|
||||||||||||||
|
|
||||||||||||||
|
MAY 1996 NEWSLETTERSIMULATING THE BURDEKIN RIVER IN FLOODA CRC study to better understand and monitor long term impacts of freshwater on coral reefs is powering into a new phase, thanks to an AIMS 'supercomputer'. A new modelling program has been developed to help researchers analyse wet season flood plumes (sediment-laden fresh water at river mouths) from the Burdekin River that may reach offshore reefs between Townsville and Bowen. The next phase of work involves extending a verified three-dimensional hydrodynamic computer model showing the plume dynamics of the Burdekin River in flood, which can reach as far as 300 kms along the coast to Cairns. The present model, used for testing model formulation, (Figure 1) has verified the 1981 flood, as measured and reported by Wolanski and Van Senden (1983). This figure shows a simulation of the flood using actual discharge and wind data from this period. Wolanski and Van Senden's data, collected over two days, shows how this 'snap-shot' model can accurately correlate salinity distribution observed and predicted on January 27th, 1981. Comparisons at other times also confirms that the model is consistently reproducing field-tested observations. The far field differences observed between the plume simulation and the field measurements are probably due to other rivers, such as the Haughton and Ross rivers, discharging at the same time. The peak discharge from the 1981 flood was more than 12,000 tonnes of water per second, or a staggering of 7.5 billion tonnes of freshwater carried into the GBR lagoon over a three week period. The computer model simulation of this event was for 21 days, taking 50 hours of CPU time on the very powerful IBM590, and involving 56 trillion calculations. Initial analysis on the model shows the main influence on the river's plume at sea is the discharge volume and local winds. So each year, different plume trajectories are expected depending on changes in wind and rainfall. The computer model is currently being expanded to cover the entire continental shelf between Cairns and Bowen. This new model domain will have about 100,000 computational points (5 points in the vertical and 20,000 in the horizontal at a resolution of 2 km). We have prepared the wind data and discharge data from the Burdekin region from 1966 to 1991 to simulate flooding over this period of 25 years. These runs will give us the data for the risk assessment of the Burdekin plume waters impacting reefs of the Central Great Barrier Reef. An aspect of the model simulations is that it identifies the fate of the Burdekin River plume in isolation to other river discharges which cannot be measured in the field. This should provide useful information on catchment management implications for the Burdekin region and its impact on shelf waters. Proposed risk assessment will be over this 25 year period, but it may be interesting to also divide this 25 year simulation and examine probabilities before and after construction of the Burdekin dam. This would enable resource managers to examine if changes in the river's flow caused by the dam has affected the plume over recent years. It might also be possible to correlate prawn catch data from the region with the fate of the freshwater to see if the plume has an impact of the spatial variability of the prawn catch. The project, with help from Eric Wolanski, Terry Done, Felicity McAllister, Duncan Galloway, and Kris Summerhayes is funded by the CRC Reef Research Centre Ltd, IBM and AIMS. By Brian King and Simon Spagnol, AIMS Reference: |
