Documenting provenance of science in a state government agency

Matthew Miles, Angela London, Robin Green, Amy Teitzel, Jikke Adema

    Research output: Contribution to conferencePaperpeer-review

    Abstract

    The South Australian Department of Environment Water and Natural Resources (DEWNR) has developed a comprehensive suite of standards, tools and guidelines to improve the quality and transparency of the science we produce. The approach is named MEK (Managing Environmental Knowledge). The MEK suite of resources supports the department’s Information Management Framework and is aligned to South Australia’s Digital by Default and Open Data agendas. DEWNRs Project Management Framework (PMF) does not currently address data management in projects. The MEK fills this gap by adding steps to the project phases which enable data supply chain management. This reminds projects of the importance of early data management planning and which standards or guidelines to refer to during the execution and delivery of a project. MEK tools support project managers and scientists to explicitly document data supply chains so that provenance of intermediate data outputs and publishable products is clear and accessible. The MEK aims to (a) improve the security, integrity and availability of data produced through projects (b) establish provenance, ensure transparency and enable re-use of data products and (c) enable effective use of existing enterprise information systems. MEK resources are made available to employees through the department’s intranet. The tools include (i) data planning form for estimating resources and broad needs of a project in relation to data management, (ii) data charts that provide a visual way of describing data supply chains and (iii) data catalogue for storing detailed metadata of each element in the data charts. The guidelines that underpin these tools include: data storage describing how to make use of the various corporate systems and applications, evaluation detailing the peer review procedures for major deliverables and evaluation guidelines for other project outputs, data handling describing information classification and sensitive data handling, and publication including proofing and publishing procedures. Government agencies which rely on data to produce scientific outputs require a synergy of people, process and technologies to be effective and achieve strategic objectives. In DEWNR, an organisation which was recently formed as an amalgam of a number of departments, a disparate mix of platforms, applications and procedures have evolved over time. The result of this is that there is lack of consistency across the organisation in data management processes. The South Australian Government’s Digital by Default, Open Data and Data Sharing agendas have contributed to the need to better understand the provenance of DEWNR’s science information. Simply printing a run of documents is no longer sufficient, we must openly provide intermediate datasets and products such as models, as well as communications material. Therefore, clear provenance is critical to ensure outputs provided to downstream users are evaluated and authorised. The MEK resources are enabling the department to improve its culture around data management, deliver improved science outputs and thereby enable evidence based decision making to support the management of our State’s natural resources.

    Original languageEnglish
    Pages417-423
    Number of pages7
    Publication statusPublished - 2017
    EventMODSIM2017, 22nd International Congress on Modelling and Simulation -
    Duration: 3 Dec 2017 → …

    Conference

    ConferenceMODSIM2017, 22nd International Congress on Modelling and Simulation
    Period3/12/17 → …

    Fingerprint

    Dive into the research topics of 'Documenting provenance of science in a state government agency'. Together they form a unique fingerprint.

    Cite this