Encouraging reusability of computational research through Data-to-Knowledge Packages - A hydrological use case.

Abstract

The growing demand for reproducible research is based on the expectation that publishing research in this form will enable its reuse and the generation of new knowledge. However, reproducibility alone does not guarantee these benefits. Users still need to make considerable efforts to understand the data and analysis code before they can reuse these components in other contexts. To address this challenge, we introduce the Data-to-Knowledge Package (D2K-Package), a collection of research materials including source code and open FAIR data, virtual labs, web API services, and computational workflows. The D2K-Package's core is the reproducible basis composed of the data and source code on which an analysis is based. This core is designed such that the other components can be derived from it. The main goal of the package is to help researchers generate new knowledge by facilitating the understanding and encouraging the reuse of reproducible research. We demonstrate the applicability of the D2K-Package with a hydrological use case which can be also used for testing, and discuss its seamless integration into the research cycle.