In the early hours of 20 March 2019, Cyclone Trevor made landfall near the remote community of Borroloola in the Northern Territory. The storm had been intensifying over the warm waters of the Gulf of Carpentaria for days, and its eventual path was the product of a complex interplay of atmospheric forces. For the meteorologists at the Bureau of Meteorology, the challenge was not merely to predict where Trevor would go, but to communicate the certainty—and the uncertainty—of that prediction to people whose lives depended on it. The context of a cyclone's path is never purely physical; it is also social, economic, and political. Understanding how forecasters produce that path, and how communities interpret it, reveals the power structures embedded in the science of severe weather.

The physical context of a cyclone's trajectory begins with sea surface temperature. Cyclones draw their energy from warm ocean water, typically above 26.5°C. As the storm moves, it leaves a wake of cooler water, which can weaken it. But the steering currents that guide a cyclone are determined by larger-scale weather patterns: the subtropical ridge, the monsoon trough, and upper-level winds. In the Australian region, cyclones often track westward or southwestward, but variations in these steering flows can produce erratic paths. For Cyclone Trevor, a strong ridge to the south pushed the system westward, while an approaching trough later turned it southeastward. Each shift in the steering flow had to be captured by numerical weather prediction models, which simulate the atmosphere using millions of calculations.

The power to predict a cyclone's path lies in the quality of data and the sophistication of models. Satellites provide continuous imagery of cloud patterns, while reconnaissance aircraft—though rarely used in Australia—can measure pressure and wind speed directly. Buoys and weather stations add surface observations. All this data is assimilated into models such as the Australian Community Climate and Earth-System Simulator (ACCESS). These models divide the atmosphere into grid cells and solve equations of fluid dynamics and thermodynamics. The resolution of the grid matters: a coarser grid might miss small-scale features that steer a cyclone. Even with high resolution, small errors in initial conditions can grow, so forecasters run ensembles—multiple simulations with slightly different starting points—to gauge the range of possible tracks.

But the steering currents that guide a cyclone are determined by larger-scale weather patterns: the subtropical ridge, the monsoon trough, and upper-level winds.

The output of these models is a probability cone, the familiar graphic that shows the most likely path and the spread of possible positions. But the cone is often misinterpreted. Many people assume that the storm will stay within the cone, when in fact the cone represents the historical error of forecasts at each lead time. For Cyclone Trevor, the cone narrowed as landfall approached, but the uncertainty remained significant. The power of the forecast lies not just in its accuracy, but in how it is communicated. Emergency services use the cone to decide evacuation zones, but residents may see a wide cone and assume the threat is low, or a narrow cone and assume certainty. The context of the forecast—its history, its limitations—must be part of the message.

The social context of cyclone tracking is shaped by past experiences and cultural knowledge. Indigenous communities in northern Australia have long read the signs of approaching storms: changes in wind, cloud, and animal behaviour. For the Yanyuwa people of the Gulf region, seasonal calendars mark the time of 'strong winds' and 'big rains'. These traditional knowledge systems coexist with modern meteorology, but they are not always integrated into official warnings. The power to define what counts as evidence—and whose knowledge is authoritative—rests largely with scientific institutions. When Cyclone Trevor approached Borroloola, the official warnings were based on satellite data and model output, not on local observations of the sky. This can create a gap between the forecast and the community's trust in it.

The economic context adds another layer. Cyclones disrupt mining, agriculture, and tourism, and the cost of evacuations and preparedness is substantial. For remote communities, the decision to evacuate involves not only the cyclone's path but also the availability of transport, shelter, and resources. In 2019, the evacuation of Borroloola required coordination between the Northern Territory Emergency Service, the Australian Defence Force, and local councils. The power to order an evacuation rests with authorities, but the decision to comply rests with individuals. Studies after Cyclone Trevor showed that some residents stayed because they feared looting, or because they had no way to transport pets or livestock. The forecast alone does not determine behaviour; it is filtered through personal circumstances and trust in institutions.

Finally, the political context of cyclone tracking involves questions of funding, infrastructure, and climate change. As the planet warms, the potential for more intense cyclones increases, though the total number may not rise. In Australia, the Bureau of Meteorology has faced budget cuts that affect its ability to maintain observation networks and upgrade models. The power to invest in forecasting is a political decision, and it has consequences for the accuracy of warnings. Cyclone Trevor was a Category 4 storm when it crossed the coast, but it struck a sparsely populated area. A similar storm hitting a major city would test the limits of the warning system. Understanding the context and power of cyclone tracking means recognising that the science is embedded in a web of social, economic, and political forces that shape both the forecast and its impact.