The Science Delivery Division of the Department of Science, Information Technology, Innovation and the Arts (DSITIA) notes that there is an equal likelihood of above-median and below-median rainfall over the next three-month period (March to May), i.e. no strong signal for drier or wetter conditions. Read more (PDF, 168K, last updated 09:30AM, 12 March 2013)*
Seasonal forecasts are based on the current and projected state of the El Niño-Southern Oscillation (ENSO) phenomenon and on factors which alter the impact of ENSO on Queensland rainfall (e.g. the Pacific Decadal Oscillation (PDO)). This time of year is known as the ‘autumn predictability gap’ when ENSO-based forecasts are least reliable.
Currently, one tropical cyclone (‘Oswald’) has made land-fall in Queensland this season (see BoM Special Climate Statement (PDF)* and the Queensland 2013 Flood Recovery Plan (PDF)*). The possibility remains of more tropical cyclones making land-fall over Queensland during the remaining two months (March and April) of the cyclone season.
So far this summer (November to February), much of central and south-eastern Queensland has been wet, having received decile 7 or above rainfall. In contrast, extensive areas of inland Queensland, including northern regions of the state, have recorded rainfall below decile 3 for the summer so far.
There are various approaches used to provide rainfall outlooks. These approaches tend to differ in terms of the components of the climate system that are considered and, for this reason, each approach may convey a different outlook, particularly for specific locations.
DSITIA produces two statistical climate risk assessment schemes. They are:
The experimental SPOTA-1 scheme provides long-lead probabilities of summer (November to March) rainfall for Queensland from mid-April through to mid-November each year. The final assessment of rainfall probabilities for the 2012/13 summer, based on the experimental SPOTA-1 scheme as at 1 November 2012, indicated that decile 3 to 7 rainfall is most likely for much of Queensland this summer (November to March). This assessment is based, in part, on an index of March sea-surface temperature anomalies which reflected a ‘cool’ state of the PDO. This assessment also takes into account a monthly ENSO index, based on the sea-surface temperature gradient (west to east) across the South Pacific Convergence Zone (i.e. between eastern Australia and the central Pacific).
DSITIA’s SOI Phase scheme provides probabilities of rainfall for the coming three-month season based on SOI values over the previous two months. The SOI Phase scheme currently indicates that the probability of above-median rainfall across most of Queensland is 40 to 50 per cent for the next three-month period (March to May). This analysis is based on the SOI being in a ‘Consistently Near-Zero’ phase at the end of February, as discussed further in the Australia's Variable Rainfall poster (PDF, 9.3M, last updated 03:09PM, 29 July 2016)*, or the Department's archive of historical rainfall maps).
Each of the above schemes may have their own particular following. Although such schemes cannot provide outlooks with absolute certainty, users of the information who follow a skilful scheme should benefit from doing so in the long-term. Users should consider the historical track record of any scheme, and such historical information is becoming increasingly available.
DSITIA's Long Paddock website provides the historical archive of SPOTA-1 reports and past commentaries on the SOI Phase scheme. Users should also consider the wide range of information available each month describing the current state of the ocean/climate system, for example the 'ENSO Wrap-Up' (PDF)*.
ENSO influences other climate variables apart from rainfall (e.g. temperature, pan evaporation and vapour pressure). This means that the impact of ENSO on crop or pasture growth can be stronger than on rainfall alone. The impact of ENSO on pasture growth is also dependent upon current pasture condition and soil water status. DSITIA’s AussieGRASS model takes these factors into account in producing seasonal pasture growth probabilities.
* Requires Adobe Reader