Seawater intrusion is a global environmental problem. Owing to its dynamic evolution

practical monitoring requires methods that are both accurate and convenient. In recent years

geophysical methods have played an important role in monitoring coastal aquifers. As a commonly used geophysical technique and an important branch of electrical prospecting

high-density electrical resistivity methods are widely applied in geological exploration and hydrogeology

including monitoring the freshwater-seawater interface. Data interpretation for high-density electrical resistivity surveys has also been a longstanding focus of research. With advances in computer technology

machine-learning algorithms have increasingly been introduced into high-density electrical resistivity data interpretation. Among them

back-propagation(BP) neural network

extreme learning machine

and convolutional neural network are 3 commonly used and promising machine-learning approaches for monitoring subsurface fluids with high-density electrical resistivity methods. This paper summarizes and discusses their basic principles

research progress

and current applications.