Data infrastructures for estuarine and coastal ecological syntheses

Holistic understanding of estuarine and coastal environments across interacting domains with high-dimensional complexity can profitably be approached through data-centric synthesis studies. Synthesis has been defined as “the inferential process whereby new models are developed from analysis of multiple data sets to explain observed patternsacross a range of time and space scales.” Examples include ecological—across ecosystem components or organization levels, spatial—across spatial scales or multiple ecosystems, and temporal—across temporal scales. Though data quantity and volume are increasingly accessible, infrastructures for data sharing, management, and integration remain fractured. Integrating heterogeneous data sets is difficult yet critical. Technological and cultural obstacles hamper finding, accessing, and integrating data to answer scientific and policy questions. To investigate synthesis within the estuarine and coastal science community, we held a workshop at a coastal and estuarine research federation conference and conducted two case studies involving synthesis science. The workshop indicated that data-centric synthesis approaches are valuable for (1) hypothesis testing, (2) baseline monitoring, (3) historical perspectives, and (4) forecasting. Case studies revealed important weaknesses in current data infrastructures and highlighted opportunities for ecological synthesis science. Here, we list requirements for a coastal and estuarine data infrastructure. We model data needs and suggest directions for moving forward. For example, we propose developing communitystandards, accommodating and integrating big and small data(e.g., sensor feeds and single data sets), and digitizing ‘ dark data’ (inaccessible, non-curated, non-archived data potentially destroyed when researchers leave science).