4.3.2.2. Water Hazards and Water Supply

If rainfall becomes more extreme, flooding, landslides, and erosion would become greater and/or more frequent (Fowler and Hennessy, 1995; IPCC 1996, WG II, Sections 4.2, 12.4). Locations most at risk include steep or unstable hill country, river valleys, and floodplains. Most of the region's settlements are adjacent to rivers and harbors. Furthermore, in low-lying coastal areas, sea-level rise will compound the problem by reducing river flood discharge rates and thus exacerbate flood heights and durations.

Where urban areas are exposed to flood risk, either directly or through failure of upstream river management works, the potential human and infrastructural vulnerability is extreme and has significant financial implications. Major costs to dam owners and public agencies would arise from any upward revision of Probable Maximum Flood estimates. The costs to private property and public buildings and infrastructure of floods under the existing climate are already high; for example, the damage caused by the 1974 floods in Brisbane was reported to be A$660 million (Minnery and Smith, 1996). A study by these authors of the Hawkesbury-Nepean corridor, an urban area near Sydney, estimated that the potential damage from the "100-year" flood rose more than tenfold for a doubled CO₂ scenario in which the flood return periods decreased by a factor of four (Table 4-1).

Climate change is expected to affect water demand, water supply, and water quality (IPCC 1996, WG II, Sections 14.2, 14.3). For many parts of the region, rainfall and water supply are generally adequate. However, the drier inland areas of Australia are vulnerable to potential water shortages during the seasonal minimum and during droughts; any additional shortages in these regions arising from climate change would sharpen competition among various economic, social, and environmental uses and hence increase the effective cost of water. Considerable demand arises from urbanization and the diversion of large amounts of water for economic purposes such as mining and irrigated agriculture. This competition may be exacerbated by trends toward population growth, higher valuation of natural waters, and possibly shifts to more intensive farming systems. If, on the other hand, there were an increase in water availability due to climate change, it might well encourage demand for more irrigation, with obvious short-term benefits-although in the longer-term this could lead to increased salinization in semi-arid regions.

In central Australia, low rainfall and high evaporation forces the few towns-such as Alice Springs and Yulara-and other tourist centers, cattle stations (ranches), and Aboriginal and mining settlements to rely on fresh or desalinated brackish groundwater (Knott and McDonald, 1983; Jacobson et al., 1989; Jacobson, 1996). Economic growth and population growth will put added pressure on these supplies, which are recharged in part by occasional heavy rainfall events. Rainfall changes would have complex impacts on this groundwater supply because any reduced recharge would lower water tables and water supplies, and vice versa. The effects on water quality are not clear.

The water resources of atolls and low-lying islands are usually restricted to rainwater or limited sub-surface supplies. These are sensitive to climatic variations and environmental impacts, and in many cases are already stressed by unsustainable demand and pollution. With climate change, the freshwater lenses beneath coral atolls are exposed to saltwater intrusion from rising sea levels and possibly increased storm events, as well as to possible rainfall reductions where these occur. The indigenous populations of the Torres Strait, the Cocos (Keeling) Islands, and Pacific islands associated with New Zealand are likely to be affected.

The major Murray-Darling River system, in southeastern Australia, is heavily regulated by dams and weirs and supplies 10 billion m³ of water annually for human use, mainly for irrigation. This amounts to about 40% of the mean annual flow. Application of various scenarios (e.g., CSIRO, 1992) has suggested a possible combination of decreased mean rainfall, higher temperatures and evaporation, and a higher frequency of extreme events (i.e., more floods and droughts) (e.g., Schreider et al., 1996; see Figure 4-2); if so, the inflows to the system may be reduced, and if irrigation demands remain fixed or increase, the remaining river flows would substantially decrease (see, e.g., Hassall and Associates, 1997, and Box 4-1).

The Asian-Pacific Integrated Model (AIM) team has simulated changes in drought and flood intensities for a wide region, including Australia and New Zealand; AIM (1997) shows results using the Geophysical Fluid Dynamics Laboratory (GFDL) Q-flux climate model (experiment A2 or A3). Based on the estimated magnitude of 10-year return period monthly discharge events, they show small and spatially-varying changes in flood discharge over Australia but apparently significant reductions in minimum flows in the Murray-Darling Basin. However, the rainfall simulation of experiment A2 over Australia is questionable (Whetton et al., 1994).

Because of the higher variability of Australian rainfall, the storage capacities of Australian water supply dams are typically about six times larger than those of European dams for the same percentage of mean annual streamflow and probability of water shortfall. In sharp contrast, New Zealand's higher and more reliable rainfalls enable its hydro-electricity system to operate on a total reservoir capacity equal to only about six weeks of national demand-albeit with the attendant risk of a rapid decline in supply in times of drought, as was experienced during 1992 (Fitzharris, 1992).

IPCC Homepage