CyberGIS and Geospatial Data Science for Advancing Geomorphology

Abstract

Theoretical and practical issues in geomorphology have not been adequately addressed due to a lack of formalization and digital representation of spatial and temporal concepts, given the limitations associated with modern-day geographic information systems (GIS). Rapid advancements in geospatial technologies have resulted in new sensors and large volumes of geospatial data that have yet to be fully exploited given a variety of computational issues. Computational limitations involving storage, preprocessing, analysis, and modeling pose significant problems for Earth scientists. Consequently, advanced cyberinfrastructure is required to address geospatial data-science issues involving communication, representation, computation, information production, decision-making, and geovisualization. We identify and discuss important aspects of exploiting advances in cyberinfrastructure that involve computational scalability, artificial intelligence, and uncertainty characterization and analysis for addressing issues in the Earth sciences. Such developments can be termed cyber geographic information science and systems (cyberGIS). We discuss this important topic by addressing the significant overlap of concepts in GIS and geomorphology that can be formalized, digitally represented, implemented, and evaluated with cyberGIS. We then introduce the fundamentals of cyberinfrastructure and cyberGIS, including a discussion of the utilization of artificial intelligence and deep learning. We finally provide one case study demonstrating operational cyberGIS capabilities.