National Committee on Water EngineeringPlease note that this paper is held here for archival purposes, the most up to date papers are currently held on the Engineers Australia website.
The Australian climate is noted for its marked variability of extremes. The possibility of rapidly occurring climate change due to both natural and anthropogenic factors necessitates the acknowledgement of the uncertainty of future climate and the development of contingency planning for adaptation to a changing climate. This paper aims to provide an Australian context to the current global concern over climate change. This paper also attempts to provide a more rational approach to managing water resources and hydrological issues given the historical variability of the Australian climate and the uncertainty of its future climate.
Global circulation and climate is characterised by a combination of both choatic and persistent (non-chaotic factors). Climate variability is best characterised according to the timescale of that variability, as below:
Whilst inter-annual climate variability such as enhanced and depleted seasonal rainfalls induced by the El Nino-Southern Oscillation (ENSO) are largely predictable, longer term trends in climate are the result of complex interacting physical processes that are not well understood. The predictability of climate at characteristic timescales is elaborated below;
Despite being identified some eighty years ago (Walker, 1923), recent research has only just begun implementing the predictability of seasonal rainfall and runoff as a function of the El Nino-Southern Oscillation (ENSO). The development of the extreme phases of ENSO (the ‘warm’ El Nino and ‘cold’ La Nina) can be predicted with some accuracy up to 9 to 12 months in advance. Once an extreme phase is initiated, the predicted impacts on rainfall across Australia are well known and reliable (in terms of the seasons affected and the degree of rainfall enhancement/depletion). More recent research has identified a further inter-decadal control on ENSO behaviour which modulates the predictability of rainfall and is discussed in the following section.
Recent research has identified longer-term persistent anomalies in the Pacific Ocean which affect the strength of ENSO impacts across Australia (Mantua et al., 1997; Power et al., 1999). Two indices have been developed, termed the ‘Pacific Decadal Oscillation’ (or PDO) and the ‘Inter-decadal Pacific Oscillation’ (or IPO). These indices reveal persistent anomalous warming and cooling of the Pacific Ocean which in turn affects both the predictability and impact of extreme ENSO phases across Australia. The developed indices reveal that the persistent states (warm and cool) exist for average periods of about 10 years. Whilst the causal factors of this variability remain unidentified, their persistent nature means that significant insight can be gained into expected climate variability over periods of up to 10 years in advance.
Climate variability is also apparent at multi-decadal (20-50 years) time scales. Figure 1 shows the global temperature record over the twentieth century. It is apparent that superimposed on the general warming trend is a multi-decadal signal indicating a cooling of global temperatures between 1945 and 1975. These changes in global temperature trends have recently been associated with changes in climate on a global scale. In terms of the Australian climate, a substantial and rapid change in climate state is apparent across eastern Australia in 1945 (Franks and Kuczera, 2002). Figure 2 shows a marked change in flood frequency at one representative gauge from New South Wales. Similarly, a substantial shift in climate has been noted to have occurred across western Australia corresponding to 1975. Recent research has indicated that this multi-decadal variability of global trends may be the result of chaotic interactions between the oceans and atmosphere, whilst other observers have noted the coherence of this signal to changes in solar forcing (Reid, 1991; White et al., 1997). Whilst the casual mechanisms of this mode of variability remain undefined, the effects have been substantial. Irrespective of mechanism, this mode of variability is a re-occurring aspect of natural climate variability which places significant uncertainty on attempts to predict future climate over longer periods.
Figure 1 – Global SST anomalies Figure 2 – Changing flood risk pre- (1920-1991) and post- 1945
As noted earlier, climate at any given location is a singular result of the inter-play of many complex physical processes. Some of these processes are increasingly better understood however the casual physics behind the majority of climate processes are not. Attempts to predict the future climate are almost exclusively based on the use of General Circulation Models (GCMs), which purport to contain accurate descriptions of the physics of climate. These models do not accurately reproduce the modes of climate variability listed above. Furthermore, there are more fundamental deficiencies in GCMs that limit their utility as reliable predictors of future climate. These deficiencies include, but are not limited to, the descriptions of;
Given these and other uncertainties, contemporary GCM climate predictions are not reliable. GCMs should be viewed as interpretive and experimental tools, not predictive models. Whilst some uncertainties can be propagated through climate models, structural flaws (such as the absence of a key physical process) are not assessed, because GCMs are never rigorously tested as hypotheses. It is therefore clear that the future Australian climate is entirely uncertain.
In the absence of reliable predictions of future climate, much can nonetheless be done to minimise the impacts of both climate variability and change. Practising hydrologists and water resource managers can evaluate hydrological and water resource designs in the light of historic climate variability at a range of temporal scales (seasonal through to multi-decadal). In the case of predictable climate variability, adaptive management of water resources can be developed to minimise impacts. In the case of largely unpredictable climate variability and change, informed risk analysis can be pursued to provide robust hydrological design in the light of the uncertainty of future climate. Such approaches are detailed below;
The predictability of seasonal/inter-annual variability through the ENSO process, some nine months in advance enables adaptive management of water resources. The recent insight of decadal controls on ENSO should further increase the reliability of El Nino/La Nina predictions. Whilst ENSO reliability is known throughout the meteorological community, its relevance to hydrological and water resource managers is only slowly becoming apparent. This is despite the observation that hydrologically-induced variability is often significantly greater than rainfall variability due to the non-linearity of hydrological processes.
The observation of marked multi-decadal variability in the historical record is more problematic. Due to the limited history of hydrological gauging in Australia, it is currently impossible to assess the probability of a return to an anomalous dry period (such as occurred in NSW between 1910 and 1945). Current insight into multi-decadal climate variability does not enable adaptive management in the light of a climate prediction – however, recognising the possibility of such a shift in future climate enables risk assessments to provide worst-case ‘wet’ and ‘dry’ scenarios for planned infrastructure and hydrological design. In this way, climate tolerance can be built into hydrological designs - if we cannot predict the future occurrence or non-occurrence of such shifts, we must recognise that they do nonetheless occur and that we must incorporate the possibility of such a shift into hydrological designs. Traditional hydrological design is typically empirical, viewing historical hydrological data as a stochastic variable arising from a single climate state (i.e. a static climate). Typically these methods ignore the type of persistent climate states that are apparent in the Australian records, and hence may either over- and under-estimate the true risk of protracted wet and dry epochs.
The possibility of climate change induced by greenhouse-gas emissions is the least predictable aspect of climate. Given the prohibitive uncertainties of climate modelling, again additional climate tolerance is required – if we are unable to predict the future climate, then we should at least acknowledge the probability of marked changes in regional climates, in tandem with the occurrence of natural multi-decadal shifts in climate. By viewing future climate as an unknown, hydrological and water resource designs may be more robustly achieved through risk assessment under alternative scenarios of wet/dry future climate.
Floods in Australia are a natural and frequently occurring event, often affecting thousands of people. Floods can cause widespread disruption to commercial and agricultural activities and property damage resulting in millions of dollars of economic loss. While it is impossible to eliminate floods, the economic and social impact of floods can be mitigated by a combination of structural and non-structural measures. Structural measures include flood mitigation dams, retarding basins, channel levees, channel improvements and preferential flooding of low value land to mitigate flood peaks downstream. Non-structural methods include land-use regulation, public education, and the implementation of flood warning systems.
In the light of the uncertainty of future climate, a comprehensive review of existing flood mitigation measures should be achieved. Economic analysis of the worth of additional measures should also be investigated to identify the most cost-effective deployment of resources. In this way, areas most vulnerable to an increase in flood risk may be prioritised.
The future risks from floods may also be effectively reduced through engaging with planners and state authorities. Measures to reduce the vulnerability to flooding include improving structural building regulations and by enforcing more stringent zoning regulations. Engineers have a key role in this process through the estimation of flood risk based on historical data and with the uncertainties of future climate taken into account. Engineers should also be actively engaged in a three-way partnership with flood insurers and state authorities to improve flood alleviation measures.
Currently, Australia operates set procedures to compensate flood victims in cases of hardship. This is a pragmatic solution, but not necessarily the best. The State may not be well geared up to assess how much compensation to pay or to administer it efficiently, whereas insurers have the systems in place to pay out fair levels of compensation. Additionally, individual States do not cover their potential obligations; there are no examples of a State deciding to arrange for reinsurance cover from foreign reinsurers – all the costs are kept within the boundaries of the country. This could be dangerous if there is a major catastrophe or series of catastrophes because the State may find itself in financial difficulties, or at the very least find it difficult to attract capital investment in future (Crichton, 2000).
Reinsurance offers a very cost-effective way to spread the risks across the economies of other countries so that if one country is hit by disaster, other countries automatically step in with support. This would necessitate the involvement of engineers in quantifying flood risks in a standard and methodological manner. An effective partnership between the State, insurers and flood engineers would enhance the availability of affordable flood insurance, whilst ensuring the necessary penetration of cover through minimal State regulation. Surpluses could additionally be used to enhance flood mitigation measures. These practices are currently under development within many countries where concern of the flooding implications of climate change is growing. Whilst insurance does not necessarily mitigate flood risk, it can go some way towards mitigating the financial burden of elevated flood risks.
Droughts are also a commonplace occurrence across Australia. Future changes in climate may increase the number of droughts experienced. Droughts affect different commercial, agricultural and potable water supply activities differently. In terms of urban water supply, whilst rainfall across Australia is typically lower than that over other continents, Australia’s relatively small population and concentration in coastal regions means that under all but the most severe of droughts, sufficient water is available for water supply. Improving urban water supply security in anticipation of more protracted droughts in the future may be most effectively achieved through a number of approaches. On a local scale these include the development of water sensitive urban design (water re-use and water harvesting schemes), with the aim of reducing demand on existing and future water resources infrastructure. Water use efficiencies may also be achieved through adoption of water sensitive technologies.
Agricultural droughts may be more problematic due to the large-scale of agricultural activities and their wider spatial distribution. Whilst some water efficiency gains may be achieved through optimal irrigation and water use, it is likely that improved water security may be best achieved through developing improved infrastructure. This is not without significant problems due to the often contentious issues of environmental flows in natural waterways and the effect of dams on ecological aspects of river health. Additionally, infrastructure is typically expensive to build and maintain. Any option of further developing infrastructure would require a holistic cost-benefit analysis in tandem with possible alternatives to identify regional areas, if any, that might be judged to be economically and ecologically sustainable.
Marked climate variability is an intrinsic aspect of the Australia climate. The probability of future climate variability and change represent key threats to many Australians. Current predictive insight into the future Australian climate is, at best limited but also possibly misleading as climate processes are complex and are still not well understood. Confidence in current GCM predictions is almost exclusively overplayed. A common insurance industry practice is to define risk as the probability of occurrence multiplied by the consequence of that occurrence. If the scientific community cannot provide reliable predictions of future climate (and in particular climate extremities), then the focus must turn to reducing the consequences of possible changes in climate. In this light, engineers can best contribute by acknowledging the uncertainty of future climate and by incorporating known climate variability into current practices for drought and flood prone designs. Increasing climate tolerance should be an intuitively appealing response to predictive uncertainty.
The potential impacts of climate variability and change on Australia are wide-ranging and rigorous risk analysis of these threats requires analyses that cross traditional disciplinary boundaries. It is believed that a truly multi-disciplinary approach is needed to evaluate these threats. To this aim, mechanisms are required that integrate knowledge and provide a suitable forum for a more holistic approach to climate impacts. This may be best served by the organisation of specific conferences and workshops devoted to these aims, but with broad representation from relevant stakeholders. The Institution of Engineers, Australia, is ideally suited for coordinating such a role, as a key institutional body responsible for practical adaptive solutions to climate impacts.
Crichton, D., 2001 The Implications of Climate Change for the Insurance Industry – an update and outlook to 2020. Building Research Establishment, Watford, England.
Franks, S.W. and Kuczera, G., 2002. Flood frequency analysis – Evidence and implications of secular climate change, New South Wales, Wat. Resour. Res., in press.
Mantua, N.J., S.R. Hare, Y. Zhang, J.M. Wallace and R.C. Francis, 1997. A Pacific decadal climate oscillation with impacts on salmon, Bull. Amer. Meterol. Soc., 78, 1069-1079.
Power, S., Casey, T., Folland, C., Colman, A., and Mehta, V., 1999. Inter-decadal modulation of the impact of ENSO on Australia, Clim. Dynamics, 15(5), 319-324.
Reid, G.C., 1991. Solar total irradiance variations and the global sea surface temperature record, J. Geophys. Res., 96, 2835-2844.
White, W.B, Lean, J., Cayan, D.R. and Dettinger, M.D., 1997. Response of global upper ocean temperature to changing solar irradiance, J. Geophys. Res., 102(C2), 3255-3266.
For further information, please contact:
National Committee on Water Engineering
The Institution of Engineers, Australia Telephone: (02) 6270 6555
11 National Circuit Facsimile: (02) 6273 1488
BARTON ACT 26000. WWW: ieaust.org.au
Prepared on behalf of the National Committee on Water Engineering by Stewart Franks (ceswf@civeng.newcastle.edu.au).
Last updated May 2002.
Last Modified:
Thursday, October 7, 2010 20:45
Comments to: peter.brady@eng.uts.edu.au