Geospatial Foundation Models: Recent Advances and Applications

Foundation models are deep learning models trained on massive datasets. Recent advancements have made them capable of performing a wide range of general tasks, including language processing, summarization, code generation, problem-solving, and reasoning. Geospatial foundation models are specifically trained on large-scale geospatial and temporal data. While the capabilities of generalpurpose foundation models have been demonstrated through numerous popular applications, such as natural language generation, question answering, and text summarization, the applications of geospatial foundation models are still in the exploratory stage. In this article, we summarize recent advancements in geospatial foundation models and describe various applications.