Motivation
Space and time matter not only for the obvious reason that everything happens somewhere and at some time, but because knowing where and when things happen is critical to understanding why and how they happened or will happen. Spatial data science is concerned with the representation, modeling, and simulation of spatial processes, as well as with the publication, retrieval, reuse, integration, and analysis of spatial data. It generalizes and unifies research from fields such as geographic information science, geoinformatics, geo/spatial statistics, remote sensing, and transportation studies, and fosters the application of methods developed in these fields to outside disciplines ranging from the social to the physical sciences. In doing so, research on spatial data science must address a variety of new challenges that relate to the diversity of the utilized data and the underlying conceptual models from various domains, the opportunistic reuse of existing data, the scalability of its methods, the support of users not familiar with the language and methods of traditional geographic information systems, the reproducibility of its results that are often generated by complex chains of methods, the uncertainty arising from the use of its methods and data, the visualization of complex spatiotemporal processes and data about them, and, finally, the data collection, analysis, and visualization playing out in near real-time. Spatial data science does not only utilize advanced techniques from fields such as machine learning or big data storage and retrieval, but it also contributes back to them. Recent work, for instance, has shown that spatially-explicit machine learning methods substantially outperform more general data when applied to spatial data even though this spatial component may seem of secondary importance at first glance.
The Center for Spatial Studies at the University of California, Santa Barbara is hosting a symposium entitled “Setting the Spatial Data Science Agenda.” The meeting will bring together academic and industry representatives from fields such as geographic information science, geoinformatics, geo/spatial statistics, remote sensing, and transportation studies, with interest in setting an interdisciplinary research agenda to advance spatial data science methods and practice, both from scientific and engineering viewpoints. We also invite experts from related fields and those that are producers or users of spatial data in the social and physical sciences.
Goals
Instead of being restricted by a historically grown partition into small and overlapping communities that deal with spatial data in one way or the other, the overarching goal of this symposium is to put spatial data science at the forefront of a unified field that explores the current research and application landscape to define an agenda for spatial data science for the next 10 years.
Means
About 40 invited and funded experts from academia and industry will convene to share and develop visions, insights, and best practices. Plenary presentations and intense exchanges in small breakout discussion groups offer opportunities for knowledge transfer.
Please direct any questions to Karen Doehner or Krzysztof Janowicz