However for our agricultural sector and our regional economies, accurate climate forecasts for the next year or even decade could revolutionise planning and resource management.
Recognising the potential of what is known as “decadal forecasting”, the CSIRO has dedicated a team to a ten year, $36 million project aiming to predict significant variations in the future climate and forecast periods of both good conditions and bad.
CSIRO, Oceans and Atmosphere Climate Science Centre, chief research scientist, Dr Richard Matear said the project looked at longer term predictions that are characterised by leading “climate states” such as ENSO.
“Predicting whether an elNino or LaNina climate state is going to occur is a challenge,” he said.
“Once the state has flipped and we are going into an elNino, our systems are very good at forecasting the next six to nine months, but at this time, July or August, it is really hard to forecast.”
Dr Matear said the group was using ocean observations, including temperature at depth, to better drive the accuracy of forecasting upwards of a full year out.
“We think the information in the ocean will be super important to provide the information on the climate state and get us over the barrier we have now, beyond seasonal forecasting,” he said.
“The ocean is really the memory of the climate system, so understanding and observing the ocean, and understanding the process by which the ocean influences the atmosphere we hope will provide forecasting in a longer time scale.”
Dr Matear said ocean data was collected from a variety of sources including the Argo array, satellite and sea surface temperature monitoring.
“We probably have 4000 Argos floating around monitoring the ocean,” he said.
“it has been a revolution in understanding the oceans structure and how it is evolving in time.”
Dr Matear said a fundamental step in the project was comparing the predictions forecasting models made against actual observations of weather to check the accuracy, this required large amounts of data storage and computing power.
“Because we need to run lots of model simulations, it is a challenge,” he said.
“By having many predictions and models we can run the statistics of extreme events, what is the probability of it being really dry, or really wet.”
Dr Matear said the team had already developed an initial forecasting system they were about to start testing against prior observations.
While the project concentrated on decadal forecasting, Dr Matear said there was some overlap with Bureau of Meteorology efforts to improve seasonal forecasting.
“CSIRO have the opportunity to probe the system and learn how to make better forecasts and then have that information feed back into the Bureau’s forecasting system and improve it into the future.”