Nowadays, we have the emergence and abundance of many different data repositories and archival systems for scientific data discovery, use, and analysis. With the burgeoning data sharing platforms available, this study addresses how natural resources and environmental scientists navigate these diverse data sources, what their concerns and value propositions are towards multiple data discovery channels, and most importantly, how they perceive the characteristics and compare the functionalities of different types of data repository systems. Through a user community research of domain scientists on their data use dynamics and insights, this research provides strategies and discusses ideas on how to leverage these different platforms. Further, it proposes a top-down, novel approach to search, browsing, and visualization for dynamic exploration of environmental data.
With the advancement of information technologies and sensor networks, ubiquitous sensing is revolutionizing scientific data collection and accelerating research discovery. Driven by traditional and new sensing array paradigms (North Carolina State University, 2011) and unmanned aircraft applications, natural resources and environmental data from various monitoring sources are growing bigger, faster, and more diverse than ever before. To harness the immense power and extract valuable insights from data in this fast-developing field, robust data repository systems and rigorous database functionalities are vastly needed.
Previous research a decade ago has studied the processes that led to the creation, analysis, and publishing of ecological sensing data (e.g. Wallis et al., 2008). With the goals of developing “digital curation infrastructure " and identifying requirements for “data digital libraries, " these studies investigated habitat ecologists’ and other environmental scientists’ data practices associated with embedded sensor networks (Borgman et al., 2006;Borgman et al., 2007a;Borgman et al., 2007b). They emphasized the early involvement of data archivists in the lifecycle of collaborative ecological research. They further promoted the application of more broadly conceived digital library systems and efforts to improve scientific data discovery and reuse.
Today, with digital technologies revolutionizing scientific data collection and processing, the expectations for systematic data sharing to address grand environmental challenges are growing strong. Consequently, diverse data sharing and discovery platforms are quickly emerging and becoming abundant. Commonly known as research data repositories, they provide permanent storage and access of data through large database infrastructures to promote sharing, increased access and better visibility of research records. These include domain or disciplinary-specific, government-sponsored, scientist-hosted, and library-managed, as well as other types of data repositories and archives. With this rapid new development, much related and timely questions such as how they perform and whether they serve the needs of scientific users have not been addressed.
In the context of natural resources and environmental science, this research discusses and compares different types of data repository systems by exploring their characteristics, functionalities, strengths and weakness, as well as development strategies from a human-centric standpoint. It identifies the perspectives, experiences, and needs of scientific researchers in complex data science projects. It also provides suggestions on how to improve data repository network for user-oriented search and discovery.
The goals are to identify challenges in scientific data instrumentation and opportunities for repository system development and service improvement. By adopting the classic critical incident methodology (Flanagan, 1954) and creative scenario-building approach, this current research aims to answer the following questions:
- What are the most valuable data features, search properties, and integrative characteristics for scientific data discovery, use, and analysis? 2. How do scientists perceive the characteristics and compare the functionalities of different types of data repository systems?
To address these questions, the researcher first conducted a focus group with the team scientists of the Center for Natural Resources Assessment and Decision Support, followed by multiple individual interviews with other domain scientists in the College of Natural Resources and Environment at Virginia Tech. A total of six scientists were interviewed for a deep and cased oriented analysis of their data practices. This sample size meets the qualitative research design and sampling strategies of Yin (1994) andCreswell (2007). All interviews were semi-structured following a carefully designed protocol incorporating critical incident, story telling, and scenario building techniques to explore the self-reflective experience, typical user behaviors, and practical examples of scientists’ data work. All interviews were conducted face-to-face during the end of 2015 and over the spring of 2016. Each interview lasted from 1-2 hours and was fully recorded, manually transcribed, and carefully analyzed through open coding and axial coding for qualitative insights.
From an empirical standpoint, the Support, 2015). In such context, the research participants from the college and the center approach critical natural resources issues from many diverse angles and thus offer a cross-boundary, multidisciplinary perspective of data scholarship.
From a theoretical standpoint, this study employs social informatics perspective (Kling, 2007) to draw socio-technical understanding of human-technology relations in action. In particular, it examines how data work and knowledge practice are embedded within, enabled by, as well as constrained by technic
This content is AI-processed based on open access ArXiv data.
