Cyclone Evacuation Decision after a Costly False Alarm
A coastal province is 36 hours from the forecast landfall of a Category 3 cyclone. The forecast track has a 70% probability cone 120 km wide, and landfall is expected at night during a high tide. Fourteen villages with 45,000 people lie in the cone. There are three cyclone shelters with total capacity 8,000; one shelter is reachable only at low tide. The fishing fleet has 300 boats. SMS warnings reach about 60% of households; loudspeakers and religious leaders are the main alternatives. Last season, authorities evacuated 20,000 people for a cyclone that veered 90 km offshore; no one was hurt, but evacuation costs and disruption were heavily criticised in local media. This year, several village leaders say they will wait for 'real rain' before moving. The provincial disaster office must decide on evacuation timing and public messaging within the next six hours.
1. What evacuation decision and message should the provincial disaster office issue in the next six hours?
Read the worked answer
Issue a staged evacuation now, not a single all-clear-or-evacuate order. First, move the most vulnerable people and the villages with the longest travel times, including the shelter reachable only at low tide, before the tide cuts access. Second, start general evacuation of the 14 villages during daylight so people are not moving at night into a high-tide storm surge; aim to complete before dark. Third, because official shelters hold only 8,000 of 45,000, identify and publicise safe alternatives: strong community buildings, raised schools, and vertical evacuation to upper floors of reinforced structures. Secure or move the 300 boats if safe. The message should be probabilistic and action-oriented: 'The current forecast shows landfall near you at night with a high-tide surge. We are evacuating now because the risk is too high to wait. The track may change; we will update you.' Use SMS, loudspeakers, religious leaders and village committees, and repeat the timing and routes. Do not wait for real rain; waiting until rain is visible leaves no margin for nightfall and tide.
2. How should the office rebuild credibility after last season's false alarm and counter village leaders' wait-and-see stance?
Read the worked answer
Rebuild credibility through transparency and participation, not defensiveness. First, acknowledge the cost and disruption of last season's evacuation and thank people. Second, explain in plain language that a forecast is a probability, not a promise: last season's evacuation was a reasonable protective action when the track threatened the coast, even though the cyclone veered. Distinguish a false alarm from a correct warning under uncertainty. Third, share the current probability cone and expected timing, and explain why night landfall and high tide make this different. Fourth, involve village leaders and religious leaders in two-way dialogue; ask them to relay the message and report back what people are saying. Fifth, commit to a post-season review and to improving triggers, and avoid blaming communities for wait-and-see behaviour. Finally, make the warning system more local: village committees with loudspeakers, clear flags or bells, and a named person responsible for each hamlet. Credibility grows when people see that warnings are consistent, explained and followed by feedback.
3. Which indicators should be monitored to adjust the evacuation and shelter plan in the next 24 hours?
Read the worked answer
Monitor a small set of decision-relevant indicators every three to six hours: updated forecast track, intensity and arrival time; storm-surge and tide forecasts; observed rainfall and road flooding on evacuation routes; shelter occupancy against capacity; the status of the low-tide shelter access; number of vulnerable people registered and transported; boat haul-out progress; SMS delivery and loudspeaker coverage; and reports from village committees about people refusing to move. Set decision points in advance: if the track shifts closer, extend evacuation to additional villages; if it shifts away, do not cancel evacuations already underway until the threat is clearly past, but communicate the reduced risk. If roads flood, redirect to safe vertical evacuation. Use the indicators to update the public message with specific reasons, not just that the situation is changing. Document decisions and timestamps for the after-action review.