Published works

Australia is subject to significant impacts from natural hazards, the risk of which is increasing due to population growth and climate change (e.g. Newman et al., 2017).  If we could, it would be cheaper, and cause less harm, to deal with some of these threats in advance to make our communities more resilient. The difficulty is knowing how to prioritise what to plan for, and how to make best use of available resources. This difficulty can be addressed by developing risk reduction strategies, which requires a good understanding of not only current, but also future risk, together with the effectiveness of risk reduction options. 

The projected increase in risk, together with the awareness of the complexity of the underlying dynamics affecting this risk, and the need to assess the impact of various risk reduction options, has led to the recognition that there is an urgent need to better understand the components of disaster risk and their dynamics. The use of decision support systems (DSSs) is advantageous in this context, as such systems (1) are transparent and can quantify the expected benefits of mitigation investiture across multiple criteria, enabling strong arguments for the selection of particular mitigation options to be made, (2) can be used to assess the likelihood and consequences of disasters across multiple criteria, resulting in less bias when assessing the relative benefits of mitigation options, and (3) can make use of formal decision science techniques to find portfolios of mitigation options most effective under a range of (future) conditions.

However, based on a literature review undertaken at the start of the project, DSSs for disaster risk reduction have tended to focus on disaster preparedness and the immediate and post-crisis response to emergencies, instead of strategic planning of mitigation options. Of those DSSs that have focused on mitigation, none have considered dynamics and drivers across the three components of disaster risk – hazard, exposure and vulnerability, or the integrated assessment of a range of risk reduction options, both structural and non-structural.

The DSS developed, applied and used in this project, UNHaRMED (Unified Natural Hazard Risk Mitigation Exploratory Decision Support System), aims to fill this gap. Over the past ten years, the University of Adelaide (UoA) and the Research Institute for Knowledge Systems (RIKS) have worked on a framework for supporting disaster risk management, building on the strength of both organisations, which have significant expertise in developing and using DSSs with and for policy-makers. UoA, in addition, has a strong background in hydrological and complex systems modelling and RIKS in modelling future land use developments, relevant for simulating dynamic exposure profiles.   

The aim of this project was therefore to combine the strengths of both organisations in developing a DSS to assist government agencies to better understand how risks arising from multiple natural  hazards change over space and time under different plausible future conditions (e.g. climate change, population growth, economic development), as well as the relative effectiveness of different risk reduction strategies (e.g. structural measures, land use planning, land management, building code changes etc.). The development of UNHaRMED has been supported by the inputs of many stakeholders around Australia, shaping what the tool should be able to do and what it should look like. Together with the (proposed) end-users, the generic system has been applied to four regions of Australia: Extended Adelaide, Greater and Peri-Urban Melbourne, Tasmania and South-West Western Australia.

Accompanying the UNHaRMED system is a framework that facilitates its development and supports its use by organisations such that it: (i) is able to deal with complex problems in a systematic and transparent manner; (ii) makes best use of available sources of data and information; (iii) is adaptable/flexible; (iv) deals with multiple, competing objectives; (v) identifies mitigation options that represent the best possible (optimal) trade-offs between objectives; (vi) deals with uncertainty; (vii) enhances understanding of the side effects and impacts of different combinations of policy options; and (viii) adopts an interdisciplinary approach across various policy fields.

To demonstrate its use potential, several policy cases and scenario studies have been conducted with UNHaRMED in those regions for which applications have been developed. These cases show how UNHaRMED can add value over commonly used existing practices.

The main lessons learnt from these studies were:

• Changes in future risk are non-linear, both spatially and temporally, highlighting the importance of quantifying the spatio-temporal dynamics of future risk explicitly.
• It is vital to consider changes in both population growth and socio-economic development (affecting future exposure) and climate change (affecting future hazard) in the assessment of future risk, especially as the impact of the former can be significantly larger.
• The use of integrated modelling approaches, coupled with scenario analysis, is vital for enhancing understanding of future risk.
• The data requirements for UNHaRMED are significant, and the uncertainty present in the modelling results is dependent on the quality of the input data. In the absence of accurate data, reasonable assumptions can be made, but it is important to evaluate the modelling results in light of these assumptions.
• Decision support systems like UNHaRMED need to be tailored to specific decision contexts with the aid of stakeholder input. A joint participatory and modelling approach strengthens the modelling and facilitates its understanding, relevance and uptake amongst stakeholders. Moreover, it enables the incorporation of important aspects relevant to the decision-making process that cannot be modelled.
• While the application of UNHaRMED offers significant flexibility, it is likely that changes to the software are required in order to meet specific end-user needs in different decision contexts.

Next steps beyond the project include the further development of UNHaRMED from a research prototype to an operational decision support tool and its use and implementation by governments across Australia. Several activities beyond this project have already been conducted in South Australia and a pathway for implementing the system within the Western Australian government is currently being discussed.