As the campaign to drive the invaders out of the islands to the north of Australia intensified, it became increasingly obvious that the methods of analysis applied to the weather of the higher latitudes were not applicable to those of the tropic. Following the transfer of Allied Headquarters to Brisbane in 1942, the Bureau sent R. A. E. Holmes and W. J. Gibbs (later to become Director of Meteorology) there as well, to form the Meteorological Section of Allied Air Headquarters providing advice on weather conditions over the area of operations (Gibbs in a personal communication).

With the increasing reliance of the military on weather forecasts for operational purposes, its staff decided to mount a major attack on the question of forecasting in the tropics using the additional skill and advice of those personnel serving in the forward areas. Relying on observations from their own as well as enemy sources (obtained by radio interception), they were able to delineate the various air-masses which affected the region and define the boundaries between them. By discarding the method of isobaric analysis, used in the southern parts of the country, in favour of a system of drawing streamlines, the staff found that these boundaries were zones where the air currents converged and that it was this that created the clouds and associated poor flying conditions (Mellor [58]). This information was distributed to other forecasting offices by means of the Tropical Weather Research Bulletin (TWRB) beginning in May 1944.

In the meantime, day-to-day forecasting experience showed that diurnal variations in cloud and rainfall patterns and orography also played a major role in the development of tropical weather systems (Gibbs [31]). This knowledge, when combined with the skills in streamline analysis gained from the TWRB, permitted a greatly improved system of tropical forecasting to be introduced throughout the South-West Pacific. Indeed, the TWRB was hailed by the British Air Ministry as one of the "two most outstanding contributions to tropical meteorology up to this time" (Mellor [58]).

As part of their ongoing investigation into the properties of radar, the CSIR Radiophysics Laboratory (RPL) requested Bureau assistance with research into the phenomenon known as anomalous propagation or super-refraction. This was an atmospheric condition in which radar waves were transmitted over very long distances by particular layers within the lower levels of the atmosphere. The staff of the RPL were interested to know if Bureau staff could more accurately predict the synoptic patterns under which these conditions would occur and Pat Squires and Allan Cornish were seconded to provide the necessary assistance. This resulted in a training manual [56] co-authored by Squires and D. F. Martyn being published towards the end of the war and the suggestion that the cooperation should continue with the return to peacetime conditions (Warren [85]).

Bureau of Meteorology