Deforestation — Drivers, Carbon Impacts & Global Policy

Updated May 2026 Forest loss & degradation Amazon · Congo · Southeast Asia REDD+ · Paris Agreement
Forests cover ~31% of Earth's land surface, store ~861 billion tonnes of carbon, and harbour 80% of terrestrial biodiversity. Since 1990, the world has lost approximately 420 million hectares of forest — an area larger than the European Union. Deforestation is the second-largest source of greenhouse gas emissions after fossil fuels, accounting for roughly 10–15% of global CO₂ emissions. The Amazon alone has lost ~20% of its original extent and is approaching a tipping point beyond which large portions may flip irreversibly to savanna.
~4.1B ha
Total global forest area today (FAO FRA 2020); ~31% of total land area; down from ~5.9B ha in pre-agricultural times
420M ha
Net forest loss 1990–2020 (FAO FRA 2020); equivalent to an area larger than the EU; rate decelerating but still ~4.7M ha/yr net loss 2010–2020
~10–15%
Share of global GHG emissions from deforestation and forest degradation (IPCC AR6 WG3 2022); equivalent to ~5–8 Gt CO₂/yr
15.3B
Trees cut per year globally (Crowther et al. 2015, Nature); net loss of ~10B trees/yr after replanting; world has lost 46% of trees since start of human civilisation
80%
Share of terrestrial biodiversity found in forests (IPBES 2019); deforestation is the #1 driver of terrestrial species loss globally
~$2.5T/yr
Estimated economic value of global forest ecosystem services (World Bank / Ecosystem Services Valuation Database); far exceeds commodity value of timber and farmland

★ The Scale of the Problem — What We Are Losing

Forests are the planet's most complex terrestrial ecosystems. A mature tropical rainforest contains more species per hectare — insects, plants, mammals, fungi, microorganisms — than any other ecosystem on Earth. The Amazon basin alone is home to approximately 10% of all species on the planet, including an estimated 390 billion individual trees across 16,000 species. The Congo Basin forest is the second-largest tropical forest and one of the most important carbon stores in the world, covering 3.7 million km² across six countries.

Between 2001 and 2023, the world lost approximately 147 million hectares of tree cover (University of Maryland / Global Forest Watch), with tropical primary rainforest loss — the most ecologically and climatically significant type — running at roughly 3–4 million hectares per year in recent years. Brazil, the Democratic Republic of Congo, and Indonesia consistently account for the largest absolute losses. While deforestation rates in the Brazilian Amazon fell sharply between 2004 and 2012 in response to policy intervention, rates rebounded significantly under the Bolsonaro government (2019–2022) before declining again under the Lula government from 2023.

Global Tropical Forest Loss by Country (2001–2023)

Source: Hansen et al. / Global Forest Watch (University of Maryland); tree cover loss data 2001–2023; includes all tree cover loss >30% canopy density threshold.

Annual Primary Tropical Forest Loss — Key Years

Source: Global Forest Watch / University of Maryland; primary forest loss = humid tropical primary forest only, the highest-integrity category. INPE (Brazil PRODES system) for Amazon data.

Forest Cover by Region

South America842M ha — Amazon, Atlantic Forest, Cerrado
Europe (incl. Russia)1,015M ha — boreal taiga dominates
North & Central America750M ha — temperate & boreal mixed
Africa636M ha — Congo Basin + savanna forests
Asia-Pacific622M ha — SE Asia highly threatened
Change 2010–2020 (net)−4.7M ha/yr globally (FAO FRA 2020)
Source: FAO Global Forest Resources Assessment 2020.

Types of Forest Loss

DeforestationPermanent conversion to non-forest use; ~50% of gross loss
DegradationSelective logging, fire, edge effects; forest remains but depleted; often invisible in headline stats
Shifting cultivationSmall-scale clearing; some recovery; major in Africa & SE Asia
Commercial agricultureSoy, cattle, palm oil, cocoa — dominant in tropics
Infrastructure / miningRoads, dams, mines — catalyse further loss
Wildfire (climate-driven)Growing share, esp. Amazon, Siberia, North America

Amazon — The Critical Numbers

Original Amazon forest area~420M ha (pre-1500)
Deforested to date~20% (~85M ha); mainly Mato Grosso, Pará, Rondônia
Tipping point threshold (estimated)~25–40% deforestation + climate warming → dieback
Peak deforestation (2004)~27,772 km² in one year (INPE PRODES)
Low point after PPCDAm policy (2012)~4,571 km² — 83% reduction in 8 years
Bolsonaro-era peak (2021)~13,235 km² — back above 2008 levels
2023 (Lula / Amazon Fund restored)~11,568 km² — declining trend resumed
Source: INPE PRODES; Lovejoy & Nobre 2018 (Sci. Adv. — tipping point); Nobre et al. 2016; Staal et al. 2020 (Nat. Comms).
Half the world's forests are already gone — and we've accelerated the clearing: Humans have cut down approximately half of all the forests that existed on Earth when civilisation began. Of the 3.9 trillion trees estimated to exist today (Crowther et al. 2015), roughly 46% of the original count has been lost. But the rate of loss is not slowing proportionally — primary forest loss in the tropics was higher in the 2010s than in the 2000s, driven by agricultural expansion, infrastructure, and increasingly by climate-driven fire. The forests we are losing most rapidly are also the most valuable: mature, primary tropical forests that have existed for thousands of years and harbour species found nowhere else.

★ What Is Driving Forest Loss? Commodity Agriculture Dominates

Research consistently shows that commercial agriculture — cattle ranching, soy cultivation, palm oil, and cocoa — drives the majority of tropical deforestation. A 2017 analysis in Environmental Research Letters found that commercial agriculture drives 71% of tropical forest loss; subsistence farming accounts for a further 24%; with logging, mining, infrastructure, and fire together responsible for the remainder. This concentration of drivers in a small number of commodity supply chains is both the problem and the opportunity: it means that a relatively small number of multinational corporations, governments, and financial institutions could, in theory, dramatically reduce deforestation rates by reforming supply chain standards and investment policies.

Proximate Drivers of Tropical Deforestation

Source: Kissinger et al. 2012 (Drivers of Deforestation and Forest Degradation — PROFOR); Curtis et al. 2018 (Science — mapping drivers globally); Geist & Lambin 2002 (BioScience).

Deforestation Frontiers (WWF / Global Forest Watch)

FrontierPrimary DriverAnnual LossStatus
Brazilian AmazonCattle, soy, infrastructure~1.0–1.5M ha/yrCritical
Congo BasinSmallholder farming, charcoal, logging~500–700K ha/yrCritical
Borneo / IndonesiaPalm oil, pulp & paper, mining~300–500K ha/yrCritical
Cerrado (Brazil)Soy, cattle — legally allowed~800K ha/yrCritical
Chaco (S. America)Cattle, soy expansion~400–600K ha/yrCritical
Papua New GuineaPalm oil, logging concessions~150K ha/yrHigh
West Africa (Ghana, Côte d'Ivoire)Cocoa, mining, agriculture~200K ha/yrMedium-High
Atlantic Forest (Brazil)Sugarcane, cattle — mostly residual clearing~50–80K ha/yrHeavily fragmented
Mekong Region (Laos, Myanmar)Rubber, cash crops, dams~200K ha/yrHigh
Source: WWF Deforestation Fronts 2021; Global Forest Watch 2023; Butler 2022 (Mongabay); Curtis et al. 2018 (Science).

Cattle Ranching — The Dominant Driver in Latin America

Cattle ranching accounts for approximately 80% of Amazon deforestation. Brazil has the world's largest commercial cattle herd (~215 million head) and is the world's largest beef exporter. The economics of Amazonian cattle ranching historically depended on land speculation — clearing forest to establish legal land title and then selling the land. Soy monoculture frequently follows cattle on the new agricultural frontier.

% Amazon deforestation linked to cattle~80% direct; further % indirect via soy feed
Brazil cattle herd~215M head (2023 IBGE)
Beef exports from deforested Amazon land~30–40% of Brazilian beef exports (Trase / Global Canopy)
EU Deforestation Regulation (EUDR)Applies to beef, soy, palm oil, cocoa, coffee, wood; due-diligence proof required from Dec 2024 (now delayed to late 2025)

Soy — Feeding the World on Deforested Land

Brazil is the world's largest soy producer (~160 million tonnes in 2023), having expanded production dramatically into the Cerrado and Amazon over the past 30 years. While the Amazon Soy Moratorium (2006) largely ended direct soy-for-Amazon deforestation, the Cerrado — Brazil's other threatened biome — remains largely unprotected. The Cerrado has lost ~50% of its original vegetation and continues to lose ~800,000 hectares per year, with almost no regulatory protection comparable to the Amazon.

Brazil soy production (2023)~160M tonnes — world's largest producer
Amazon Soy Moratorium (2006)Signatories agreed not to buy soy from newly deforested Amazon; largely effective within the Amazon biome
Cerrado coverage by Amazon MoratoriumZero — leakage of deforestation to Cerrado well-documented
Cerrado remaining native vegetation~50% remains — ongoing loss ~800K ha/yr (MapBiomas)

Palm Oil — The High-Yield Problem

Palm oil is the world's most widely used vegetable oil, found in roughly half of all packaged products in supermarkets. Oil palm is extraordinarily productive — 4–5 tonnes of oil per hectare versus 0.5–0.7 t/ha for soy — making it difficult to replace without converting even more land. The problem is that it is grown predominantly in tropical countries where it competes directly with primary rainforest, often in Malaysia and Indonesia, and increasingly in West Africa and Colombia.

Indonesia palm oil deforestation (2001–2019)~3.3M ha of primary forest and peatland (Gaveau et al.)
RSPO certified palm oil (2023)~21% of global production; certification quality contested
No Deforestation, No Peat, No Exploitation (NDPE)~85% of global palm oil buyers have some form of NDPE commitment; enforcement varies widely

★ Deforestation as a Climate Driver — The Carbon and Rainfall Double Shock

Forests affect climate in two distinct but equally important ways: as carbon stores, and as drivers of regional hydrology and temperature. The carbon story is well understood — forests contain approximately 861 Gt of carbon (above- and below-ground biomass plus soil), and when they are burned or left to decompose after clearing, this carbon enters the atmosphere as CO₂ and methane. But the hydrological story is less widely appreciated and potentially just as consequential: large forests generate their own rainfall through transpiration and the "biotic pump" mechanism, and their loss disrupts regional precipitation patterns far beyond the forest boundary itself.

The Amazon rainforest transpires an estimated 20 billion tonnes of water per day — more water than the Amazon River carries to the ocean. This moisture rises, condenses, and falls as rain hundreds or thousands of kilometres away — the so-called "flying rivers" that supply rainfall to southern Brazil's agricultural heartland (the Cerrado soy belt and the coffee regions). Modelling studies suggest that deforestation of 25–40% of the Amazon, in combination with continued global warming, could trigger a self-reinforcing dieback of the eastern Amazon — irreversibly converting it to savanna over decades, with catastrophic consequences for the global carbon cycle and South American agriculture.

Forest Carbon Stocks — Where the Carbon Is

Source: Harris et al. 2021 (Nature Climate Change — Global Carbon in Tropical Forests); Pan et al. 2011 (Science — global forest carbon); Baccini et al. 2012 (Nature Geosci.); FAO 2020.

Deforestation Carbon Emissions vs. Other Sources

Source: IPCC AR6 WG3 2022 (Chapter 7 — Agriculture, Forestry, Land Use); Global Carbon Project 2023; Friedlingstein et al. 2022 (ESSD); Gasser et al. 2020.

Amazon Tipping Point Science

Tipping point threshold (Lovejoy/Nobre 2018)20–25% deforestation; currently at ~20%
Trigger mechanismLoss of moisture recycling → reduced rainfall → tree die-off → more fire → more loss (self-reinforcing)
Fonseca et al. 2023 (Nature, 200+ authors)Amazon is "closer than ever" to tipping point; interaction of deforestation + warming + drought
Carbon release if Amazon dieback (est.)~90–140 Gt CO₂ — equivalent to 7–10 years of global fossil fuel emissions
Flying rivers (moisture export)~20B tonnes water/day transpired; Rondônia deforestation already reducing rainfall in SE Brazil
Source: Lovejoy & Nobre 2018 (Sci. Adv.); Staal et al. 2020 (Nat. Comms.); Zemp et al. 2017 (Nat. Comms.); Gatti et al. 2021 (Nature).

Forest Carbon Flux — Sink or Source?

Intact forests are a net carbon sink, absorbing more CO₂ than they release. But degraded and fragmented forests can become net carbon sources — even without full deforestation. A landmark 2021 study in Nature found that the eastern Amazon has already become a net carbon source — releasing more CO₂ than it absorbs — due to the combination of deforestation, edge effects, and climate-driven drought and fire stress.

Global forest carbon sink (intact forests)~2.6 Gt C/yr absorbed (Pan et al. 2011, Science)
Gross deforestation emissions~4.2–5.5 Gt CO₂/yr (Harris et al. 2021)
Net land-use change flux~1.6 Gt C/yr net source (after forest regrowth credit)
Eastern Amazon (Gatti et al. 2021, Nature)Net carbon source for the first time in records; SE Amazon worst-affected
Source: Gatti et al. 2021 (Nature); Harris et al. 2021 (Nat. Clim. Change); Pan et al. 2011 (Science); Global Carbon Project 2023.

Forest Degradation — The Hidden Loss

Selective logging, edge effects, and fire damage forests without fully clearing them — creating a "degraded" forest that may still register as forested in satellite data but has lost much of its carbon, biodiversity, and ecological function. Degradation is estimated to emit as much carbon as deforestation itself but is far less visible in policy frameworks.

Degradation emissions (est.)~2–3 Gt CO₂/yr — comparable to deforestation
Edge effects~50% of tropical forest within 1 km of an edge is significantly degraded (Brinck et al.)
Logging road network expansion~250,000 km/yr of new roads in tropical forests (Laurance et al. 2014); every road opens new frontier
REDD+ inclusion of degradationREDD+ (under Paris Art. 5) covers "+" = conservation and enhancement, but degradation measurement remains technically challenging
The 1.5°C target is essentially unachievable if tropical forests continue to be lost: The IPCC AR6 Synthesis Report (2023) states explicitly that achieving and sustaining net-zero emissions requires not just eliminating fossil fuel emissions but also halting and reversing forest loss — with forests providing net carbon removal of 1–2 Gt CO₂/yr by 2050 in most 1.5°C scenarios. Currently, forests are a net source of emissions when degradation and fire are included. Reversing this requires both stopping deforestation and allowing secondary forests to recover — a process that takes 20–80 years for mature ecosystem function to return.

★ Forests as the Cradle of Biodiversity — What We Lose When We Clear

Forests harbour an estimated 80% of all terrestrial species on Earth. Tropical forests, covering just 7% of the Earth's surface, contain more than half of all species. The Amazon basin alone holds approximately 10% of all species — an estimated 40,000 plant species, 1,300 bird species, 3,000 freshwater fish species, 427 mammal species, and millions of as-yet undescribed insect and microbial species. Many of these species have ranges measured in single valleys or mountain slopes — meaning local deforestation can cause immediate, irreversible extinction.

The IPBES 2019 Global Assessment estimated that approximately 1 million species are currently threatened with extinction — the highest number in human history — with habitat loss (primarily deforestation) identified as the primary driver of terrestrial biodiversity loss. Species Area Relationship theory predicts that a 90% reduction in habitat area leads to roughly a 50% reduction in the number of species that can persist — and many tropical forest fragments are already well below the threshold size needed to maintain viable populations of large mammals, top predators, or wide-ranging species.

Forest Biodiversity — Key Metrics

Share of terrestrial species in forests~80% (IPBES 2019)
Amazon plant species (est.)~40,000 — 16,000 tree species alone (ter Steege et al. 2013)
Amazon bird species~1,300 — 15% of all bird species on Earth
Amazon freshwater fish species~3,000 — more than in all of the world's oceans combined
Species threatened primarily by deforestation54% of all IUCN Red List threatened species (Maxwell et al. 2016)
Atlantic Forest remnant~12–16% remains of original extent; yet still harbours 2,000 bird spp. and unique endemic fauna
Extinction debt (future committed extinctions)Even if deforestation stopped today, hundreds of species in existing fragments are committed to extinction over coming decades
Source: IPBES 2019 Global Assessment; Maxwell et al. 2016 (Nature — threat synergies); ter Steege et al. 2013 (Science); Myers et al. 2000 (Nature — biodiversity hotspots); Dirzo et al. 2014 (Science — defaunation).

Biodiversity Hotspots Under Forest Pressure

Hotspot% Endemic Species% Original Habitat RemainingPrimary Threat
Atlantic Forest (Brazil)8,000 endemic plant spp.~12–16%Agriculture, urban expansion
Madagascar & Indian Ocean Islands>90% endemic~10%Slash & burn, charcoal
Sundaland (Borneo, Sumatra)~25,000 endemic plant spp.~7%Palm oil, pulp & paper
Philippines>5,000 endemic plants~7%Agriculture, mining
Western Ghats & Sri Lanka~5,900 endemic plants~6%Tea, rubber, agriculture
Indo-Burma>7,000 endemic plants~5%Agriculture, logging, dams
New Caledonia~2,500 endemic plants~5%Nickel mining
Mountains of SW China>3,500 endemic plants~8%Agriculture, development
Source: Myers et al. 2000 (Nature — original hotspots paper); Mittermeier et al. 2004 (Hotspots Revisited); IUCN Red List; Conservation International Hotspots Database.
Defaunation — losing animals even before losing trees: Even before a forest is fully cleared, hunting and fragmentation drive the loss of animals — particularly large vertebrates (tapirs, jaguars, large birds, primates) that perform critical ecosystem functions: seed dispersal, pollination, predator regulation, and nutrient cycling. A forest stripped of its large animals becomes "empty forest" — the trees may stand but the ecological processes that sustain them over centuries are broken. The loss of large seed dispersers (e.g. elephants, tapirs, large hornbills) means many large-seeded tree species, which are often the most carbon-dense, fail to regenerate after disturbance.

★ The Economics of Deforestation — Why Forests Keep Being Cleared

From a purely local economic perspective, clearing forest often makes financial sense for landholders, companies, and governments who capture the commodity value of the cleared land while externalising the global climate and biodiversity costs. A hectare of Amazon rainforest cleared for cattle pasture may be worth $150–400 in beef production value annually — while storing 150–250 tonnes of CO₂. At prevailing voluntary carbon market prices of $10–50/t, the carbon value alone could exceed the agricultural value. At Paris-aligned carbon prices of $100–200/t, deforestation would be economically irrational. The challenge is that landholders and governments do not receive carbon payments unless enrolled in formal REDD+ programmes — which cover a tiny fraction of tropical forests.

REDD+ — Reducing Emissions from Deforestation and Degradation

REDD+ (Reducing Emissions from Deforestation and Forest Degradation in Developing Countries, plus conservation, sustainable management and enhancement of forest carbon stocks) is the primary international mechanism for paying developing countries to protect their forests. Established under the UNFCCC in 2007–2010, REDD+ has mobilised billions in climate finance but faces fundamental measurement, reporting, and verification (MRV) challenges, permanence risk, and leakage concerns.

REDD+ finance committed (2008–2022)~$12B in pledges (Climate Finance Network); delivered ~$5–7B
REDD+ under Paris Agreement (Article 5)Formally endorsed; countries can generate ITMO credits for REDD+ results; rules under Art. 6.2
Amazon Fund (Brazil)~$1.7B raised (Norway, Germany, others); suspended 2019–2022; reactivated 2023 under Lula; $1.6B new pledges in 2023
COP26 Glasgow Declaration (2021)141 countries pledging to end deforestation by 2030; $19B public & private finance; progress lagging
Key REDD+ challengeLeakage: protecting one area just displaces deforestation elsewhere unless national-scale monitoring exists
Source: Climate Finance Network 2022; GCF REDD+ portfolio; Norman & Nakhooda 2015; Forest Declaration Assessment 2023.

EU Deforestation Regulation (EUDR)

The EU Deforestation Regulation (Regulation 2023/1115) — adopted June 2023 — requires companies placing products on the EU market to prove they were not produced on land deforested after 31 December 2020. The regulated commodities are cattle, cocoa, coffee, palm oil, soy, wood, rubber, and derived products. Companies must provide geolocation data for all supply chain plots, creating potentially the world's most powerful trade-linked deforestation instrument. Original enforcement deadline: December 2024. Delayed to late 2025 after significant pushback from Brazil, Malaysia, Indonesia, and some EU members.

Commodities coveredCattle, cocoa, coffee, palm oil, soy, wood, rubber
Deforestation-free cut-off date31 December 2020
Large company compliance deadline30 December 2025 (delayed from Dec 2024)
SME deadline30 June 2026
Penalty for non-complianceUp to 4% of EU annual turnover; product seizure
Source: EU Regulation 2023/1115 (Official Journal of the EU); European Commission DG ENV; Forest Declaration Assessment 2024.

Brazil's PPCDAm — A Proven Model

Brazil's Action Plan for Prevention and Control of Deforestation in the Legal Amazon (PPCDAm) reduced Amazon deforestation by ~83% between 2004 and 2012 — one of the most effective environmental interventions in history. It combined satellite monitoring (INPE/PRODES), rural credit conditionality, increased enforcement, expanded protected areas, and indigenous territory demarcation.

Peak deforestation (2004)27,772 km²/yr
Trough (2012)4,571 km²/yr (–83%)
Emissions avoided (2004–2012)~3.2 Gt CO₂ (Nepstad et al. 2014, Science)
Key lessonPolicy works — but requires political commitment; reversed rapidly under Bolsonaro

Private Sector Commitments — Gap vs. Reality

Consumer Goods Forum (CGF) pledge (2010)Net-zero deforestation by 2020; largely unmet
New York Declaration on Forests (2014)Halt deforestation by 2030; private sector signatory rate behind schedule (Forest Declaration Assessment)
Forest 500 (Global Canopy 2023)68% of most influential companies still have no deforestation commitment
Supply chain traceability (soy, palm)~80% of global palm oil buyers have NDPE policy; traceability to plot-level much lower
Carbon market forestry credits (VCM)REDD+ credits face severe integrity challenges (Guardian/CarbonPlan 2023 investigation); major buyers withdrawing

Forest Finance Gap

Annual public climate finance for forests (2021)~$2.2B (Climate Policy Initiative 2023) — a fraction of need
Required finance to stop deforestation (est.)~$130–200B/yr by 2030 to meet Paris pledges (Taskforce on Nature Markets)
Gap~100×; forest finance is the single largest gap in the climate finance landscape
Carbon value of tropical forests at $150/t CO₂~$100T+ total; ~$5–8T annual flow if forests were valued as carbon assets
Debt-for-nature swapsGrowing: Ecuador, Belize, Gabon; restructure sovereign debt with conservation covenants