Advances in geocomputation and geospatial artificial intelligence (GeoAI) for mapping

TLDR

Geocomputation and GeoAI are advancing GIS and Earth observation, yet challenges remain in integrating diverse geospatial features and applying them to natural, built, and social environments, while geospatial and Earth data are essential for revealing patterns and informing decision‑making. This editorial offers a comprehensive overview of GeoAI applications in mapping, classifying them into four domains—buildings and infrastructure, land use, natural environment and hazards, and social issues—and outlines future research challenges and opportunities. The authors categorize GeoAI mapping applications into four domains and summarize case‑study data sources into seven types, including in‑situ, geospatial datasets, crowdsourced data, remote sensing, photogrammetry, LiDAR, and statistical data. GeoAI has enhanced traditional geospatial analysis and mapping, transforming how complex human–natural systems are understood and managed.

Abstract

Geocomputation and geospatial artificial intelligence (GeoAI) have essential roles in advancing geographic information science (GIS) and Earth observation to a new stage. GeoAI has enhanced traditional geospatial analysis and mapping, altering the methods for understanding and managing complex human–natural systems. However, there are still challenges in various aspects of geospatial applications related to natural, built, and social environments, and in integrating unique geospatial features into GeoAI models. Meanwhile, geospatial and Earth data are critical components in geocomputation and GeoAI studies, as they can effectively reveal geospatial patterns, factors, relationships, and decision-making processes. This editorial provides a comprehensive overview of geocomputation and GeoAI applications in mapping, classifying them into four categories: (i) buildings and infrastructure, (ii) land use analysis, (iii) natural environment and hazards, and (iv) social issues and human activities. In addition, the editorial summarizes geospatial and Earth data in case studies into seven categories, including in-situ data, geospatial datasets, crowdsourced geospatial data (i.e., geospatial big data), remote sensing data, photogrammetry data, LiDAR, and statistical data. Finally, the editorial presents challenges and opportunities for future research.