Few natural events raise a greater sense of fear across the Australian community than drought. Australia is the driest inhabited continent and its pattern of rainfall is highly variable from year to year and over decades. Drought is best thought of as a period when accumulated rainfall is significantly less than that anticipated. It is of particular concern because of its impact on agricultural production and rural communities. Even major cities are affected as water supplies diminish during periods of extended drought.
CSIRO scientists say that later this century there could be three times as many droughts every decade in Australia because of climate change. Using computer models based on different levels of carbon dioxide emissions, they claim to predict temperature, rainfall and drought in the year 2070.
The CSIRO prediction is particularly alarming and should receive close scrutiny. A more sober assessment of past rainfall patterns and the way the complex climate system actually works do not support the claims.
The Bureau of Meteorology publication, "Drought, Dust and Deluge: A Century of Climate Extremes in Australia", identifies 6 major drought events during the 20th century and gives the historical context. Two of these events extended over about 12 months but the others were multi-year droughts.
The 1914-15 and 1982-83 droughts were linked to major El Nino events, the abnormal warming of surface waters of the central and eastern Equatorial Pacific Ocean. El Nino events have no known association with global temperature trends. They come about through two contributing factors: a reduction in upwelling of very cold sub-surface water, and eastward flow of warmer water from the western Pacific Ocean.
Warmer surface temperatures over the central and eastern Pacific Ocean affect the atmosphere in two ways. Firstly, they caused rain-producing clouds to move eastward from Indonesia and New Guinea to the Pacific Ocean; this reduces the tropical moisture supply to the weather systems over Australia. Secondly, the normal subsidence of air over Australia, which contributes to drying of the atmosphere and is the cause of the generally low rainfall, is increased.
The overall effect of any El Nino event is to reduce rainfall over Australia although not all El Nino events result in drought. The pattern of rainfall deficiency is not consistent with each event because rain occurs in association with individual weather systems. Some regions can receive adequate rainfalls even when there are more general deficiencies elsewhere. The major El Nino event of 1997-98 had global impact but timely rains mitigated the effect on Australia.
Multi-year drought events were experienced from 1885-1902 (the Federation Drought), 1937-45, 1965-68 and 1991-95. Each of these was linked to El Nino-like sea surface temperature patterns across the equatorial Pacific Ocean. Above average temperatures dominated the central Pacific and at times extended to the eastern Pacific. These multi-year events are sometimes called "long El Nino".
Varying sea surface temperature patterns over the tropical Indian Ocean also affect Australian rainfall, particularly the northwest of the continent. Overall, the influence of tropical sea surface temperature patterns on Australian rainfall is second only to the seasonal cycle.
Ocean temperature records have identified an apparent "climate shift" to warmer surface waters across the tropical Pacific Ocean during 1976. The cause of the general warming has been linked to a decrease in overturning of the ocean surface layer and reduced upwelling of colder sub-surface water across the tropical Pacific and Indian Oceans.
The reduction of Australian rainfall over the last three decades is an outcome of the generally warmer sea surface temperature of the Pacific and Indian Oceans and the characteristic El Nino-like conditions. It cannot be linked to increasing carbon dioxide concentrations.
Our understanding of rainfall patterns over Australia would lead to the conclusion that any future changes in rainfall, such as more droughts and more floods, must result from increased variability of the sea surface temperature patterns of the tropical oceans. In particular, if the CSIRO computer-based predictions are to have any veracity, we would expect the computer models to respond to anthropogenic greenhouse forcing by projecting more frequent and more intense El Nino events.
For the major 1997-98 El Nino event, computer models underestimated the speed and eventual strength of the tropical warming. Climate forecasters cannot yet be confident in the skill of computer model projections a few months in advance. The reality is that computer models do not simulate El Nino events with skill. The UN's Intergovernmental Panel on Climate Change (IPCC), in its 2001 Third Assessment Report, makes this point.
Computer simulations extending over many years reproduce some of the characteristics of El Nino-like events but the amplitude is significantly less than observed. Also, computer models have not demonstrated an ability to reproduce the ocean dynamics associated with observed El Nino events.
In summary, the computer models are unable to simulate the most basic regulator of Australian rainfall variability, either under existing conditions or with increasing carbon dioxide concentration.
The reality is that Australian rainfall responds to the dynamics of the ocean circulations and changing tropical sea surface temperatures. The ocean circulations are a poorly observed and understood component of the climate system. We can make no predictions about future Australian rainfall until we better understand the oceans and represent their motions in computer models.
The CSIRO predictions of more droughts, based as they are on rudimentary computer models, lack credibility.
No comments:
Post a Comment