This study contributes to the conference theme of information science processes and practices by examining the research data discovery contexts of ecological and social scientists who reuse datasets for research. The aim of the research was to gain insight into the data discovery contexts of these scientists and to understand the data needs related to data discoverability and reuse. The study identified four dimensions of data needs, including research processes, making sense of data, data reuse, and data access. Additionally, a conceptualization of data needs within the context of data reuse was proposed, which has not been thoroughly examined in previous studies. The study employed a mixed-method approach within the post-positivist research paradigm to identify the different contexts in which data is discovered. A combination of survey and in-depth interview techniques were used to investigate the broader contexts of data discovery in people s information-seeking processes. The critical incident technique was used to elicit the contexts of data discovery, and the interview protocol was structured based on the stages of a data lifecycle. Interviews were conducted with 24 participants from three organizations, including TERN, ADA, and CSIRO. Participants held diverse job roles and were at different career stages. The study identified four dimensions of data needs and examined their relationship with the roles of data managers and end-users. The findings contribute to the existing literature on data needs and emphasize the potential usefulness of research data and the need for paradata. The study also suggests that anticipating the contexts of data reuse involves considering what data users may find useful. Ensuring data quality is crucial for successful data reuse, which involves having access to organizational data expertise, providing data in various formats and platforms, and ensuring sufficient data coverage. Data needs are influenced by the specific research objectives, which in turn affects the criteria for selecting and reusing data. Clear data licensing conditions are crucial to facilitate data reuse.