Dry conditions in most parts of Australia tend to be associated with El Niño. The link between rainfall and streamflow and ENSO is statistically significant in most parts of eastern Australia (Chiew et al., 1998; Power et al., 1998). Relationships between river flows and ENSO also have been identified for some seasons in parts of New Zealand (McKerchar et al., 1998). Because of the relatively high variability of Australian rainfall, the storage capacities of Australia's large dams are about six times larger than those of European dams for the same mean annual streamflow and probability of water shortfall. In contrast, New Zealand's hydroelectricity system has a total storage capacity of about 6 weeks of national demand because of its higher and more reliable rainfall (Basher et al., 1998).

The Murray-Darling River basin is the largest in Australia and is heavily regulated by dams and weirs. About 40% of mean annual flow is used for human consumption, principally through irrigation; there is high interannual variability. Application of the CSIRO (1996a) scenarios, with their wide range of rainfall changes as a result of inclusion of both the older slab-ocean GCM and the more recent coupled AOGCM simulations, suggests a possible combination of small or larger decreases in mean annual rainfall, higher temperatures and evaporation, and a higher frequency of floods and droughts in northern Victorian rivers (Schreider et al., 1996). A study of the Macquarie River basin in NSW indicates inflow reductions on the order of 10-30% for doubled CO₂ and reduced streamflows if irrigation demand remains constant or increases (Hassall and Associates et al., 1998). Adelaide and Perth traditionally have been regarded as the most vulnerable metropolitan areas to future water supply problems, including increasing levels of salinity (Schofield et al., 1988; Williams, 1992; PMSEIC, 1998; MDBC, 1999), although Perth recently has decided to spend AU$275 million for drought-proofing (Boer, 2000). Water supplies are adequate for many coastal regions in Australia. However, drier inland areas are vulnerable to water shortages during the annual dry season and drought.

Studies by Kothavala (1999) and Arnell (1999)—using results from the U.S. National Center for Atmospheric Research (NCAR) Community Climate Model (CCMO) GCM and the HadCM2 and HadCM3 AOGCMs, respectively—show increases in drought across eastern and southern Australia. Kothavala found that the Palmer Drought Index showed longer and more severe drought in northeastern and southeastern Australia. Arnell (1999) found marked decreases in runoff over most of mainland Australia but some increases over Tasmania. For the Murray-Darling basin, he found decreases in mean flow by the 2050s ranging from about 12 to 35%, with decreases in the magnitude of 10-year maximum and minimum monthly runoff.

The only recent water supply study for New Zealand is that by Fowler (1999), based on the RSNZ (1988) and Mullan and Renwick (1990) regional climate change scenarios and three equilibrium slab-ocean GCM simulations. These models give scenarios of rainfall increases in the Auckland region, leading to the conclusion that changes in water resources most likely would be positive. Scenarios based on recent AOGCM simulations have yet to be evaluated.

Atolls and low-lying islands (e.g., some in the Torres Strait and in association with New Zealand) rely on rainwater or limited groundwater resources for water supplies. These resources are sensitive to climate variations and in some cases already are stressed by increasingly unsustainable demand and pollution caused by human activity. Saltwater intrusion into aquifers might occur through sea-level rise, more frequent storm events, possible reductions in rainfall, and increased water demand as a result of higher temperatures (see Basher et al., 1998; Chapter 17).

Until recently, water planning in Australia was driven by demand and controlled by engineers, not by economics (Smith, 1998b). This situation has changed with growing population and demand (rural and urban/industrial), including rapid growth in irrigation of high-value crops such as cotton and vineyards. There also is an increasing awareness of stress on riverine ecosystems as a result of reduced mean flows, lower peak flows, and increasing salinity and algal blooms. Higher temperatures and changed precipitation as a result of climate change generally would exacerbate these problems and sharpen competition among water users (e.g., see Hassall and Associates et al., 1998). In 1995, the Council of Australian Governments reviewed water resource policy in Australia and agreed to implement a strategic framework to achieve an efficient and sustainable water industry through processes to address water allocations, including provision of water for the environment and water-trading arrangements. The Agriculture and Resource Management Council of Australia and New Zealand subsequently commissioned a set of National Principles for the Provision of Water for Ecosystems, with the following stated goal: "To sustain and where necessary restore ecological processes and biodiversity of water-dependent ecosystems." Implementation of water reforms and national principles has resulted in the definition of conceptual frameworks and practical methods for assessing the water requirements of environmental systems.

In Australia, flow recommendations commonly are developed after water infrastructure projects and dams have been in place for some time and environmental flows implemented in river systems that already are experiencing a modified or regulated flow regime (Arthington, 1998; Arthington et al., 1998). This situation is most applicable to adaptation to climate change in existing regulated flow regimes.

The Australian National Principles require that provision of water for ecosystems should use the best scientific information available on the hydrological regimes necessary to sustain aquatic ecosystems. Ideally, environmental flow recommendations are based on establishment of quantitative relationships between flow characteristics and desired geomorphological, ecological, or water-quality outcomes. Methods are available to estimate flow-related habitat requirements of aquatic invertebrates, fish, and aquatic and riparian plants (e.g., wetted perimeter, transect methods, instream flow incremental methodology (IFIM)—see Kinhill, 1988). However, there are no standard methods for assessing flows that are relevant to maintenance of key life history processes. In the absence of robust biological indicators of response to flow regulation, recent research has advocated the use of statistical descriptors of flow regimes. These methods include maintenance of critical flow characteristics within one or two standard deviations of mean parameters (Richter et al., 1996).

In New Zealand, various pressures on riverine ecosystems have been recognized, including those from agriculture, urban usage and sewage, hydroelectricty and water supply dams, forestry and mining, and introduced pests and weeds. Management of water is covered by the Resource Management Act (RMA) of 1991. Under this Act, the intrinsic values of ecosystems, including their biodiversity and life-supporting capacity, must be considered. The emphasis has changed from multiple-use management to environmentally sustainable management (Taylor and Smith, 1997), and Maori values are explicitly recognized (Ministry for the Environment, 1999). Drought associated with ENSO has placed stress on water supplies in various parts of the country, and it is recognized that climate change could lead to further stresses, especially if there is an increase in the frequency of El Niño events (Taylor and Smith, 1997). The RMA provides a statutory basis for integrated catchment management in that regional Councils control land use, water use, and water quality. Regional plans under which water is allocated have a term of only 10 years, allowing for review to adapt to issues such as river flow changes caused by climate change.

In the context of climate change, it is relevant to ask: How much can critical features of the flow regime be changed before the system becomes seriously stressed? Finding answers to this question for a range of Australian rivers is central to the assessment and management of water allocations to sustain water-dependent systems. Climate change has yet to be systematically injected into this process, but at least the mechanisms are now in place to develop appropriate water allocations and price incentives to use water to the best advantage. There also is scope for increased application of seasonal climate forecasts in water resources management, as a tool to aid adaptation to climate variability.

Natural salinity and high water tables have been present in Australia for centuries. However, because of changes in land management—notably land clearing and irrigation—salinity is now a major environmental issue in Australia (Ghassemi et al., 1995; MDBC, 1999). About 2.5 Mha are affected in Australia, with the potential for this to increase to 12.5 Mha in the next 50 years (PMSEIC, 1999). Much of this area covers otherwise productive agricultural land. The area damaged by salinity to date represents about 4.5% of presently cultivated land, and known costs include US$130 million annually in lost agricultural production, US$100 million annually in damage to infrastructure (such as roads, fencing, and pipes); and at least US$40 million in lost environmental assets (Watson et al., 1997; PMSEIC, 1998). The average salinity of the lower Murray River (from which Adelaide draws much of its water supply) is expected to exceed the 800 EC threshold for desirable drinking water about 50% of the time by 2020.

Although climate is a key factor affecting the rate of salinization and the severity of impacts, a comprehensive assessment of the effects of climate change on this problem has not yet been carried out. Revegetation policies and associated carbon credit motivational policies designed to increase carbon sinks are likely to have a significant impact on recharge. However, global warming and dryland salinity policies need to be coordinated to maximize synergistic impacts.

In many coastal areas and oceanic islands, development and management of fresh groundwater resources are seriously constrained by the presence of seawater intrusion. Seawater intrusion is a natural phenomenon that occurs as a consequence of the density contrast between fresh and saline groundwater. If conditions remain unperturbed, the saline water body will remain stationary unless it moves under tidal influences. However, when there is pumping of freshwater, sea-level change, or changing recharge conditions, the saline body will gradually move until a new equilibrium condition is achieved (Ghassemi et al., 1996). If the sea level rises to its "best-guess" or extreme predicted value over the next century, this would significantly increase intrusion of seawater in coastal and island aquifers.

Water quality would be affected by changes in biota, particularly microfauna and flora; water temperature; CO₂ concentration; transport processes that place water, sediment, and chemicals in streams and aquifers; and the timing and volume of water flow. More intense rainfall events would increase fast runoff, soil erosion, and sediment loadings, and further deforestation and urbanization would tend to increase runoff amounts and flood wave speed. These effects would increase the risk of flash flooding, sediment load, and pollution (Basher et al., 1998). On the other hand, increases in plantation and farm forestry—in part for carbon sequestration and greenhouse mitigation purposes—would tend to reduce soil erosion and sediment loads.

Eutrophication is a major water quality problem in Australia (State of the Environment, 1996). This is a natural process, but it has been greatly accelerated in Australia by human activities, including sewage effluent and runoff from animal farms, irrigation, and stormwater. Low flow, abundant light, clear water, and warmth all encourage algal growth, which affects the taste and odor of water and can be toxic to animals, fish, and humans. Thus, local climate warming and the potential for reduced streamflow may lead to increased risk of eutrophication.

IPCC Homepage