W Hugo1*, A Rogers1, A Sahula1, H Wilson1
*1South African Environmental Observation Network, De Havilland Crescent, Persequor Park, Pretoria, 0001 South Africa)


There is a growing expectation that research output, being increasingly open, standardised, and managed in formal research data infrastructures, will be useful to policy and decision makers without much additional intervention and modification (Chen at al., 2017, OECD, 2012). We believe that this is unlikely to be feasible in the majority of cases (Hugo & Rogers, 2017). There is, then, a need for mechanisms whereby scientific evidence and operational observation data can be translated into decision and policy support metrics or indicators. The difficulty in achieving this has been highlighted more than a decade ago (Reid, 2004).

There are several reasons why improved access to scientific evidence, in particular, does not lead to improved decision and policy support:

  • The language, (vocabularies, semantics, and heuristics) adopted by the research community in a specific discipline may not translate into meaningful implementation language (Preston et al., 2015);
  • The researchers may not be in a position of influence (which includes aspects such as writing policy briefs, undertaking personal initiatives, and building up public or industry concern and interest) (Fox and Sitkin, 2015);
  • The frequency, timing, and/or certainty associated with research output is at odds with decision and policy- making cycles. Research typically progresses until there is a defensible level of certainty in statistical assessment of a result, while policy and management decisions are made within a regular cycle or as events require;
  • Scientists are not trained for, or measured by, the typical work required for decision and policy support: synthesis of scenarios and cost-benefits of such scenarios given sometimes significant uncertainty in the input data, and the need to balance cross-disciplinary concerns. Scientists tend to be specialists, while decision and policy support require a generalist approach;
  • Observation data is increasingly commoditized and no longer requires direct handling by scientists - examples being satellite observation data, weather data, and the like;
  • In the field of disaster risk specifically, scientists are potentially liable should their warnings (or lack thereof) lead to loss of infrastructure, lives, and livelihoods, and in some countries, the process of issuing warnings is regulated (Alemanno and Lauta, 2014);
  • Open availability of data and information, without value judgement, moderation, or expert interpretation often do more harm than good (Watanabe, 2012).
Several examples of frameworks aimed at translation of science into policy exist. These are sometimes formal - such as the very detailed framework developed by IPCC for translating climate science into policy - (IPCC, 2007), but also informal - such as the work to develop Essential Biodiversity Variables, loosely designed to support several Aichi Targets (Pereira et al., 2013), and the UN Sustainable Development Goals (UN, 2016).

In the paper, we propose a semantic framework for Risk and Vulnerability, and explain how the framework could assist with the development of loosely coupled, decision-ready variables for a number of risk and vulnerability- related hazards. In addition, a proposal is made in respect of the certification required within such a loosely coupled architecture, and the necessity for trust to be verifiable for the contributors to the architecture.

The semantic framework addresses aspects of proper definition and derivation of variables (in practice working towards essential variables for risk and vulnerability), the state of readiness or usability of data services (moving from raw data to ‘analysis ready’ and ‘indicator’ or ‘decision ready’ data), and aspects of trust in the value chain. The semantic framework is supported by guidance and best practice in respect of standards and specifications for participating data providers and decision support platforms.

The outputs of this work will be submitted to the Disaster Risk Working Group of the Digital Belt and Road(DBAR) initiative, and hopefully lead to enhanced use of scientific evidence and data services in support of risk and vulnerability assessment, monitoring, and mitigation. It is also anticipated that the case study for Risk and Vulnerability can be extended to include other research themes in the Future Earth initiative.


The authors wish to acknowledge contributions from the South African Weather Service (Prof Hannes Rautenbach), National Disaster Management Centre (Mr Dechlan Pillay), the Department of Environmental Affairs (Dr T Makholela and Mr T Ramaru), and Ms A le Roux and Mr G Mans of the CSIR Built Environment, all of whom have assisted with discussion and refinement of the semantic framework. The work is financially supported by the South African Department of Science and Technology.


Alemanno, Alberto and Lauta, Kristian Cedervall, The L’Aquila Seven: Re-Establishing Justice after a Natural Disaster (July 1, 2014). European Journal of Risk Regulation, 2/2014. liability-SSRN-id2461327.pdf

Chen Fang, Rajib Shaw, Md Anwarul Abedin, Salvano Briceno, Amod Mani Dixit, Manu Gupta, Wim Hugo, Jia Gensuo, Virginia Murray, Aung Myint, Pereira Joy Jacqueline, Atta-ur Rahman, Vinod K.Sharma, Sugeng Triutomo, Wang Xiaoqing, Deepthi Wickramsinghe (2017). Challenges of Disaster Risk Reduction in the Belt and Road: Contribution of DBAR, Bulletin of Chinese Academy of Sciences, 2017, 32(Z1): 52-61

Craig R. Fox and Sim B. Sitkin (2015). Bridging the divide between behavioral science & policy, Behavioral Science & Policy, Spring 2015, dividebetween-behavioral-science-and-policy.pdf

Hugo, W. and Rogers, A. (2017). Bridging the Gap Between Policy and Research Infrastructure: Risk and Vulnerability Case Study, Geophysical Research Abstracts Vol. 19, EGU2017-12568, 2017, EGU General Assembly 2017

IPCC (2007). Conceptual framework for the identification and assessment of key vulnerabilities,


Pereira et al. (2013). Essential Biodiversity Variables, Science 18 Jan 2013: Vol. 339, Issue 6117, pp. 277- 278 DOI: 10.1126/science.1229931

Preston, B.L., Mustelin, J. & Maloney, M.C. Mitig Adapt Strateg Glob Change (2015) 20: 467. doi:10.1007/s11027-013-9503-x

Reid WV (2004) Bridging the Science–Policy Divide. PLoS Biol 2(2): e27. doi:10.1371/journal.pbio.0020027

UN (2016). Sustainable Development Goals,

Watanabe, T. (2012). Lessons learned from two nuclear accidents taken place in Japan (TOKAIMURA and FUKUSHIMA) in terms of publicity of data and information, 23rd International CoDATA Conference, Taipei, Taiwan. and-fukushima-terms