For more than a quarter of a century, Australian scientists have been active in developing numerical weather prediction models for operational usage. One of the earliest modellers was R.H. (Reg) Clarke, who in 1967 produced the first numerical weather prediction for the southern hemisphere using the primitive equations of motion. In this paper we trace the development of limited area numerical weather prediction over the Australian region from the early days to the present. Our primary emphasis will be on the techniques used to solve the primitive equations; however, we also examine other important aspects of numerical weather prediction. These include the choice of lateral boundary conditions, the problem of making the best use of the unique mix of conventional and remotely-sensed data available over the Australian region, and the inherent predictability limitations of the atmosphere. Finally, we discuss the directions limited area numerical weather prediction might take in Australia during the 1990s.
LESLIE et al. (Sun,) studied this question.