This year’s fire season in Australia feels unpredictable. One week brings torrential rain, the next a burst of heat, followed by a sharp cold snap.
This kind of “climate whiplash” — rapid swings between wet and dry conditions — is making it harder to know when the landscape is safe and when it’s ready to burn.
The latest official forecast reflects that same uncertainty. While large areas are likely to experience wetter-than-normal conditions for the rest of spring, parts of inland Western Australia, South Australia, and Victoria face an elevated risk of fire due to persistent soil dryness and above-average temperatures.
Even regions showing a “normal” outlook aren’t immune; catastrophic fires can still occur under the right mix of heat, wind, and fuel dryness.
But what if we could see how dry the vegetation is in near real time before fires start? That’s now possible thanks to a publicly available tool implemented by Geoscience Australia.
This tool draws on more than a decade of research that began when I was a Ph.D. student in Spain and has evolved through collaborations with colleagues at the Australian National University and emergency services.
It’s a glimpse into the future of bushfire management, where satellites and other technologies act like a nervous system for the continent, constantly sensing and responding to changes in the landscape long before smoke fills the sky.
As Australia faces longer, more erratic fire seasons, tools like this may help us all stay one step ahead of the flames.
How do we measure fire danger?
Fuel moisture content is the percentage of water inside leaves and twigs compared with their dry weight. When it drops below a certain level, vegetation ignites more easily, burns hotter, and spreads faster.
It is one of the key predictors of fire danger because, for example, it strongly influences whether a source of ignition, such as a lightning strike, stays small or grows into a fast-moving wildfire.
Traditionally, measuring fuel moisture content relies on cutting samples in the field and weighing them wet and dry. This is a precise, but time-consuming, method that can only cover small areas.
Satellite-based estimates have existed for years, providing valuable, continent-wide data for seasonal bushfire outlooks.
But their coarse resolution means they’re too broad for guiding local decisions such as planning prescribed or cultural burns, and only a few of these earlier approaches were made operational.
Turning satellite light measurements into moisture maps
The new bushfire management tool my team helped develop fixes this problem.
It uses a machine-learning model trained on fuel moisture content estimates developed as part of a precursor tool — the Australian Flammability Monitoring System.
This older system produced continental fuel moisture content at 500-meter resolution every four days since 2001 using data from NASA’s MODIS satellite sensor. It was validated and calibrated using extensive field measurements and biochemical reference data.
This provided a solid foundation for the new model, which provides near-real-time, 20 m-resolution maps showing how wet or dry vegetation is across the continent. It does so by drawing on satellite imagery from the European Space Agency’s Sentinel-2 satellites, updated every five days with historical coverage from 2015 onward.
The new version applies the same physical principles to higher-resolution imagery.
Water in leaves strongly absorbs radiation in the shortwave infrared region. This means dry vegetation reflects more light than hydrated, green plants. Meanwhile, as plants lose moisture, the amount of visible and near-infrared radiation they reflect also changes. Chlorophyll breaks down, leaf structure deteriorates, and the red and infrared light signals weaken.
By learning from these patterns, the model can infer how much water is inside the vegetation. This effectively turns satellite light measurements into maps of live fuel moisture across Australia.
To cross-calibrate cultural indicators of flammability, such as plant color, scent, and seasonal cues with satellite observations, we have collaborated with Indigenous fire practitioners.
Indigenous-led field surveys across New South Wales have confirmed strong alignment between these traditional indicators and satellite results.
This two-way learning strengthens both scientific and cultural understanding. It ensures national monitoring systems are informed by generations of Indigenous knowledge about landscape health. It also enhances fire practice and community resilience through cutting-edge Earth observation tools.
Seeing the danger long before it ignites
With every update, the tool provides a continent-wide snapshot of how flammable Australia’s vegetation is.
Thanks to its higher spatial resolution, it can reveal subtle gradients in vegetation dryness that coarser sensors simply miss. This helps fire agencies and Indigenous communities pinpoint where prescribed or cultural burns can be carried out safely — and where fuels remain too moist to ignite.
The data are also being used in insurance and risk modeling for new developments to quantify the number of “flammable days” in a given region.
Meanwhile, the CSIRO is also collaborating with the Bushfire Research Centre of Excellence to better represent changing fuel conditions in fire-behavior models. As part of this effort, our new fuel moisture content product is being integrated into the CSIRO’s fire spread simulation model.
In bushfire management, the advantage lies in seeing the danger long before it ignites. As Australia’s fire seasons lengthen and the weather swings between extremes, knowing how wet or dry our landscapes are may become as important as forecasting heatwaves or storms.
This marks a shift from reacting to fires to anticipating and preventing them. Other projects, such as OzFuel, will accelerate this by filling other crucial gaps in sustained, high-resolution monitoring of fuel conditions across Australia.
The author acknowledges the many scientists, fire practitioners, and Indigenous partners whose work made the new fuel moisture content tool possible, and thanks Geoscience Australia for its operational implementation.
Marta Yebra, Director, Bushfire Research Centre of Excellence, Australian National University
This article is republished from The Conversation under a Creative Commons license. Read the original article.
Follow us on X, Facebook, or Pinterest
