Computational challenges in global environmental research infrastructures

Environmental science research is increasingly dependent on the collection and analysis of large volumes of data gathered via wide-scale deployments of sensors and other observation sources. Meanwhile, researchers are being called upon to address global societal challenges that are inextricably tied to the stability of our native ecosystems. These challenges are intrinsically interdisciplinary in nature, forcing scientists to collaborate across traditional disciplinary boundaries. The role of research infrastructure in this context is to support researchers in their interactions with a host of different data sources and analytical tools, as well as with each other, but no single environmental research infrastructure can hope to fully encompass the entire research ecosystem that has arisen to support the study of environmental science. The challenge therefore is for new environmental research infrastructures to exhibit sufficient technical interoperability between the different services they offer so as to permit researchers to freely and effectively interact with the full range of research assets potentially available to them, allowing them to collaborate and conduct innovative interdisciplinary research regardless of the particular research community to which they belong. Realising this ideal however requires a broad understanding of the fundamental commonalities of environmental science research infrastructure services as well as the development and wide adoption of common foundational services. It also requires a pragmatic bridging between the different standards and controlled vocabularies currently in use or preparation by different scientific communities, a process that can be expedited by the use of a standard reference model and the use of a formal framework for semantically linking similar concepts in different contexts