Towards good governance of data: A case study in geoscience data governance

Abstract

Scientific data constitute the objective information derived from activities in science and technology, serving as an essential component of scientific research outcomes. With the ongoing growth in investments in science and technology, coupled with the continuous improvement in scientific and technological innovation capabilities, a substantial volume of scientific data is consistently generated. The abundance of scientific data are gradually transforming into a crucial strategic resource within the domain of science and technology, driving the shift in scientific research paradigms towards a data-intensive approach. Geosciences stand out as a quintessential example of a data-intensive scientific discipline. With the rapid advancement of Earth observation and Internet of Things technologies, copious volumes of data are being amassed across various spheres of the Earth. This progressive accumulation heralds the emergence of the Big Earth Data era, playing a central role in propelling a novel research paradigm within geoscience and advancing research in Earth system science. As a result of policy restrictions, cultural factors, and technological barriers, a significant portion of geoscience data remains unavailable to the general public and scientific research personnel. The unavailableness of these data limits the realization of their public and scientific value and leads to a substantial waste of invaluable data resources. Scientific data centers, operating as the central hubs for data storage, management, and operations, play a crucial role in connecting data contributors and data users, thus occupying a pivotal position in data governance in Big Data era. To optimize the potential scientific value of geoscience data, it is crucial to leverage scientific data centers as pivotal platforms for promoting broader sharing of geoscience data. This objective can be achieved through a combination of top-down policy support and bottom-up incentive mechanisms, ultimately culminating in the establishment of good governance of scientific data. Drawing inspiration from well-established data-sharing principles like FAIR, CARE, and TRUST, scientific data centers should delve deeper into the unique data requirements and challenges inherent in contemporary geoscience research. Subsequently, they should formulate targeted management strategies that span the entire data lifecycle, encompassing broader open sharing and the development of sustainable data governance solutions. Furthermore, as integral components of information infrastructure, scientific data centers should seize the opportunities presented by machine learning (ML) and artificial intelligence (AI) to advance the development of information infrastructure. This requires a swift transformation from their traditional roles as passive data repositories to active data laboratories. It is essential to recognize that a diverse range of data sources, particularly non-traditional ones such as social sensing data, often exhibit fragmented, heterogeneous, and low-quality characteristics, making them susceptible to forming “information islands”. To address this challenge, scientific data centers should ingeniously incorporate AI and ML, innovatively creating innovative data integration frameworks that bridge both the natural and human societal realms. This approach, in turn, facilitates the generation of high-quality integrated data products that enable geoscience research. In addition, scientific data centers should also serve as leaders and pioneers in the Big Data revolution, advancing cutting-edge Big Data analysis methods and serving as the driving force behind AI for geoscience.