At the end of most fire seasons, someone is asked to reconstruct what happened. We think the review is where the next season is won or lost, and it cannot be done from memory.
When the rains arrive, the fire season ends and the review begins. The questions are reasonable. How many fires did we have? How fast did we respond? How many hectares burned? Were the patrols done? Which ranges did well, and which struggled?
The answers, in many divisions, are rebuilt after the fact. Officers scroll back through chat groups, dig through email for satellite alerts, read patrol registers and ask crews what they remember. The result is a report that looks precise but rests on recollection. Lessons drawn from it are, honestly, guesses.
Fire season is exactly the setting in which memory is least reliable. Incidents overlap. The same fire is reported by a tower, a satellite pass and two rangers, and in the retelling it becomes one fire or three. Times are approximate because nobody wrote them down while the crew was on the line. A patrol register shows that a patrol was claimed, not where the patrol went.
None of this is anyone’s fault. People were fighting fires, not keeping records. The fix is not to ask them to write more; it is to capture the record as a side effect of the work itself.
| Question | From memory | From the record |
|---|---|---|
| When did we know? | “Sometime in the afternoon” | The satellite pass, tower call or ranger photo, with its time |
| How fast did we respond? | An estimate | Acknowledgement, dispatch and arrival times per incident |
| Were patrols done? | A signed register | A GPS track and photos per patrol |
| How much burned? | Reconstructed later | Kept current through the season |
| What went wrong on the worst day? | Competing recollections | A replay of who knew what, and when |
Season totals are useful for the annual report. They are not where the lessons are. The lessons are in the sequence of a single bad fire: when the first sign appeared, how long it took to become an incident, who owned it, when a crew was sent, and what was happening elsewhere at the same moment. A replay of that sequence turns a vague sense that “we were slow” into a specific finding, such as the gap between a ranger’s photo and the desk acknowledging it.
Replay also changes the tone of a review. Arguments about whose account is right give way to a shared look at the same timeline. That makes it easier to fix processes without blaming people.
ForestWatch records each fire from the satellite pass to containment, with sources. Ranger photos and voice notes on WhatsApp become located incidents with an owner and a clock; fire-line photos, tracks and containment are confirmed on the record; and each patrol carries a GPS track and photos as proof of work. Response times, hectares and patrol coverage are kept in a running season report, and any fire can be replayed from first detection to containment, alongside division scorecards and an audit trail.
For the field side of this, read the playbook for ranger fire reports that don’t get lost; for the pre-season work, the fire season preparedness checklist. Everything else is in ForestWatch guides and tools.
A review that runs on evidence is less comfortable. It is also the only kind that makes the next season better.
They are often reconstructed after the season from chat groups, emails, registers and recollection, so times, counts and patrol claims are approximate.
Timed records captured during the season: when each fire was detected, acknowledged, dispatched and contained, patrol GPS tracks and photos, and a replay of the worst fires.
For learning, yes. Totals suit the annual report; the sequence of a single bad fire shows exactly where time was lost.
See it on your own data. ForestWatch — Every hotspot, ranked by risk. Book a 30-minute working session with an engineer.
