Above average temperatures more likely in WA

Daytime temperatures in the west of WA and overnight temperatures over most parts of the State, have an increased chance of being above average during the coming three months, the Bureau of Meteorology announced today.

The current pattern of outlook probabilities has been more heavily influenced by recent above average temperatures in the Indian Ocean. For the August to October period, the chances of above average seasonal daytime temperatures are between 60 and 65% over western WA (see map). So in years with ocean patterns like the current, about six August to October periods out of ten are expected to be warmer than average across these parts of the country, with about four out of ten being cooler. However, for the rest of the State, the shifts in the odds are too weak to offer any firm guidance as to the most likely outcome for mean maximum temperatures during the period from late winter to mid-spring. Outlook confidence is related to the consistency of the influence of Pacific and Indian Ocean temperatures on seasonal temperatures. During August to October, history shows this influence on maximum temperatures to be moderately consistent across most of WA, apart from the Kimberley (see background information). Overnight temperatures have an increased chance of being above normal for the season across much of WA. The chances of above average minimum temperatures are mainly between 60 and 70%, rising to near 75% across a large area in the west of the State. History shows the oceans' influence on minimum temperatures during August to October to be moderately consistent over the southern half of WA, but only weakly consistent in the north.

Click on the maps above for larger versions of the maps. Use the reload/refresh button to ensure the latest forecast maps are displayed.

THE NEXT ISSUE OF THE SEASONAL OUTLOOK IS EXPECTED BY 17^th AUGUST 2004.

Corresponding rainfall outlook

Maximum temperature departures from average for April to June 2004 - base period 1961-1990

Minimum temperature departures from average for April to June 2004 - base period 1961-1990

• The Bureau's seasonal outlooks are general statements about the probability or risk of wetter or drier than average weather over a three-month period. The outlooks are based on the statistics of chance (the odds) taken from Australian rainfall/temperatures and sea surface temperature records for the tropical Pacific and Indian Oceans. They are not, however, categorical predictions about future rainfall, and they are not about rainfall within individual months of the three-month outlook period. The temperature outlooks are for the average maximum and minimum temperatures for the entire three-month outlook period. Information about whether individual days or weeks may be unusually hot or cold, is unavailable.

• This outlook is a summary. More detail is available from the contact people or from SILO (www.bom.gov.au/silo/products/SClimate.shtml).

• Probability outlooks should not be used as if they were categorical forecasts. More on probabilities is contained in the booklet The Seasonal Climate Outlook - What it is and how to use it, available from the National Climate Centre. These outlooks should be used as a tool in risk management and decision making. The benefits accrue from long-term use, say over 10 years. At any given time, the probabilities may seem inaccurate, but taken over several years, the advantages of taking account of the risks should outweigh the disadvantages. For more information on the use of probabilities, farmers could contact their local departments of agriculture or primary industry.

• Model Consistency and Outlook Confidence: Strong consistency means that tests of the model on historical data show a high correlation between the most likely outlook category (above/below median) and the verifying observation (above/below median). In this situation relatively high confidence can be placed in the outlook probabilities. Low consistency means the historical relationship, and therefore outlook confidence, is weak. In the places and seasons where the outlooks are most skilful, the category of the eventual outcome (above or below median) is consistent with the category favoured in the outlook about 75% of the time. In the least skilful areas, the outlooks perform no better than random chance or guessing. The rainfall outlooks perform best in eastern and northern Australia between July and January, but are less useful in autumn and in the west of the continent. The skill at predicting seasonal maximum temperature peaks in early winter and drops off marginally during the second half of the year. The lowest point in skill occurs in early autumn. The skill at predicting seasonal minimum temperature peaks in late autumn and again in mid-spring. There are also two distinct periods when the skill is lowest - namely late summer and mid-winter. However, it must always be remembered that the outlooks are statements of chance or risk. For example, if you were told there was a 50:50 chance of a horse winning a race but it ran second, the original assessment of a 50:50 chance could still have been correct.

• The Southern Oscillation Index (SOI) is calculated using the barometric pressure difference between Tahiti and Darwin. The SOI is one indicator of the stage of El Niño or La Niña events in the tropical Pacific Ocean. It is best considered in conjunction with sea-surface temperatures, which form the basis of the outlooks. A moderate to strongly negative SOI (persistently below –10) is usually characteristic of El Niño, which is often associated with below average rainfall over eastern Australia, and a weaker than normal monsoon in the north. A moderate to strongly positive SOI (persistently above +10) is usually characteristic of La Niña, which is often associated with above average rainfall over parts of tropical and eastern Australia, and an earlier than normal start to the northern monsoon season. The Australian impacts of 23 El Niño events since 1900 are summarized on the Bureau's web site (www.bom.gov.au/climate/enso/).