Application of AI Identification Method and Technology to Boron Isotope Geochemical Process and Provenance Tracing of Water Pollution in River Basins

Abstract

River water is the most important water source that people can use. Since the 20th century, human influence on river courses has become increasingly serious. The quantitative analysis of water quality is even more difficult. According to the characteristics of Fenhe water chemistry, pollution time and pollution control factors, the contribution rate of people in the polluted water body is not clear. Therefore, this paper aims to use AI identification methods and technologies to study water pollution and provenance tracing. The combination of major elements, trace elements and stable isotopes was used to study the chemical characteristics, water quality status, and sources of pollution of the Fenhe water in the Fenhe area. Because the water contains a large number of pollution sources, it is difficult to find the source using traditional methods. Using correlation analysis, principal component analysis, multi-factor regression analysis, trend analysis and other methods, the macroelements and trace elements in the water body of the Fenhe River were analyzed. The boron sources in the Fenhe river were qualitatively and quantitatively analyzed using mass spectrometry equilibrium equation. Using the boron isotope value of the river, it showed a spatial variation of upstream (+5.1‰) < middlestream (+8.6‰) < downstream (+9.5‰) in dry season, and showed a spatial variation of upstream (+6.1‰) < downstream (+7.2‰) < middlestream (+9.0‰) in the wet season. The contribution of silicate to B is calculated by subtracting the contribution of other resources from the comprehensive contribution rate. It is found that the contribution of silicate is about 38.8%, 22% in dry season and 49.2%, 17% in wet season. The research results have provided a reliable scientific basis for the protection of water resources and pollution control in the Fenhe River Basin. Therefore, the above research confirms the role of AI identification method in the process of boron isotope geochemistry and provenance tracing of water pollution in river basins.