Regional-Scale Seasonal Prediction Over Eastern Australia and the Coral Sea

Significant Achievements, Anticipated Outcomes and Future Work

The APAC facility has been used to for work in 3 research areas – seasonal prediction, extreme rainfall and climate change, and the development of a relocatable, coupled high resolution atmosphere-ocean model for the Royal Australian Navy . The tool used in each of these areas is a numerical model of the atmosphere known as RAMS. RAMS is suitable for use at high spatial resolution. The model is nested in the output from either analyses of the global atmosphere or global climate model simulation of the atmosphere.

Seasonal Prediction: Previous work in this area has shown that RAMS may be used to downscale the output from global analyses and possibly from the current operational CSIRO multi-seasonal prediction system. During the 2003, the APAC system was used to complete a series of hindcasts designed to quantify the skill of RAMS in providing these predictions up to 9 months ahead. To achieve this, RAMS was nested in NCEP analyses and two 10-year series of 12-month simulations were carried out. The output from these simulations is currently being analysed. Early results indicate that the model has some skill in differentiating between wet and dry years and in predicting the spatial distribution of rainfall 9 months ahead.

Climate Change: RAMS has also been used to quantify the expected increase in the intensity of extreme rainfall events due to the enhanced greenhouse effect. Output from the CSIRO Mark 3 GCM has been used to select 100 current-climate extreme rainfall events and 100 future-climate (corresponding to 2040) extreme rainfall events as input to RAMS. South- east Queensland and northern NSW were chosen for the study as this region has the highest flood-risk in Australia. These simulations highlight the importance of mesoscale processes and terrain effects in extreme rainfall events. They have shown that high-resolution, mesoscale models of the atmosphere are able to represent, both quantitatively and spatially, the climate of extreme rainfall events. Key findings from this work are that the most extreme rainfall events are likely to increase in intensity and frequency as a result of the enhanced greenhouse effect. These increases in intensity are approximately 30% for the climate of 2040. Increases in intensity are predicted to occur for areas less than 6000 km2 but for larger areas the change in the intensity of extreme rainfall events is negligible. The results also show that the spatial distribution of the change in intensity of extreme rainfall events is not uniform. Extreme events increase in intensity over the mountainous terrain but tend to decrease elsewhere. The largest increases in extreme rainfall intensity occur in the mountainous regions that currently experience the most extreme rainfall events. If this result is correct, it has major implications for the planning of major infrastructure projects such as large dams, and for flood planning for the communities downstream from these catchment areas.

Relocatable Ocean-Atmosphere Model: RAMS is being evaluated as part of the BlueLink Ocean Forecasting Initiative of the Royal Australian Navy, CSIRO and the Bureau of Meteorology. RAMS will form the atmospheric component of the Relocatable Ocean Atmosphere Model (ROAM) that is being developed as an "in-house" model for Navy operations. As part of this work, RAMS has been run in a semi-operational manner, since September, in a variety of locations with a grid spacing of 2 km. The locations considered include southwest WA, the Northwest Shelf, Brisbane and southeast Qld, the Solomon Islands and Auckland. These simulations have provided guidance in setting up the model to be robust and accurate in a wide variety of geographical and meteorological conditions. As a result of this work, some aspects of the model’s physics have been upgraded. Severe weather such as severe thunderstorms and heavy rain are the most difficult fields for a meteorological model to simulate. The experimental, semi-operational runs indicated that RAMS was able to capture the intensity of the heavy rain and thunderstorms that affected Brisbane in January. A stable version of RAMS is now available and hindcasts for the period October 2003 – March 2004 will soon be commenced so that the model skill can be evaluated statistically.

Computational Techniques Used

The numerical model used for this research is the Regional Atmospheric Modelling System(RAMS). RAMS is a high- resolution, compressible, non-hydrostatic model of the atmosphere. The physical processes represented by the model include an atmospheric boundary layer, soil and vegetation effects, long and short wave radiation, and the complex cloud processes that result in precipitation (ice, liquid and water vapour). RAMS has a two-way nesting capability that allows the user to model the region of interest at higher resolution than the surreounding region.

Publications, Awards and External Funding

External Funding and Awards

Not applicable.

Publications

D. Abbs, 2003: How climate change threatens that dream life on the coast. Science and Technology, Canberra Times. 4 September 2003.

Abbs, D., 2004: The Effect of Climate Change on the Intensity of Extreme Rainfall Events. In: Proceedings of Coast to Coast 2004. Hobart.

Abbs, D., 2004: A High-Resolution Modelling Study of the Effect of Climate Change on the Intensity of Extreme Rainfall Events. In: Proceedings of the 44th Annual Conference of the Floodplain Management Authorities of NSW. Coffs Harbour.