Why Does Public Action Change When a Tropical Cyclone Moves from Category 2 to 3?

Geoffrey Chen

When Numbers Start Making Decisions · Season Five, “Things That Have Not Happened Yet” · Article 4

1. Two names on either side of 165 kilometres per hour

Australia’s Bureau of Meteorology classifies tropical cyclones from Category 1 to Category 5. A Category 2 system has typical strongest gusts of 125–164 km/h. Category 3 begins at 165 km/h and extends to 224 km/h. From Category 3, the system is called a severe tropical cyclone. Bureau of Meteorology: Tropical cyclone categories

Only one kilometre per hour separates 164 from 165. Wind does not undergo a physical transformation when an integer is crossed. Public language does: Category 2 becomes Category 3, “destructive” becomes “very destructive”, and “severe” appears. If a coastal council sees the expected gust updated from 164 to 165, media coverage, business closures, household preparation and deployment may all intensify.

Does that make the scheme arbitrary? Risk is continuous, but collective action needs a common language. Without categories, people along hundreds of kilometres of coast would struggle to recognise quickly when wind damage enters a more serious range. The proper questions are what the category measures, what it omits and which actions genuinely depend on it.

2. Category primarily measures wind

The Bureau’s categories are based on maximum mean wind speed near the centre, with typical strongest gusts used in public descriptions. Category 2 has mean winds of 89–117 km/h; Category 3 has 118–159 km/h. Category 3 can produce structural and roof damage, destroy some caravans and increase the likelihood of power failure.

These values are not the wind speed every community will experience. A cyclone has a track and spatial extent. Strongest winds occupy particular parts of the system, while terrain, buildings and local gusts change impacts. Category is also an estimate at a moment. A system can intensify or weaken before and after landfall.

More importantly, category does not express every hazard. The Bureau warns that even a Category 1 cyclone can have major effects through rain and flooding. Storm surge, waves, erosion and physical size cannot be fully represented by one wind category. Dangerous flooding can continue after a cyclone weakens into a tropical low.

Category 3 is therefore not the point at which overall danger first exists. It is the point at which wind damage enters the “severe” classification.

3. The warning clock is not the same as the category line

People can assume a formal warning begins only when a cyclone reaches Category 3. The actual logic is different. A tropical cyclone watch is issued when gales are expected to affect land within 48 hours. A warning is issued when gales are expected within 24 hours or are already occurring. The key variables are damaging winds, time and affected location—not a wait for Category 3.

That distinction is decisive. A Category 2 system nearing land may demand more immediate local action than a Category 3 system far offshore. A lower-category system carrying extreme rain or surge should not have evacuation delayed because its wind number is lower.

Emergency agencies, ports, schools, hospitals and businesses combine the Bureau’s track map, watch or warning, local water conditions and road access. Category creates context. It should not impersonate a complete instruction.

If an organisation adopts a rule that it closes only at Category 3, it must show that the threshold suits its place and risk. The Bureau did not make that organisational decision and cannot absorb its responsibility.

4. The forecast category is itself uncertain

Before landfall, track and intensity are predictions. A displayed area of possible movement does not mean each point within it has the same probability, and it does not guarantee that places outside are safe. Rapid intensification can sharply reduce preparation time.

An update from Category 2 to Category 3 does not necessarily mean the earlier forecast was negligent. New satellite, radar, aircraft or ocean information may change the best estimate. Institutions must be able to escalate when evidence changes rather than defend an old number for the sake of consistency.

Conversely, a cyclone forecast to reach Category 3 that arrives at Category 2 does not automatically prove that closure or preparation was wasted. The decision should be judged from the probabilities, consequences and options available at the time. A forecast has no practical value if action is allowed only after the worst case becomes certain.

“Better to overreact” cannot justify unlimited measures either. Repeated closures, compulsory evacuations and resource movements impose real costs and can create warning fatigue. Systems should examine false alarms, misses, available lead time and uneven effects and adjust their action matrices over time.

5. People on either side bear different costs

An upgrade can close shops and schools, cancel flights, halt ports and move hospital patients. These decisions protect life while affecting hourly workers, people without cars, carers and small firms. Those with more resources can often leave earlier; those with less may need public help earlier.

One category should therefore not produce only a uniform slogan. Authorities must state which place faces which hazard, when action is needed, what assistance is available and when measures will be lifted if the track changes.

Communication around a category boundary can also distort attention. Media may treat “severe” as the only newsworthy transition and understate Category 2 flooding. Organisations may attach every Category 3 forecast to identical action because they fear criticism for insufficient response. The number coordinates, but it can also erase relevant differences.

6. Explanation and correction cannot wait until afterwards

An emergency cannot deliver a full model report to every resident. It can still communicate the current category, expected development, track uncertainty, main hazards, watch or warning area, next update time and the authority issuing the action.

Ending a warning is also a decision. If winds decline while flooding and road hazards remain, announcing only that the cyclone is now Category 1 may encourage people to return too early. Information must distinguish hazards that fall with category from those that persist.

After the event, reviews should separately examine intensity prediction, track, warning time, local action and public understanding. Final casualties cannot serve as the only score for an entire system. Nor can every weakness be excused by saying the cyclone changed quickly. If particular communities repeatedly receive information late or evacuation routes repeatedly fail, the defect is institutional as well as meteorological.

7. Categories organise a shared time

The deeper function of a cyclone category is not merely describing wind. It brings many strangers into a shared rhythm of preparation. Forecast models, emergency agencies, hospitals, schools and households coordinate around a classification that continues to update. No individual possesses every relevant fact, yet distributed knowledge can be called into action through categories, maps and warnings.

The danger is obedience to “Category 3” without knowing that it mainly refers to wind or who actually issued an instruction. Responsibility must remain named. The Bureau owns the meteorological assessment. Emergency authorities own protection and evacuation decisions. Businesses own arrangements for workers. A number can coordinate authority; it should not anonymise it.

Conclusion: Category 3 should change the language, not decide every action alone

Australia should retain its five cyclone categories and the “severe” label from Category 3. Clear levels communicate escalating wind damage quickly and are indispensable for coordination across regions.

But any action matrix should use at least four dimensions: wind category, arrival time, location and track, and other hazards. Watches and warnings are not the same thing as categories. Closure, evacuation and deployment must come from accountable authorities with activation and release conditions.

There is no natural cliff at 165 km/h. The threshold translates continuous wind risk into a public language. A mature institution neither abandons the line because it was designed by people nor gives it every judgement because it carries an official name. Before the future has happened, we need simple signals. The greater the consequence, the more clearly those signals must retain what they do not say.

Primary sources

Discover more from Geoffrey Chen

Subscribe to get the latest posts sent to your email.