New satellite data reveals wildfire emissions are 65% higher than thought. Here’s what that means for climate strategy, nature-based solutions and green tech.
Most global fire emissions numbers you’ve seen are wrong – and not by a small margin. New satellite data shows that landscape fires are pumping about 3.4 billion tonnes of carbon into the atmosphere every year, around 65% higher than previous estimates.
This matters because businesses, investors and policymakers are planning net-zero strategies on top of carbon budgets that quietly undercount a major source of emissions. If you work in climate, land-use, or green technology, you don’t want a 65% error hiding in the “natural” side of the ledger.
The latest update of the Global Fire Emissions Database (GFED5) doesn’t just tweak the numbers. It changes how we should think about wildfire emissions, land-use change and nature-based climate solutions – especially in a year when fire seasons in boreal forests, the Mediterranean, and parts of Latin America keep breaking records.
In this article I’ll unpack what GFED5 found, why fire emissions are higher than thought, and how this should shape your decarbonisation plans, climate risk assessments and nature investments.
1. What the new global fire data actually shows
The core finding is direct: global fire emissions average about 3.4 GtC (billion tonnes of carbon) per year over 2002–2022, not the ~2.1 GtC we thought before. That’s a jump of roughly 65%.
A few key points behind that headline:
- Fire emissions now sit at about one-third of fossil fuel carbon emissions (fossil fuels are ~10 GtC per year)
- Around 0.4 GtC per year of fire emissions come from deforestation and tropical peat fires – these are not quickly reabsorbed and behave more like fossil emissions
- Total fire emissions have been roughly stable over the last 20–30 years, but the type and location of fires are changing fast
The reality is simple: the world isn’t necessarily burning more area overall, but more of what burns is high‑carbon forest, and less is low‑carbon grassland and savannah. That shift matters far more for the climate than the raw global area burned.
2. Why previous wildfire emission estimates were too low
Fire emissions are calculated using a basic equation:
Fire emissions = burned area × fuel consumed per unit area
We’ve understood the physics for decades. The problem was the satellites and the resolution we used to measure burned area.
The “small fires” blind spot
For about 20 years, global burned-area estimates were dominated by NASA’s MODIS sensors. MODIS works at roughly 500-metre spatial resolution and is designed to avoid false alarms. A pixel is marked as burned only when fire covers a large part of it.
That’s fine for big forest fires. But it misses a huge number of small fires:
- Patchy burns in savannahs and grasslands
- Smallholder farmers burning crop residues
- Narrow fire lines created for land clearing
- Controlled burns that are small but frequent
With the new GFED5 update, researchers pulled in higher‑resolution satellite data from missions such as ESA’s Sentinel series. Once those small, fragmented burns are included, global burned area almost doubles compared to earlier estimates.
Here’s the kicker: even though each of those small fires is low intensity, together they add up to a large amount of fuel consumed and a much larger global emissions total than we thought.
Not all burned land is equal
The other side of the equation is fuel consumption – how much biomass actually burns per square metre. This varies by orders of magnitude:
- Arid grasslands / savannahs: low biomass, roughly 100–500 gC per m² when burned
- Dense tropical or boreal forests, especially with peat soils: can exceed 1,000–5,000 gC per m²
So two regions can burn the same area but emit wildly different amounts of carbon. Earlier datasets undercounted small fires in open landscapes. The new data corrects that, but it also reveals which high‑carbon landscapes are becoming more fire‑prone.
3. The global fire paradox: less area burned, more carbon at risk
At first glance, recent trends look almost reassuring: total global burned area has been declining in many satellite records. But that’s only half the story.
Why burned area is going down
Burned area is shrinking mainly in savannah and agricultural regions, driven by:
- Land-use change: savannahs converted to cropland in Africa and elsewhere, breaking large contiguous fuel beds into smaller parcels
- Fragmented landscapes: roads, fields and settlements act as firebreaks, limiting the spread of fire
- Tighter air-quality rules: more restrictions on crop residue burning, particularly in regions like the EU
These are mostly low‑intensity fires with modest fuel loads. Losing them reduces smoke and local air pollution, but doesn’t dramatically change the global carbon picture.
Why climate risk is still rising
At the same time, high‑intensity forest fires are increasing in several key biomes. GFED5 shows:
- Forest fire emissions trend upwards, especially in boreal and some tropical regions
- Boreal forests have recorded record-breaking fire years (2023 being the highest on record for satellite-era emissions)
- Fires are becoming more intense, meaning deeper burning into organic soils and higher tree mortality
This shift is driven by climate change:
- Longer fire seasons extend the window when fuels can burn
- Hotter, drier conditions dry out vegetation and soils
- More lightning in some regions adds natural ignition on top of human-caused fires
So even though the global burned area may be falling, the climate and biodiversity risk is going up because we’re burning more carbon-dense ecosystems and damaging their capacity to recover.
4. Why “fires are carbon-neutral” is no longer a safe assumption
The old textbook line was that fires are largely carbon-neutral over time: vegetation burns, releases CO₂, then regrows and takes it back up. That’s still roughly true for many savannah and grassland systems that have burned regularly for millennia.
But for a growing share of global fire emissions, that assumption is now wrong.
Where fire emissions behave like fossil carbon
GFED5 estimates that around 0.4 GtC per year of fire emissions come from deforestation fires and tropical peatland fires. These are critical because:
- Forest is often converted to permanent agriculture or pasture, so there’s no full carbon recovery
- Peat soils contain deep, ancient carbon that took thousands of years to accumulate
- When peat burns, it can smoulder for weeks or months and release enormous amounts of CO₂
Those emissions are effectively one-way, just like burning fossil fuels. They contribute directly and persistently to rising atmospheric CO₂.
Fire, permafrost and long-term feedbacks
In the far north, fires in boreal forests and tundra are starting to interact with permafrost – the huge frozen carbon store locked in high-latitude soils.
High‑severity fires can:
- Remove insulating organic layers that keep soils cold
- Darken the surface, increasing heat absorption
- Speed up permafrost thaw, which then releases CO₂ and methane
This turns some boreal fire events into the opening move of a long-term carbon feedback, not just a short regrowth cycle.
Put simply: as climate change reshapes fire regimes, the share of “safe, short-cycle” fire emissions is shrinking. A rising fraction of fire carbon is now long‑lived and structurally similar to fossil emissions.
5. What this means for climate strategy, green tech and investors
If you’re working on net-zero plans, nature-based solutions or climate risk, the GFED5 update shouldn’t stay in the academic world. It has direct strategic implications.
5.1. Net-zero and carbon accounting
Most companies and many countries:
- Treat landscape fire emissions as background noise or “natural variability”
- Focus their accounting on fossil fuels, industrial processes and deforestation
Given that fire emissions are about one‑third of fossil emissions, that’s no longer defensible.
What to do:
- Integrate fire risk into land-use baselines for any forestry, agriculture or offset project
- Be sceptical of credits from regions with increasing fire intensity, especially in boreal and some tropical forests
- Demand fire‑explicit monitoring (using GFED-type datasets or similar) for nature-based projects
5.2. Nature-based solutions and resilience
High-quality nature-based climate solutions now need to be fire‑smart by design:
- Prioritise landscapes with stable historic fire regimes, such as many savannah systems where managed burning can be part of the solution
- In fire-prone forests, invest in fuel management, mosaic burning and landscape planning to reduce the risk of high‑severity megafires
- Consider fire insurance, buffer pools and dynamic baselines in carbon project design
I’ve seen too many offset project documents with a single line saying “fire risk: low” in regions that have just had their most intense fire seasons on record. That’s not risk management; that’s wishful thinking.
5.3. Opportunity space for green technology
Higher and better-resolved fire emissions data create real opportunities for green tech and climate services:
- Remote sensing platforms that provide near‑real‑time fire risk and emissions monitoring for landowners and insurers
- Early-warning systems that combine climate forecasts, fuel conditions and ignition data to target prevention resources
- Precision agriculture and regenerative practices that reduce the need for residue burning and improve soil health
- Resilient energy planning in regions where wildfires threaten grids, pipelines and renewable infrastructure
If your product or service helps reduce high-intensity fire risk or makes land systems more resilient, you’re operating in a space whose climate value is larger than previously recognised.
6. How to work with fire data in your organisation
You don’t need to be a fire scientist to start using this information. A practical approach looks like this:
-
Map exposure
Identify where your assets, supply chains or projects overlap with:- Boreal forests
- Tropical forests and peatlands
- Fire‑active savannah and agricultural regions
-
Understand the fire regime
Ask: is this area historically low‑intensity (savannah) or high‑carbon (forest/peat)? Are fire seasons lengthening? Are recent years outliers? -
Build scenarios
Consider how a shift from low‑intensity grass fires to high‑intensity forest fires would affect:- Emissions profiles
- Project permanence and returns
- Local communities and biodiversity
-
Update your climate narratives
Ensure your net-zero and sustainability plans explicitly acknowledge landscape fire emissions and how you’re addressing them.
Done well, this turns fire from a background risk into a managed part of your climate strategy.
Where this leaves us
The new GFED5 dataset doesn’t say the planet is suddenly emitting 65% more carbon than we thought. It says we’ve been underestimating the role of fire in the carbon cycle and overestimating how much of that carbon is on a harmless, short regrowth loop.
Fire emissions now sit alongside fossil fuels, land-use change and industrial processes as a major pillar of the global carbon budget. For anyone serious about climate action or green investment, ignoring that pillar is no longer an option.
If you’re planning decarbonisation pathways, building nature-based solutions or developing climate technology, ask yourself one question: does our strategy match the fire reality on the ground?
If the answer is “not yet”, this is the moment to fix it—before the next record fire season does it for you.