Key points are not available for this paper at this time.
Data about the Earth are too difficult to access, making EO exploitation inaccessible for non-experts and tedious for experts. Reasons are manifold, from intrinsic complexity to technically overloaded handling where data are presented more in a generator-centric than a user-centric manner. Under the headline of Analysis-Ready Data (ARD) significant research is going on to find ways of stripping off unneccesary burden from services. While progress is being made on improving metadata, such as by CEOS, the data perspective still is underrepresented. With our research we aim at contributing to closing this gap in particular on gridded data, so rasters and datacubes. Starting point is the data and processing model the ISO/OGC coverage standards offer; in these ecosystems, three use cases are inspected: determination of service quality parameters, automated data fusion, and ML. We find that some ARD aspects are covered, but we also spot several issues that deserve investigation and standardization effort. Broadly, these fall into the following categories: conceptual clarification, ex: pixel-in-center / pixel-in-corner; enhancing existing (and otherwise proven) standards, ex: establish a framework for units of measure amenable to autoamtic conversion, similar to coordinate reference systems; improved standards governance, ex: avoiding competing standards known to be not interoperable. In our talk we present results achieved from work in OGC Testbed-19 and EU FAIRiCUBE. We discuss gaps found and present suggestions for improvement towards easier and more reliable consumption of EO data by humans and machines.
Peter Baumann (2024) studied this question.