Rural Industries Research & Development Corporation

Research Compendium 1993 - 1994

Research Project

SIMULATION AND PREDICTION OF LARGE SCALE RAINFALL VARIABILITY

Objectives

• To simulate, with a global climatic model, years with observed enhanced rainfall over Australia, and to reproduce with the necessary accuracy the spatial, temporal and quantitative rainfall variations observed.
• To identify positively the precursor mechanisms, particularly the role of local sea surface temperature (SST) anomalies around the coast of Australia.
• To conduct predictive experiments involving either hindcasts or actual predictions of enhanced rainfall situations.

Background

Numerous observational studies have established that interannual rainfall anomalies over many parts of the globe can be related to sea surface temperature (SST) anomalies in the equatorial Pacific Ocean. This suggests that, if such anomalies can be predicted seasons in advance, then prediction of rainfall outcomes for the same timescales could be made.

In addition, studies have also indicated there may be a relationship between SSTs to the north and the west of Australia and rainfall during winter. This relationship may be a source of seasonal predictive skill beyond that associated with the Pacific SSTs. Climate models are now being used to simulate these types of relationships and are also being tested as a means of predicting rainfall anomalies given SST predictions.

Research

Relationships between Australian rainfall anomalies and SST variations have been investigated by a series of model experiments involving:

• a coarse resolution model and observed SSTs for the period 1950-1988;
• a moderate resolution model and observed SSTs for the period 1979-1988;
• a high resolution model using observed SSTs for 1993; and
• a very fine resolution, limited area model using the observed SSTs for July 1978.

Real-time rainfall predictions for 1991 and 1992 were conducted using predicted SSTs from a research group in the USA.

Outcome

The coarse and medium resolution models are capable of reproducing some of the large scale SST-related climate but less so at the regional scale. An improved high resolution model provides a better representation of seasonal mean rainfall over Australia and also appears to better simulate observed anomalies during 1993. July 1978 was a time when the relationship between rainfall and SSTs to the north and west of Australia was apparently very strong.

The results from both the coarse and very fine resolution models did not reveal any significant impact of mid-latitude SST anomalies in the Indian Ocean on simulated rainfall. In 1991 and 1992 extensive droughts occurred over eastern Australia and southern Africa. These outcomes were broadly predicted correctly, particularly the situation over Australia in 1992 when contrasting dry/wet outcomes late in the year were obtained.

Implications

Current models adequately simulate the large scale relationships between El Nino events and rainfall in the tropics. There is little evidence of a strong relationship between Indian Ocean SSTs and simulated winter rainfall. The development of higher resolution models appears to be necessary if more accurate dynamic seasonal predictions are attempted in the future. Also needed will be accurate ocean models which can lead to more comprehensive and accurate prediction schemes.

RIRDC Project No: CSG-2A

RESEARCHER: Dr I N Smith

ORGANISATION: CSIRO Division of Atmospheric Research

PHONE: 03 586 7539

PUBLICATIONS:

Hunt, B G, Zebiak, S E and Cane, M A, 1994, Experimental predictions of climatic variability for lead times of twelve months. Int.Jour.Climatol., 14, pp507- 526.

Yonetani, T and Smith, I N (in preparation) Model simulations of seasonal climate anomalies during 1993. To be submitted to Journal of the Meteorological Society of Japan.

Smith, I N, 1995, A GCM simulation of global climate interannual variability 1950-88, J.Climate.

Last updated: 10 October 1996
Copyright © RIRDC
Index of RIRDC Projects URL
http://www.rirdc.gov.au/pub/compendium/93-94/index-b.html