Climate Change 2001:
Working Group II: Impacts, Adaptation and Vulnerability
Other reports in this collection

12.8.3. Natural Systems

Figure 12-5: Risk response surface incorporating cumulative probability plots (in shaded box) for climate change magnitudes as indicated on x- and y-axes. Indicated percentage probabilities are probabilities of climate change in northern Victoria in 2070 lying within each shaded area (thus, there is a 100% probability of climate within the shaded square, and a 50% probability of climate within the innermost region). Probability (in percent) of irrigation water demand exceeding farm supply cap in any one year, for indicated climate change, is indicated by oblique lines. Critical threshold (heavy line) is set at a 50% chance of exceeding the cap (Jones, 2000).

Figure 12-6: Change in probability of exceeding critical threshold (exceeding farm cap in 50% of all years) over time. Note that if farm cap were reduced through reductions in irrigation supply under climate change, the probability of the cap being exceeded would be increased (Jones, 2000).

A large fraction of the region is composed of unmodified or nonintensively managed ecosystems where adaptation will depend mostly on natural processes. Vulnerability will occur when the magnitude or rate of climate variations lies outside the range of past variations. In some cases, adaptation processes may be very accommodating, whereas in others adaptation may be very limited. Ecosystems in the region handle a wide variety of climatic variability, in some cases with very large swings, but generally this variation occurs on short time scales—up to a few years. This does not necessarily confer adaptability to long-term changes of similar magnitude.

An important vulnerability identified by Basher et al. (1998) is the problem of temperatures in low to mid-latitudes that reach levels never before experienced and exceed the available tolerances of plants and animals, with no options for migration. The southwest of western Australia is a case in point (see Section 12.4.2). Another potential vulnerability arises from changes in the frequency of events. Examples include a climatic swing of duration exceeding a reproductive requirement (e.g., water birds in ephemeral lakes—Hassall and Associates et al., 1998) and damaging events occurring too frequently to allow young organisms to mature and ecosystems to become reestablished.

Vulnerability also is expected to exist where species or ecosystems already are stressed or marginal, such as with threatened species; remnant vegetation; significantly modified systems; ecosystems already invaded by exotic organisms; and areas where physical characteristics set constraints, such as atolls, low-lying islands, and mountain tops. Coral reefs, for instance, may be able to survive short periods of rising sea level in clear water but are less likely to do so in turbid or polluted water or if their growth rates are reduced by acidification of the ocean (Pittock, 1999). Vulnerability of coastal freshwater wetlands in northern Australia to salinization resulting from increasing sea level and the inability of some of these wetlands to migrate upstream because of physical barriers in the landscape is described in Sections 12.4.5 and 12.4.7.

Unfortunately, there is relatively little specific information about the long-term capacity for and rates of adaptation of ecosystems in Australia and New Zealand that can be used to predict likely outcomes for the region. Therefore, a large degree of uncertainty inevitably exists about the future of the region's natural ecosystems under climate change.

12.8.4. Managed Systems

In the region's agriculture, many farming systems respond rapidly to external changes in markets and technology, through changes in cultivars, crops, or farm systems. Mid-latitude regions with adequate water supplies have many options available for adaptation to climate change, in terms of crop types and animal production systems drawn from other climatic zones. However, at low latitudes, where temperatures increasingly will lie outside past bounds, there will be no pool of new plant or animal options to draw from, and the productivity of available systems is likely to decline.

Adaptation options will be more limited where a climatic element is marginal, such as low rainfall, or where physical circumstances dictate, such as restricted soil types. Even where adaptations are possible, they may be feasible only in response to short-term or small variations. At some point, a need may arise for major and costly reconfiguration, such as a shift from or to irrigation or in farming activity. Indeed, one adaptation process already being implemented in Australia to cope with existing competition for water is water pricing and trading, which is likely to lead to considerable restructuring of rural industries (see Section 12.4.6).

Management of climate variability in Australia currently involves government subsidies in the form of drought or flood relief when a specific level of extreme that is classified as exceptional occurs (Stafford Smith and McKeon, 1998; see also Section 12.5.6). An adaptive measure being applied in Australia and New Zealand is to improve seasonal forecasting and to help farmers optimize their management strategies (Stone and McKeon, 1992; Stone et al., 1996a; White, 2000), including reducing farm inputs in potentially poor years.

However, if a trend toward more frequent extremes were to occur, such measures might not allow farmers to make viable long-term incomes because there may be fewer good years. The question arises whether this is merely a string of coincidental extremes, for which assistance is appropriate, or whether it is part of an ongoing trend resulting from climate change. The alternative policy response to the latter possibility may well be to contribute to restructuring of the industry.

The PMSEIC (1999) report notes that there are opportunities for new sustainable production systems that simultaneously contribute to mitigation objectives through retention of vegetation and introduction of deeper rooted perennial pasture. Tree farming in the context of a carbon-trading scheme would provide additional opportunities, and this strategy could be linked to sustainability and alleviation of dryland salinity.

Nevertheless, the report recognizes that totally new production systems may be required for sustainability. These systems will need to capture water and nutrients that otherwise would pass the root zone and cause degradation problems. The design of such systems will entail research into rotating and mixing configurations of plants; manipulating phenology; modifying current crops and pastures through plant breeding, including molecular genetics; and possibly commercializing wildlife species and endemic biological resources. The report concedes, however, that there probably still will be agricultural areas where attempts to restore environmental and economic health will meet with little success.

Vulnerability and the potential for adaptation can be investigated quantitatively if a system and the climate change impacts on that system can be modeled. This was done by Pittock and Jones (2000) and Jones (2000) for climatically induced changes in irrigation water demand on a farm in northern Victoria. Here a window of opportunity is opened for adaptation via identification of a future time when adaptation would be necessary to reduce the risk of demand exceeding irrigation supply.

Jones (2000) used a model that was based on historical irrigation practice. Seasonal water use was used to estimate an annual farm cap of 12 Ml ha-1, based on the annual allocated water right. Water demand in excess of this farm cap in 50% of the years was taken to represent a critical threshold beyond which the farmer cannot adapt. Conditional probabilities within projected ranges of regional rainfall and temperature change were utilized, combined with a sensitivity analysis, to construct risk response surfaces (see Figure 12-5). Monte Carlo sampling was used to calculate the probability of the annual farm cap being exceeded across ranges of temperature and rainfall change projected at intervals from 2000 to 2100. Some degree of adaptation was indicated as desirable by 2030, although the theoretical critical threshold was not approached until 2050, becoming probable by 2090 (see Figure 12-6).

In a full analysis, this example would need to be combined with an analysis of the likelihood of changes in water supply (e.g., Schreider et al., 1997) affecting the allocated irrigation cap and an evaluation of possible adaptation measures.

Other reports in this collection