Applications of computational chemistry, artificial intelligence, and machine learning in aquatic chemistry research

Abstract The ever-looming water pollution has caused waterborne diseases, destruction of biodiversity, and unsafe potable water, resulting in millions of deaths every year. Although mounting efforts were exerted to tackle these serious issues, one cannot solve all the problems by manpower alone, not to mention studies that require long period monitoring on water quality (e.g., eutrophication). Therefore, it is of vital need to develop new approaches which are more intelligent, convenient, and less hazardous to perform. Computer science and engineering, which has gained rapid advancement in recent years, has been widely applied in many other fundamental disciplines such as aquatic chemistry research. For example, computational chemistry, which uses first-principles or empirical methods, has been extensively applied to predict transformation behaviors of pollutants in natural and engineered water systems. Additionally, remarkable advancements of artificial intelligence, including its subset machine learning, have become major problem-solving techniques in this area. In this context, we summarize primary applications of computational chemistry, artificial intelligence, and machine learning in aquatic chemistry research. Meanwhile, challenges along the development process were brought out to inspire greater efforts. We aimed to provide possibilities for better understanding on aquatic chemistry from a distinctive perspective, appealing to scientists and engineers to develop advanced solutions for water pollution.