Predicting wildfires could be the key to stopping them. Wildfires have increasingly become a threat to humans, wildlife and climate change. Effects are both immediate and long-term, destroying land, property and accruing substantial expenses.
Wildlife Aversion by Forecast and Early Response System (WAFERS) project lead, Abdulmohsen Aleissa explains, “In the long run, these wildfires play a major role in contributing to global climate change, which, in turn, amplifies the scale and frequency of future wildfires. This calls for immediate action to interfere with this vicious cycle of exponentially intensifying yearly wildfires.”
While most wildlife disaster responses are reactive, WAFERS strives to change this. Typically, wildfires are an extensive research area involving supercomputers, live satellite images and labor-intensive workflows.
Time-consuming steps such as liaising with on-site firefighters, retrieving satellite imagery and weather conditions, and running simulated computer models are costly. Simulations can take up to three days to run and with wildfires covering up to 60km a day it’s critical to expedite the process; that’s where the WAFERS system comes in.
The two-part WAFERS system entails a computer algorithm and custom-made drone featuring thermal and RGB cameras. The WAFERS team decided to design their own drone to control the inputs.
WAFER’s rotary drone is strong-suited for shorter flights over rugged terrain and can take off from anywhere opposed to fixed-wing drones.
After taking to the sky to collect necessary data, images are uploaded to the photogrammetry software Pix4Dmapper. A generated thermal map is then fed to the WAFERS computer algorithm for evaluation.
The algorithm automatically runs the simulation when a fire is detected to model the spread pattern over time. Other capabilities of the algorithm include “what-if analysis”, which assesses the fire’s damage potential before it takes hold. Only requiring a simple orthomosaic map of the area to estimate the pattern.
Through concepts of combustion, heat transfer and thermodynamics the powerful algorithm evaluates data and utilizes databases of weather forecasts and past fires to accurately predict the wildfires behavior.
Wildfires are dangerous and stressful situations for everyone. With most first responders being localized and may have family or friends in the blaze path, WAFERS has made every effort to take the severity and sensitivity into account. Not to mention, the supercomputers used to run these models are a small charge compared to loss of life, material and environmental damage. In 2018, Camp Fire wildfire killed 85 people, and burned 150,000 acres causing $16.5 billion in damage.
While all projects need to be tested and verified, preliminary results have been excellent. In 2020, the WAFERS team ran data from their Australian grass fire experiment and results were highly accurate.
“The biggest challenge was to minimize the amount of inputs necessary to run the simulation, while still maintaining the accuracy needed to allow firefighters to make decisive action, such as placing fire barriers, or digging trenches around the fires,”
“Additionally, we needed to make the algorithm run as efficiently as possible to make it less demanding in terms of processing power. This allows fire responders to use project WAFERS on-site, with equipment as simple as a drone and a portable laptop.” says Aleissa.
The WAFERS team won the Pix4D Climate Contest in 2019 and has been working hard to be ready for the upcoming seasons.