In recent decades, machine learning paradigm has been introduced, applied, and developed in geosciences and remote  sensing  disciplines.   Deep learning has demonstrated great capability for geodata processing, information extraction, and applications. However, it is still challenging  to further improve the machine learning paradigm  to achieve our scientific goals. There are a lot of theoretical and methodological issues in physics-informed machine learning, when developing the Geo-Artificial Intelligent (GeoAI). Meanwhile, the spatio-temporal heterogeneity  and limited  samples are also unsolved problems.

Therefore, this special issue will report the most recent work in machine learning paradigm-based research in geosciences and remote sensing. These studies will deepen our understanding on this topic and the way forward.

Typical topics are listed but to limited below:

Ø New concepts in machine learning based geosciences research

Ø Geoscience sampling data in machine learning

Ø eXplainable AI (XAI) in geosciences

Ø Intelligent geoscience data fusion

Ø Intelligent Object identification in remote sensing

Ø Geo-knowledge mining by machine learning

Ø Agricultural, hydrological, climate, and environmental applications