Fire prevention in Indonesia’s Riau province takes on many forms – from the community-based Fire Free Village Programme and the collaboration and responsiveness of fire suppression teams on the ground to the vital contribution of unseen eyes in the sky.
In the latter case, it’s the job of satellites and those who monitor them to identify ‘hotspots’. The term itself has become synonymous with efforts to prevent fires in Sumatra, entering into common parlance. And while it often happens that a hotspot identified from thousands of meters up turns out not to indicate a fire threat –the commitment to a fire-free future means that every warning sign must be investigated.
We spoke to APRIL Fire Prevention Manager Craig Tribolet about hotspots and how monitoring them as part of a comprehensive approach to fire management plays an important role in keeping Riau fire-free.
Can you define a hotspot?
We use satellite data from two separate NASA-based platforms to help monitor potential fire risk within the APRIL concessions; obviously along with other elements of our fire prevention program.
One of the instruments, MODIS, is on NASA’s AQUA and TERRA satellites. The instrument identifies any area where the temperature has reached more than 37 degrees C at night or 42 degrees C during the day. Our ground crews will then look for signs of fire within a 1.1km radius. The other more recent platform, VIIRS, is on board a NASA/NOAA satellite and uses visible infrared. When this identifies a hotspot within a spectral band, our crews ground-truth within a 375m radius. It’s important to note that the satellites don’t literally pick up a spot, they identify an area which has to be physically checked to find out where the actual hotspot is.
What kind of technology is involved in identifying hotspots and who operates the satellites?
The satellite based imaging comes from the two NASA-based platforms and is analyzed with specific hotspot algorithms via a NASA service called Fire Information for Resource Management System (FIRMS) which sends out the near real-time data within about three hours of the satellite passing over our concessions.
Hotspot data can easily be viewed by the general public using platforms like the Global Forest Watch website or the ASEAN Haze Online maps.
Who does the actual monitoring and lets you know if a hotspot has been identified?
Every morning we get a satellite download which includes all the recent hotspot data with a latitude/longitude plotting any hotspots within or adjacent to our concessions.
Our own Planning team analyses the data and plots all the latitude and longitude information on a map. Any spot that falls within or adjacent to our concessions is sent out to those individual Estate Fire and Protection Teams who are then tasked with ground-truthing each and every one of those hotspots.
On average how many hotspot alerts do you receive?
During the summer we can have maybe two or three dozen hotspots in a day. At other times we can also have none.
When we have a very large number, it’s a challenge because a hotspot identified by MODIS can be anywhere within a kilometer radius; with VIIRS it’s somewhat easier because it’s 375 meters. When it’s a very large number, we use a helicopter and it’s quite easy to check out some of them; often it turns out to be a tin roof or a barge or a mosque.
APRIL Fire Suppression Team at the Control Room
What’s the chain of events once a hotspot alert goes out?
The Estate Fire and Protection teams use a GPS to locate the hotspot based on the information from the Planning team. This is also used as evidence that we have acted on the data and undertaken an inspection.
Sometimes we need to drive a couple of hours from the estate to ground truth an area. In difficult terrain or really remote areas they will also use a drone to get an overview of the landscape. The drones are quite important because we can get one in the air within about five minutes of driving to an area and a drone can give us a pretty reliable view.
The Fire teams will also do a risk analysis bearing in mind that 100% of fires are human caused; the teams don’t just drive around blindly hoping they are going to come across a hotspot. So they prioritize high risk areas during their inspections – recent clearing or areas of illegal logging, new agricultural development, access points like roads and rivers, houses amongst other factors.
On a clear day, in looking for a fire it’s not too hard for the teams to see or smell smoke. It often can take us a few hours to verify that there is no threat which actually is harder than finding something.
When you start getting dozens of hotspot alerts, that’s when we’ll put the helicopter up to verify from the air. Of course there’s a different chain of events that kicks in if a hotspot is confirmed as a fire.
How important is this hotspot monitoring in the overall prevention of fires?
Monitoring is critical to ensure rapid response – the earlier the detection, the smaller the fire and the easier and safer to control. So near real time remote sensing tools like Hotspots can be very useful – particularly in remote areas that ground patrols may struggle to access. They form part of a broader monitoring toolkit that includes regular ground and air patrols as well as information sent from communities to our 24hr Fire Hotline.
What percentage of hotspots turns out to be fires?
The reality is that only about 5% of hotspots are ever recorded as fires – the bulk of them tend to be locations in the landscape that have heated up like bare soil, shallow water as well as human structures like tin roofs. Also, cloud and smoke can seriously limit the ability of the satellite to see fires.
But having said that, monitoring for hotspots is an absolutely vital part of the overall detection picture because our goal is zero fires. There’s not one element of our fire prevention toolkit that we can exclusively rely on. We work on the ground through inspections and community initiatives, we work with drones and helicopters, and then we work much higher up with satellites and monitoring technology.
