'And now among the fading embers,
These in the main, are my regrets,
When I am right no one remembers,
When I am wrong no one forgets.'

Indeed, there was much excellent and successful forecasting against these odds; there was also much failure, disappointment and confusion.

When RAAF Met. officers were first posted to tropical stations, they applied customary methods, endeavouring to identify boundaries between the equatorial maritime and tropical maritime aimasses of both hemispheres. They were able to perceive that there was a line or zone of demarcation between, on the one hand, the aimasses which had migrated from the north Pacific or the Asian continent, and on the other, aimasses of southern origin. This line or zone was referred to as the inter-tropical front. Later, however, this came to be known as a zone of inter-tropical convergence of airmasses. Famous Australian Spitfire ace pilot, Group-Captain C. R. (Killer) Caldwell told me that this inter-tropical zone provided some testing occasions. He related how he once flew for four and three quarter hours from Morotai to Dulag in heavy rain, and then for three and a half hours on instruments.[4]

Forecasters abandoned the notion of a well-defined front and frontal analysis as normal in higher latitudes, and concentrated on identifying convergence zones with which, in the tropics, bad weather is invariably associated. Convergence zones are lines or areas where air flow converges, producing up-motion which results in cooling of the air, condensation of moisture, cloud and rain. Conversely, zones of divergence, from which air flows outward, usually denote more stable weather conditions. As W. J. Gibbs noted:

'We became aware of the dominating influence of orography and the fact that diurnal variations in cloud and rainfall patterns strongly overshadowed interdiurnal variability. Gradually we developed our own synoptic models to replace those which proved irrelevant in the tropical atmosphere.'[5]

Bureau of Meteorology