01
Authenticate
Exchange your client ID and secret for a Bearer token using the OAuth 2.0 client credentials flow.
Fire Simulation API · For developers
Put fire behaviour inside your own product.
Send an ignition point. Get back where the fire will go, when it will arrive and how hard it will burn — delivered to your webhook, ready to show your users. No fire science team required.
Your platform
- Fire detection or incident
- Underwriting model
- Dispatch system
- GIS or dashboard
One request
POST an ignition
Latitude and longitude
Start time
Duration and interval
JSON in · ID back
FiSci runs it
Physics-based simulation
Vegetation and fuel
Terrain
Weather
Back to you
Webhook delivery
Arrival time · Rate of spread
Intensity · Flame height
Frame by frame
FiSci maintains the fuel, terrain and weather data behind every run. You send coordinates; you get fire behaviour back.
Why it matters
Your users want to know what the fire will do next. Most platforms can only show where it is.
Detection, exposure and incident data tell you where a fire is, or could be. Fire behaviour is the missing layer — and it has always been expensive, slow and specialist to build. The API makes it one request.
Building it yourself
With the API
Building fire spread modelling in-house means a fire science team, fuel and terrain data for every location you cover, live weather feeds and years of validation.
FiSci handles the vegetation, fuel, terrain and weather. Your team sends a coordinate and gets results back.
A detection, a policy location or an incident pin tells you where. It says nothing about when fire will arrive, how fast it will run or how hard it will burn.
Every point becomes a forecast: arrival times, rate of spread, fireline intensity and flame height, up to 24 hours ahead.
Off-the-shelf fire tools live in someone else’s interface. Your users leave your product to get the answer, and the answer never reaches your data.
Results land in your own stack as GIS-ready layers, displayed under your own brand inside your own product.
Physics-based simulation takes real compute time. Wire it in synchronously and your system waits on every run.
Requests return immediately and results arrive at your webhook. Queue hundreds of simulations without blocking anything.
A number from a black box is hard to stand behind when an underwriter, a dispatcher or a regulator asks how it was reached.
The same physics-based engine that runs inside Mitigate, with defensible, audit-ready outputs.
How it works
02
Send an ignition
POST a latitude and longitude, with an optional start time, duration and output interval. JSON in, JSON out.
03
Get an ID back
The API returns a simulation ID straight away, and the fire model starts running in FiSci’s cloud.
04
Receive results
When the run completes, FiSci posts the full results to your registered webhook — layer by layer, frame by frame.
The request
One call. One webhook.
A single POST starts a simulation. Only the ignition is required — everything else has a sensible default, so a first integration can be running the same day.
- ignition
- Required. The fire’s start point as longitude and latitude in decimal degrees.
- start_time
- ISO 8601 datetime. Defaults to now. Use a past timestamp to replay a known event, or a future one to model what is coming.
- duration
- Hours of fire spread to simulate. Longer for risk assessment, shorter for rapid screening.
- interval
- Minutes between output frames. Finer for smooth animation and per-step statistics, coarser for lighter payloads.
- reference
- A label of your own — an incident name, a detection ID, a policy number — carried through to the results.
Request
POST /v1/projects/123/fire-simulation
Authorization: Bearer <token>
{
"ignition": {
"longitude": 150.576,
"latitude": -33.887
},
"start_time": "2025-10-10T14:00:00+11:00",
"duration": 8,
"interval": 60,
"reference": "Wattle Creek Fire"
}Response
{ "data": { "simulation_id": "abc-123" } }What comes back
Four layers your users can act on.
Each interval of the run arrives as its own frame, so you can animate the spread, query a single moment, or pull the statistics for any point in time.
Arrival time
Isochrones showing when the fire front is projected to reach each part of the landscape — and the assets on it.
Rate of spread
How fast the fire is moving, cell by cell, through each interval of the run.
Fireline intensity
Heat output in kW/m — the measure that separates a fire crews can work from one they cannot.
Flame height
The figure crews use to judge tactics and defensive positions.
Where it fits
Built into the tools people already use.
Detection and incident systems
However a fire is first reported — a satellite hotspot, a camera, a lookout, a crew on the ground, or an ignition added by hand to your incident mapping software — show where it is likely to be in one, four and twelve hours.
Insurance and underwriting
Model exposure for a portfolio or a single policy, and replay historical fire weather against today’s book.
Emergency management
Put a spread forecast beside every incident in dispatch, so resourcing decisions start from where the fire is going.
Risk and asset dashboards
Give asset owners a live view of which sites a fire could reach, and how soon.
GIS platforms
Generate fire-risk layers on demand and serve them alongside everything else your users already map.
Scenario modelling
Batch thousands of ignitions through climate or seasonal scenarios in a pipeline, without a person in the loop.
Why build on FiSci
REST
Simple to integrate
Plain HTTP and JSON. If your team can call an API, they can run a fire simulation.
Async
Built for pipelines
Non-blocking requests and webhook delivery keep your system decoupled from simulation runtimes.
Sandbox
Test without risk
A separate test environment, limited to a set test area, for checking authentication, data formats and webhooks before you go live.
Your brand
White-label rights
Show FiSci simulation outputs under your own brand, inside your own interface.
Simulation outputs are model predictions based on the fuel, terrain and weather data available at the time of the run. They support decision-making and do not replace official warnings or instructions from emergency services.
Make your first call this week.
Request sandbox credentials and onboarding documentation from our integration team, and run your first simulation before you commit to anything.