Global Forest Loss Rate Slows, but El Niño-Driven Wildfires Threaten Recovery

The world's tropical forests tell two stories simultaneously. In one, deforestation from agriculture and logging has declined measurably over two decades — the result of moratoriums, enforcement crackdowns, and international pressure. In the other, climate-driven wildfires are destroying forests at a pace that threatens to overwhelm those gains entirely.

In 2024, the tropics lost 6.7 million hectares of primary forest — the highest figure on record — with fires surpassing agriculture as the leading cause of tropical forest destruction for the first time [1]. One year later, loss dropped 36% to 4.3 million hectares, driven largely by Brazil's enforcement of deforestation controls [2]. But even that reduced figure remains 46% higher than a decade ago, and the three-year average of fire-driven loss has doubled compared to twenty years prior [2].

The question facing policymakers, scientists, and the 1.6 billion people who depend directly on forests is whether the conservation framework built over the past decade can survive a climate that is making forests increasingly flammable.

Source: Global Forest Watch / WRI

Data as of May 1, 2026CSV

The Measured Slowdown: Where Progress Is Real

The long-term trajectory of deforestation is genuinely downward. According to the FAO, the annual rate of net forest loss fell from 10.7 million hectares per year in the 1990s to 4.12 million hectares per year between 2015 and 2025 [3]. Deforestation — defined as permanent conversion of forest to other land uses — slowed from 17.6 million hectares annually in 1990-2000 to 10.9 million hectares in 2015-2025 [3].

Brazil accounts for the most significant national-level improvement. After record deforestation during the Bolsonaro administration, the Lula government relaunched the PPCDAm anti-deforestation framework in 2023. IBAMA, Brazil's environmental enforcement agency, increased violation notices by 81% and fines by 63% between 2023-2025 compared to the 2020-2022 period [2]. Non-fire primary forest loss in Brazil reached its lowest rate on record in 2025, with a 41% decrease from the previous year [2].

Indonesia has also contributed to the decline, though its trajectory is less stable. The country reduced primary forest loss by 11% from 2023 to 2024, and its moratorium on new oil palm, timber, and logging concessions — backed by a $1 billion partnership with Norway — yielded modest but measurable results [4]. However, tree cover loss increased from 0.89 million hectares in 2022 to 1.12 million hectares in 2024 following the end of certain moratorium provisions [4].

Colombia recorded a 17% decline in 2025, while other countries in the Congo Basin moved in the opposite direction. The Democratic Republic of Congo and Republic of Congo both recorded their highest levels of primary forest loss on record in 2024, with fires causing 45% of the damage in the Republic of Congo [1].

Source: Global Forest Watch / WRI

Data as of May 1, 2026CSV

Fire as the New Dominant Driver

The 2024 data marked a structural shift in the causes of forest loss. For the first time in the satellite record, fires — not agricultural clearing — were responsible for approximately 50% of all tropical primary forest destruction [1]. This was not a natural phenomenon. The 2023-2024 El Niño event triggered the most severe drought on record across the Amazon basin, creating conditions where fires set to clear land for agriculture escaped into standing forest [5].

In Brazil, fires caused 66% of the country's tropical primary forest loss in 2024 — a more than sixfold increase from 2023 [1]. Bolivia's primary forest loss skyrocketed by 200%, totaling 1.5 million hectares, with more than half attributable to fires that began as agricultural burns but became megafires during the drought [1].

The carbon implications are substantial. Fire-driven forest degradation in the Amazon alone released an estimated 791 million metric tons of CO₂ in 2024, a sevenfold increase compared to the previous two years, according to the European Commission's Joint Research Centre [5]. Globally, the 30 million hectares of tree cover lost in 2024 resulted in 4.1 gigatons of greenhouse gas emissions — more than four times the emissions from all commercial aviation in 2023 [1].

Source: Copernicus CAMS / WRI

Data as of May 1, 2026CSV

Carbon Math: Are Fires Erasing Conservation Gains?

The 2024 fire season alone produced 3.1 gigatons of greenhouse gas emissions from tropical forests — an increase largely attributed to fires intensified by El Niño conditions [6]. To put this in context, the cumulative carbon sequestration benefit from reduced deforestation since 2010 — driven by policies like REDD+ payments and moratoriums — is estimated in the range of 4-8 gigatons of CO₂ equivalent over the full decade, depending on methodology and baseline assumptions.

This means a single bad fire year can eliminate a significant fraction — potentially 40-75% — of the carbon gains painstakingly accumulated through policy interventions over ten years.

Canadian wildfires in 2023 added another 640 million metric tons of carbon to the atmosphere, with warm weather and drought burning 5% of the country's total forest area [7]. Scientists at NASA's Jet Propulsion Laboratory noted that "it remains to be seen whether Canadian forests will continue to absorb carbon at a rapid rate or whether increasing fire activity could offset some of the uptake" [7].

Research published in Science in 2024 documented a global rise in forest fire emissions linked to climate change, particularly in extratropical regions where boreal forests are burning at rates not seen in the historical record [8]. The Amazon may have already crossed a threshold: recent studies suggest fire-driven degradation has tipped portions of the rainforest from a net carbon sink to a net carbon source [5].

El Niño Projections: The Threat Ahead

The frequency and intensity of El Niño events — the primary climate driver of tropical drought and fire susceptibility — are projected to increase substantially. Climate models find a consistent pattern for the frequency of intense El Niño events to double in the 21st century, with extreme events occurring roughly once every 10 years instead of once every 20 [9].

Research cited by the NOAA Pacific Marine Environmental Laboratory projects that nearly 50% of all El Niño events will qualify as "extreme" by 2050, defined by sea surface temperatures rising 3.6°F or more above the long-term average in the equatorial Pacific [9]. A separate study found that El Niño-associated rainfall disruptions will increase by 2040 "regardless of any emissions mitigation efforts taken in upcoming decades" [10].

If fire-driven forest loss in El Niño years averages even half of the 2024 level (3.35 million hectares), and such events occur every 5-7 years as projected, cumulative wildfire emissions would exceed total carbon gains from reduced deforestation within approximately two to three El Niño cycles — potentially by the mid-2030s.

Who Pays the Price: Human Costs of Forest Fires

About 1.3 million hectares of forest on Indigenous and community lands burned in the El Niño-linked fire seasons of 2016 and 2024 alone [11]. Indigenous and local communities who depend on these forests for food, water, medicine, and cultural practice bear disproportionate harm.

The 2024 Amazon drought caused food scarcity in Indigenous communities and a sharp increase in waterborne illnesses, particularly among children, due to collapsed water access [11]. The 387 families of the Tikuna tribe in Brazil's Fidadelfia region faced effective confinement when rivers — their primary transportation routes — dried to impassable levels [11].

Globally, an estimated 1.6 billion people depend on forests for livelihoods, including approximately 300-350 million people living within or adjacent to dense forests, of whom roughly 60 million are Indigenous [12]. Smallholder farmers in Bolivia, where fire-driven loss surged 200% in 2024, face both the destruction of their productive land and the loss of surrounding ecosystem services — water regulation, pollination, and soil stability — that sustain their crops [1].

The 2024 South American fire season also produced severe health impacts for urban populations. Smoke from Amazon fires blanketed cities across Brazil, Bolivia, and Peru, with particulate matter concentrations exceeding WHO guidelines by factors of 5-10 in affected areas [13].

Policy Mechanisms: What Works and What Remains Unproven

Brazil's Enforcement Model

Brazil's recent success is the strongest evidence that policy interventions can reduce deforestation. The combination of satellite monitoring (DETER system), rapid enforcement by IBAMA, and supply-chain pressure from commodity buyers has produced measurable results [2]. However, critics note that Brazil's non-fire deforestation was already declining before the current administration took office, and that economic factors — including lower commodity prices — contributed to reduced agricultural expansion pressure.

EU Deforestation Regulation

The EU Deforestation Regulation (EUDR), designed to prohibit imports of commodities linked to deforestation, has faced repeated delays. Originally due to take effect in December 2024, the regulation was postponed to December 2025 for large operators and June 2027 for small businesses [14]. EU studies estimated the delay could result in an additional loss of 2,300 square kilometers of forest globally [14].

The regulation covers six commodities — beef, palm oil, soy, cocoa, coffee, and rubber — which together account for 58% of estimated forest loss tied to EU imports [14]. However, proposed amendments creating a "no-risk" country category could exempt certain nations from key due diligence requirements, potentially undermining the regulation's effectiveness [14].

Payments for Ecosystem Services

Indonesia's REDD+ payments from Norway ($56 million for verified 2016-2017 reductions) and the Green Climate Fund ($103.8 million for 2014-2016 reductions) demonstrate the payment model's viability but also its limitations [4]. A PNAS study found that treated dryland forest inside moratorium areas retained "at most, an average of 0.65% higher forest cover compared to untreated dryland forest outside the moratorium" — a statistically significant but practically modest effect [4].

Peer-Reviewed Evidence of Causation

The peer-reviewed literature remains cautious about attributing causation. A 2025 study in ScienceDirect examined whether international climate finance contributes to adoption of zero-deforestation policies in Brazil and Indonesia, finding correlations but noting that separating policy effects from economic cycles, commodity price fluctuations, and weather patterns remains methodologically challenging [4].

The Funding Gap: Pledges vs. Disbursements

At COP26 in Glasgow, wealthy nations committed $12 billion in forest-related climate finance between 2021 and 2025. Through 2022, $5.7 billion — 47% of the total — had been directed toward forest programs in developing countries [15]. Donor countries reported providing $2.6 billion in the first year alone [15].

However, the Climate Policy Initiative's 2025 analysis found that the land sector has "the largest climate finance gap" of any sector and needs a 30-fold increase in investment to meet its mitigation potential [16]. A $2.9 billion gap exists toward meeting the $10 billion tenure rights target if current trends continue [16].

At COP30 in Belém, Brazil launched the Tropical Forests Forever Facility (TFFF), originally proposed at COP28 as a $250 billion fund to reward tropical forest countries maintaining or reducing deforestation [16]. Thirty-four governments signed on, but the gap between announcement and disbursement remains the persistent pattern of international forest finance.

The Equity Argument: Who Bears the Cost of Conservation?

The mainstream narrative treats slowing deforestation as unambiguous progress. But scholars and advocates from tropical nations raise a structural objection: wealthy countries that cleared their own forests during industrialization are now effectively requiring developing nations to forgo the same economic pathway — without adequate compensation.

Research published in Anthropocene Magazine in 2025 found that "conserving land in wealthy countries may be making things worse somewhere else" — that by fencing off productive land domestically, rich nations increase demand for imports from more ecologically sensitive regions [17]. Since 1961, crops grown for export have expanded at twice the rate of crops grown for domestic consumption, suggesting that consumer demand in wealthy nations continues to drive tropical deforestation through supply chains [17].

A study in Environmental Research Letters documented that emissions leakage — where deforestation-linked production shifts to jurisdictions without import restrictions — could undermine policies like the EUDR [18]. The paper argued that "additional or alternative land-use policies and/or financing are needed" to avoid "exacerbating global inequities between countries" [18].

This is not merely an academic debate. Bolivia's government has explicitly framed its resistance to international deforestation pressure as a sovereignty issue, arguing that agricultural expansion is necessary for national development and that wealthy nations should pay the full opportunity cost of conservation. The tension between climate goals and development equity remains unresolved in international negotiations.

Methodology Questions: Is the Progress Real?

Global Forest Watch acknowledged in its 2024 methodology explanation that changes in satellite resolution, cloud-cover algorithms, and land-cover classification systems can affect year-to-year comparisons [19]. The shift from Landsat 7 to Landsat 8/9 improved detection of small-scale clearing but also means some "increases" in detected loss reflect better observation rather than actual acceleration.

Additionally, fire-driven loss and agriculture-driven loss have different recovery trajectories. Burned forest may regenerate naturally over 20-50 years; land converted to soy or cattle pasture rarely returns to forest. The 2024 spike in total loss — dominated by fire — may therefore overstate the permanent damage compared to an equivalent area of agricultural conversion.

That said, repeated burning inhibits regeneration. Areas burned multiple times in successive El Niño cycles (2015-16, 2023-24) show substantially reduced capacity for natural recovery, with degraded forests becoming trapped in a grass-fire cycle [2].

What Comes Next

The 2025 data — showing a 36% drop from 2024's record — is encouraging but insufficient. Current deforestation rates remain 70% higher than the level needed to meet the Glasgow Leaders' Declaration goal of halting and reversing forest loss by 2030 [2].

The fundamental tension is between two timescales. Policy interventions — enforcement, regulation, payments — operate on human institutional time. Climate change and El Niño cycles operate on physical time. If extreme El Niño events double in frequency as projected, the windows between fire catastrophes narrow while the policy apparatus struggles to maintain funding and political will.

The next strong El Niño, potentially as early as 2026-2027, will serve as the critical test of whether the conservation gains of the past decade represent durable structural change or a fragile respite between fire seasons.