Grains Research and Development

Date: 01.06.1997

Weather forecast: Cloudbanks for the south... by Denys Slee

Last November Ian Holton forecast there would be widespread rains in southern Australia in late April to early May this year. He was right. "About six years ago I could see that the El Niño-Southern Oscillation Index didn't really work for forecasting in southern Australia, especially for the critical April to October crop-growing season," this long-range forecaster with the Bureau of Meteorology in Adelaide said. "It was too unreliable."

So he turned to the development of a forecasting system based on one of the most common rain mechanisms for southern Australia — the presence of north-west cloudbanks ahead of cold fronts.

He used a link between sea-surface temperatures in the north-eastern Indian Ocean and the generation of these cloudbanks, along with the presence and strength of major long-wave low-pressure troughs in the middle and upper atmosphere over, or just west, of the Australian continent.

If the clouds are present and are taken by the troughs across southern Australia, ahead of a cold front, then the chances of rain falling increase.

Mr Holton has developed his forecasting model on 27 years of data including sea and land surface pressures and upper level long-wave patterns. This has been validated against 13 years of independent rainfall data.

Spring signals

He says signals about development of the cloudbanks and the troughs become apparent in spring of the year before they actually occur. This enabled him to predict six months ago that good general rains would fall in southern Australia in late autumn.

"November is the time when we get a very good guide to what will happen six months later. There are no other real signals until April but even a revised forecast then shouldn't vary much from the November one.

"For example it might change from above-average to average for May to November but not from above-average to below-average."

"No forecasting system can be 100 per cent accurate," he said. "But this model should be consistently reliable for southern Australia. It has great potential for helping farmers make their important cropping decisions."

What's it looking like?

So what is in store for southern Australian farmers from May through to spring?

According to Ian Holton's experimental forecast, 1997 is a lot like 1965 and, to a lesser extent, 1972 (both El Niño years — Ed). Sea-surface temperatures in the north-east Indian Ocean are above normal and favourable for cloudbank activity and there are above-average sea-surface temperatures in the Great Australian Bight favourable to increased post-frontal moisture and stream shower activity.

Upper long-wave patterns are conducive for transporting moisture north-west to south-east during winter-early spring. The developing El Nino pattern may cause drier conditions in SA and south-west Victoria during October-November.

Which means:

  • SA agricultural districts are forecast to have average winter rainfall but it will dry off in mid- to late spring.
  • Average winter falls are forecast in Victoria for most of the season but to be below average south of the Divide in spring.
  • Southern NSW farmers (west of the Divide) are forecast to have below average rains in winter increasing to average in spring.

And here are some selected Holton wheat yield forecasts as at May 1997:

  • Wagga Wagga, 2.3 t/ha,
  • County Le Hunte (Northern-Central Eyre Peninsula), 0.8 t/ha, and
  • Lameroo district (SA southern Mallee) 1.9 t/ha.