Climate Forecasting Group

Group Leader: N. Nicholls

Objectives: To improve the understanding of Australian climate, including the effects of climate change, and to develop systems to predict climate on scales beyond the medium range.

Research activities

Studies to improve the understanding and application of intraseasonal variability, as originating from the tropics, have continued. A journal paper on the newly developed Madden-Julian Oscillation (MJO) real-time multivariate (RMM) index was submitted and accepted. This paper included some results of an analysis of Australian rainfall in association with the MJO. An external web-page providing daily-updated experimental forecasts based on multiple linear regression with the index was established. The forecasts were also provided to an international program on the comparison of both numerical and statistical long-lead forecasts. For the period over which actual forecasts have been made (November 2003 to February 2004), the RMM-based forecasts were the most skillful (at an 11-day lead). Two review chapters were written for a comprehensive new book on intraseasonal variability of the atmosphere-ocean climate system. A project with the aim of making forecasts of the weeks 1, 2, and 3 probability of tropical cyclone (TC) formation and occurrence has also commenced. Preliminary results show a more than 3:1 modulation of TC probabilities given the MJO phase.

The relationship between the MJO and the El Nino Southern Oscillation (ENSO) was investigated. MJO activity is unrelated to the state of ENSO during the austral summer (November through March), which is the season that the MJO has largest amplitude. However, in boreal summer, MJO activity is significantly correlated with El Nino, with enhanced MJO activity tending to occur in conjunction with onset of El Niño. Despite the lower level of MJO activity in boreal summer, the coupled system is sensitive to MJO activity at this time of year. Four to twelve months prior to the peak of El Nino, MJO-stress variance increases in the far western Pacific. The associated intraseasonal thermocline variability in the eastern Pacific peaks 2-4 months before El Nino. These results confirm that the MJO acts to decrease the predictability of ENSO.

Dynamics of coupled variability in the Indian Ocean were explored with the Predictive Ocean-Atmosphere Model for Australia (POAMA) and the CSIRO Mk3 models. The work suggests that the Indian Ocean is an active player in the ENSO, and that a better representation of the Indian Ocean is required to improve simulation and prediction of ENSO. As well, it has been shown that a statistical forecast scheme using sub-surface temperature data can produce improved forecasts of El Niño.

The third stage of the development of the DIAGNOSE software to assist in the rapid analysis of Australian climate anomalies was completed. The software now includes the facility to correlate time series of climate anomalies with 1000 and 500 hPa heights from NCEP/NCAR Reanalyses, to provide a rapid examination of possible causes of the anomalies. Copies of the software are being distributed to interested researchers, especially those involved in climate impact studies.

The second stage of the Indian Ocean Climate Initiative(IOCI) of the Western Australian Government commenced in July 2003. The major BMRC commitment to IOCI over the next three years will be to improve understanding of the causes of the decline in rainfall in the south-west of Western Australia. This is being undertaken through an analysis of how synoptic systems affecting the south-west may be changing. This will involve work using reanalyses and climate models. Meanwhile, the use of downscaling to examine the drying trend has continued. Both mean sea level pressure and vertically integrated precipitable water are important for explaining the observed local rainfall trend in the second half of the 20th century. The projections made by downscaling several climate change experiments performed with a coupled GCM suggest a future worsening of this drying trend with a reduction of rainfall in both winter and spring.

A number of studies of regional climate changes in Australia have been undertaken (in addition to the IOCI project). A comprehensive study of the possible climate effects of increasing surface water availability in inland Australia (eg., the “Bradfield Scheme”) has been completed. No evidence was found to support the concept that this would lead to a substantial amelioration of the climate. Mean maximum temperatures in parts of New South Wales declined over the 20^th century, and this has been cited as evidence that the warming since the mid 20^th century was not attributable to the enhanced greenhouse effect. BMRC research has now demonstrated that the 20^th century New South Wales temperatures declined sharply mid-century, associated with a sharp increase in rainfall. Subsequently, temperatures have increased and this increase is not associated with changes in rainfall, suggesting that a new mechanism is operating. It has also been shown that Australian spring snow depths have declined about 40 per cent over the last 40 years, and that this decline is attributable to warming (and not explainable by changes in rainfall). Work has commenced to explain the evaporation “paradox”, i.e. why pan evaporation has been decreasing even though temperatures have been increasing. A study of decadal scale variability of the latitude of the subtropical ridge over eastern Australia, and its impact on Australian rainfall has been completed.

The Group also undertakes collaborative projects on climate impacts. BMRC participated in a project funded by the National Health and Medical Research Council (NHMRC) and led by Queensland University of Technology into the climate impacts on Ross River virus. A statistical system to forecast the prevalence of Ross River virus in Brisbane has been developed. A collaborative project commenced with Macquarie University and Birds Australia to document avian datasets with the potential for use in climate change impact studies. This involved the preparation of a paper describing known and projected climate change impacts on Australian birds. Studies continued on climate-avian interactions in southern Australia, with two papers being produced on climate and Little Penguin breeding. Forecasts of the expected start of the Little Penguin breeding season, using SST data, were used for the first time by the Phillip Island Nature Park.

The Group has a continuing interest in climate extremes. The Fifth and final Asia Pacific Network (APN) workshop on trends in climate extremes was held 23-25 March in the Bureau, hosted by BMRC. Participants from 15 countries discussed the availability of climate data suitable for the monitoring and analysis of climate and weather extremes, and the challenges and opportunities for improved monitoring of changes in climate extremes. The participants examined how changes in extremes were related to changes in means of climate variables, and investigated how variations in extremes were related to large-scale climate variables such as sea surface temperature. A paper summarising the main findings is being prepared.