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

Area of Science:

• Scientific computing
• Data science
• Research methodology

Background:

• Reproducible research is increasingly demanded for enabling knowledge reuse.
• Current reproducibility efforts often fall short, requiring significant user effort for understanding data and code.
• A gap exists in tools that facilitate the practical reuse of research components.

Purpose of the Study:

• To introduce the Data-to-Knowledge Package (D2K-Package) as a solution to enhance the reuse of reproducible research.
• To provide a framework that integrates data, source code, computational workflows, and virtual labs.
• To facilitate the generation of new knowledge by lowering barriers to understanding and reusing research outputs.

Main Methods:

• Developed the D2K-Package, a collection of research materials including source code and open FAIR data.
• Designed the package with a core reproducible basis (data and source code) from which other components are derived.
• Incorporated virtual labs, web API services, and computational workflows for enhanced usability.
• Demonstrated applicability using a hydrological use case.

Main Results:

• The D2K-Package successfully integrates diverse research materials into a cohesive unit.
• The core reproducible basis allows for the derivation of other package components.
• A hydrological use case validated the package's functionality and potential for testing.
• The package design supports seamless integration into existing research cycles.

Conclusions:

• The D2K-Package effectively addresses the challenge of making reproducible research truly reusable.
• It lowers the barrier for researchers to understand, reuse, and build upon existing research.
• The package promotes the generation of new knowledge by enhancing the utility of reproducible research outputs.