestone karst water. Each type of water is encoded with the number 0–4. On the basis of hydrochemical data processing, a random forests model is designed and trained with the hydrochemical data. With respect to the predictive accuracy and robustness, fourfold cross-validation (CV) is adopted for the model training. The results show that the random forests model presented here provides significant guidance for the discrimination of mine water.
Geothermal water samples taken from deep aquifers within the city of Kaifeng at depths between 800 and 1650 m were analyzed for conventional water chemical compositions and stable isotopes. These results were then combined with the deuterium excess parameter (d value), and the contribution ratios of different carbon sources were calculated along with distributional characteristics and data on the migration and transformation of geothermal water. These results included the conventional water chemical group, hydrogen, and oxygen isotopes (δD-δ18O), dissolved inorganic carbon (DIC) and associated isotopes (δ13CDIC). The results of this study show that geothermal water in the city of Kaifeng is weakly alkaline, water chemistry mostly comprises a HCO3-Na type, and the range of variation of δD is between −76.12‰ and −70.48‰, (average: −74.25‰), while the range of variation of δ18O is between −11.08‰ and −9.41‰ (average: −10.15‰). Data show that values of d vary between 1.3‰ and 13.3‰ (average: 6.91‰), while DIC content is between 91.523 and 156.969 mg/L (average: 127.158 mg/L). The recorded range of δ13CDIC was between −10.160‰ and −6.386‰ (average: −9.019‰). The results presented in this study show that as depth increases, so do δD and δ18O, while d values decrease and DIC content and δ13CDIC gradually increase. Thus, δD, δ18O, d values, DIC, and δ13CDIC can all be used as proxies for the burial characteristics of geothermal water. Because data show that the changes in d values and DIC content are larger along the direction of geothermal water flow, so these proxies can be used to indicate migration. This study also shows demonstrates that the main source of DIC in geothermal water is CO2thathas a biological origin in soils, as well as the dissolution of carbonate minerals in surrounding rocks. Thus, as depth increases, the contribution of soil biogenic carbon sources to DIC decreases while the influence of carbonate dissolution on DIC increases.