QBCov: A Linked Data interface for Discrete Global Systems, a new approach to delivering coverage data on the web

We are witnessing a rapid proliferation of geoscientific and geospatial data from an increasing variety of sensors and sensor networks. This data presents great opportunities to resolve cross-disciplinary problems. However, working with it often requires an understanding of file formats and protocols seldom used outside of scientific computing, potentially limiting the data's value to other disciplines. In this paper, we present a new approach to serving satellite coverage data on the web, which improves ease-of-access using the principles of linked data. Linked data adapts the concepts and protocols of the human-readable web to machine-readable data; the number of developers familiar with web technologies makes linked data a natural choice for bringing coverages to a wider audience. Our approach to using linked data also makes it possible to efficiently service high-level SPARQL queries: for example, Retrieve all Landsat ETM+ observations of San Francisco between July and August 2016 can easily be encoded in a single query. We validate the new approach, which we call QBCov, with a reference implementation of the entire stack, including a simple web-based client for interacting with Landsat observations.