GREENHOUSE GASES FROM DEFORESTATION IN BRAZILIAN AMAZONIA: NET COMMITTED EMISSIONS

Deforestation in Brazilian Amazonia is a significant source of greenhouse gases today and, with almost 90% of the originally forested area still uncleared, is a very large potential source of future emissions. The 1990 rate of loss of forest (13.8 × 10³ km²/year) and cerrado savanna (approximately 5 × 10³ km²/year) was responsible for releasing approximately 261 × 10⁶ metric tons of carbon (10⁶ t C) in the form of CO2, or 274–285 × 10⁶ t of CO2-equivalent C considering IPCC 1994 global warming potentials for trace gases over a 100-year horizon. These calculations consider conversion to a landscape of agriculture, productive pasture, degraded pasture, secondary forest, and regenerated forest in the proportions corresponding to the equilibrium condition implied by current land-use patterns. Emissions are expressed as net committed emissions, or the gases released over a period of years as the carbon stock in each hectare deforested approaches a new equilibrium in the landscape that replaces the original forest. For low and high trace gas scenarios, respectively, 1990 clearing produced net committed emissions (in 10⁶ t of gas) of 957–958 for CO2, 1.10–1.42 for CH₄, 28–35 for CO, 0.06–0.16 for N₂O, 0.74–0.74 for NO_x and 0.58–1.16 for non-methane hydrocarbons.     

CARBON FLOW IN INDIAN FORESTS

The present study estimates the net emission of carbon from the forest sector in India. For the reference year (1986), the gross emission from deforestation in that year, plus committed emissions from deforestation in the preceding years, is estimated to be 64 × 10⁶ t of C. The carbon sequestration (or net woody biomass accumulation in trees for long-term storage) from the area brought under tree plantations and the existing forest area under forest succession is estimated to offset the gross carbon emission in India, leading to no net emissions of carbon from the forest sector. Medium-term projections for India (for the year 2011) show that under a business as usual scenario at current rates of afforestation, projected carbon emissions would continue to be balanced by sequestration.     

Landscape-Scale Analysis of Interactions between Insect Defoliation and Forest Fire in Central Canada

Analysis of Ontario's historical records from 1941–1996 showed that spruce budworm, Choristoneura fumiferana (Clem.) caused whole tree mortality within 389×10³ km². This amounted to 9.2% of the annually cumulative area with moderate-severe defoliation. Large (>2 km²) fires were reported in approximately 65,000 km² and overlapped only 2.8% of the area of reported spruce budworm (SBW) caused tree mortality. Within the 417×10³ km² defoliated by SBW at least once in 1941–1996, the maximum total area recorded as defoliated in any year was over 20 times the maximal area burnt. In the 19,950 km² experiencing both wildfire and SBW defoliation, analysis of the spectra of time lags between the two disturbance types indicated that fires occurred 3–9 years after a SBW outbreak disproportionately often. This window of opportunity for wildfire varies geographically: it starts later after SBW outbreak and lasts longer in western than in eastern Ontario. In addition, 7.5% of the areas containing SBW killed trees were burnt in western compared to 4.8% in eastern Ontario. These geographical differences may result at least partly from slower decomposition of dead fuels in the drier climates of the western SBW belt compared to the eastern SBW belt. The implications for climatic change are discussed.     

Three-dimensional modeling of synthetic cold fronts interacting with northern Alpine foehn

Summary The effect of the Alpine orography on prototype cold fronts approaching from the west is investigated by three-dimensional numerical model simulations. The numerical experiments cover a range of parameter constellations which govern the prefrontal environment of the front. Especially, the appearance and intensity of prefrontal northern Alpine foehn varies from case to case.The behaviour of a cold front north of the Alps depends much on the prefrontal condition it encounters. It is found that prefrontal foehn can either accelerate or retard the approaching front.An important feature is the pressure depression along the northern Alpine rim that results from the southerly foehn flow. In cases where this depression compensates the eastward directed pressure gradient associated with the largescale flow, the front tends to accelerate and the foehn breaks down as soon as the front passes. In contrast, the foehn prevents the front from a rapid eastward propagation if it is connected with a strong southerly wind component.No-foehn experiments are performed for comparison, where either the mountains are removed, or the static stability is set to neutral. Also shown are effects of different crossfrontal temperature contrasts.List of Symbols c _F propagation speed of a front - x, y horizontal grid spacing (cartesian system) - , horizontal grid spacing (geographic system) - t time step - z vertical grid spacing (cartesian system) - cross-frontal potential temperature difference - _i potential temperature step at an inversion - E turbulent kinetic energy - f Coriolis parameter - FGP frontogenesis parameter (see section 2.2) - g gravity acceleration (g=9.81 m s^–2) - vertical gradient of potential temperature - h terrain elevation (above MSL) - h _i height of an inversion (h _i =1000 m MSL) - H height of model lid (H=9000 m MSL) - K _M exchange coefficient of momentum - K _H exchange coefficient of heat and moisture - longitude - N Brunt-Väisäla-frequency - p pressure - Exner function (=T/) - latitude - q _v specific humidity - R _d gas constant of dry air (R _d =287.06 J kg^–1 K^–1) - density of dry air - t time - T temperature - potential temperature - TFP thermal front parameter (see section 2.2) - u, v, w cartesian wind components - u _g ,v _g geostrophic wind components - horizontal wind vector - x, y, z cartesian coordinates Abbreviations GND (above) ground level - MSL (above) mean sea level - UTC universal time coordinated With 20 Figures     

Centralization in the global avoided deforestation collaboration network

Reducing Emissions from Deforestation and Forest Degradation projects currently cover an area approximately twice the size of Germany and challenge traditional concepts of centralization and decentralization in studies of environmental governance. Emerging from the interactions of a complex network of actors, Reducing Emissions from Deforestation and Forest Degradation demonstrates that transnational governance networks of organizations can become spatially centralized. Using a historical analysis of the development of Reducing Emissions from Deforestation and Forest Degradation, we argue that the evolution of Reducing Emissions from Deforestation and Forest Degradation policy has been directed primarily from donor countries, especially in North America and Europe. Adopting a social network analysis approach, we present findings from a new dataset of collaboration on 276 Reducing Emissions from Deforestation and Degradation, avoided deforestation, and sustainable forest management projects that began some on-the-ground operations between 1989 and June 2012, finding that organizations in donor countries have from the beginning been the central actors in the Reducing Emissions from Deforestation and Forest Degradation network. We conclude that Reducing Emissions from Deforestation and Forest Degradation exhibits spatial centralization within transnational governance architectures despite institutional fragmentation, raising important normative questions about participation in transnational forest governance.     

Public expectation as an element of human perception of climate change

Human expectations regarding weather and climate sometimes lead to perceptions of climate change which are not supported by observational evidence. This paper analyses two very characteristic complaints about current climate in Switzerland, i.e., the lack of snow in winter and the lack of sunshine in summer, through a statistical investigation of climatological data. As one major problem of public perception of climate in mid-latitude regions is linked to the strong variability of the climatic parameters, the paper suggests means of presenting climatic data which include a measure of this variability. Such presentations would help overcome the common confusion between the terms weather and climate, and stress the fact that short-term extreme events are not necessarily indicative of a long-term shift in climate.     

Deforestation in the Brazilian Amazon could be halved by scaling up the implementation of zero-deforestation cattle commitments

Deforestation for agriculture is a key threat to global carbon stocks, biodiversity, and indigenous ways of life. In the absence of strong territorial governance, zero-deforestation commitments (ZDCs), corporate policies to decouple food production from deforestation, remain a central tool to combat this issue. Yet evidence on their effectiveness remains mixed and the mechanisms limiting effectiveness are poorly understood. To advance understanding of ZDCs’ potential at reducing deforestation, we developed the first spatially explicit estimates of farmers’ exposure to ZDC companies in the Brazilian Amazon cattle sector. Exposure was measured by determining the market share of ZDC firms from the first full year of ZDC adoption in 2010 until 2018. Our analysis evaluated how variation in this exposure influenced deforestation. We found the G4 Agreement, the most widespread and strongly implemented cattle ZDC, reduced cattle-driven deforestation by 7,000 ± 4,000 km² (15 ± 8%) between 2010 and 2018. Additionally, had all firms adopted and implemented an effective ZDC, cattle-driven deforestation could have dropped by 24,000 ± 13,000 km² (51 ± 28%). These results for the world’s principal deforestation hotspot suggests supply chain policies can substantially reduce deforestation. However, their effectiveness is contingent on widespread adoption and rigorous implementation, both of which are currently insufficient to prevent large scale deforestation. Increased adoption and implementation could be incentivized through greater pressure from the Brazilian government and import countries.     

Modelling the effects of albedo change associated with tropical deforestation

The effects in climate models of changes in albedo and soil moisture which are expected to follow from tropical deforestation are reviewed. The major results from experiments with tropical forest regions (a) forested (Experiment F) and (b) grass-covered (Experiment G) are compared over South America and Central Africa. Then, Experiment F is compared with Experiment 2 which has constant snowfree albedo of 0.2, close to that of grassland; in both South America and Zaire regions, the albedo changes for (G-F) and (2-F) are similar and, as Experiment 2 is an eight year run, assessments of the statistical significance can be made.The results for December to February and March to May show a decrease of rainfall over the Amazon region and also to the southeast outside the area which has changed albedos in (G-F). These decreases are common to both (G-F) and (2-F) and are statistically significant in the comparison with the eight year record for Experiment 2. Deforestation in the Zaire regions is associated with similar changes, which, though less extensive, are generally larger in relation to the albedo changes. In June to August, changes are small south of the equator because the rainbelt has moved north; in South America larger, significant decreases are obtained north of the equator.The drying with increased albedo is due to decreases in both evaporation and moisture convergence. The decreases in evaporation result both from a reduction in absorbed energy at the surface and an increase in surface resistance to evaporation due to greater soil moisture deficits arising from earlier decreases in rainfall.     

Monoterpene concentrations in and above a forest of scots pine

Diurnal and vertical ambient air measurements of the monoterpenes have been made in and above a Scots pine (Pinus sylvestris) forest of central Sweden, within the boreal northern coniferous biome. Sampling was done with Tenax TA, and analysis by GC and ion trap detection. Daytime mixing ratios were on the order of tenths of a ppbv from the forest floor to the top of the forest, and a factor of 2 or 3 lower above the forest. Mixing ratios at night were at the ppbv level, highest near the forest floor and the crown, and decreased with height above the forest. The highest total concentration observed was 8 ppbv inside the forest at 3 am (GMT). The average terpene composition was 3-carene 32%, -pinene 29%, limonene 18%, -pinene 10%, -phellandrene 7%, camphene 5%, and sabinene at less than 2%. The 3-carene/-pinene ratio varied with wind direction and speed, relative humidity, and wet/dry vegetation, but not with ozone or NO₂ concentration, solar radiation, or temperature. Variations in the observed terpene composition at the sampling site are mainly caused by the influence of other vegetation in the vicinity of the site. It would seem that wet Scots pine emits more 3-carene relative to -pinene than does dry pine.     

Community forest management in Indonesia: Avoided deforestation in the context of anthropogenic and climate complexities

Community forest management has been identified as a win-win option for reducing deforestation while improving the welfare of rural communities in developing countries. Despite considerable investment in community forestry globally, systematic evaluations of the impact of these policies at appropriate scales are lacking. We assessed the extent to which deforestation has been avoided as a result of the Indonesian government’s community forestry scheme, Hutan Desa (Village Forest). We used annual data on deforestation rates between 2012 and 2016 from two rapidly developing islands: Sumatra and Kalimantan. The total area of Hutan Desa increased from 750 km² in 2012 to 2500 km² in 2016. We applied a spatial matching approach to account for biophysical variables affecting deforestation and Hutan Desa selection criteria. Performance was assessed relative to a counterfactual likelihood of deforestation in the absence of Hutan Desa tenure. We found that Hutan Desa management has successfully achieved avoided deforestation overall, but performance has been increasingly variable through time. Hutan Desa performance was influenced by anthropogenic and climatic factors, as well as land use history. Hutan Desa allocated on watershed protection forest or limited production forest typically led to a less avoided deforestation regardless of location. Conversely, Hutan Desa granted on permanent or convertible production forest had variable performance across different years and locations. The amount of rainfall during the dry season in any given year was an important climatic factor influencing performance. Extremely dry conditions during drought years pose additional challenges to Hutan Desa management, particularly on peatland, due to increased vulnerability to fire outbreaks. This study demonstrates how the performance of Hutan Desa in avoiding deforestation is fundamentally affected by biophysical and anthropogenic circumstances over time and space. Our study improves understanding on where and when the policy is most effective with respect to deforestation, and helps identify opportunities to improve policy implementation. This provides an important first step towards evaluating the overall effectiveness of this policy in achieving both social and environmental goals.     

The Value Of The β Coefficient In The Relaxed Eddy Accumulation Method In Terms Of Fourth-Order Moments

The commonly measured value of in the relaxed eddy accumulationmethod of about 0.56is shown to arise from the non-Gaussiannature of turbulence. Fourth-orderGram–Charlier functions forthe two-dimensional probability distributionsof variation in the horizontal component of wind velocityand concentrations of water vapour, carbondioxide and methane with respect to thevertical component of wind velocity are used to examinethe value of .An analytical solution for ispresented in terms of fourth-order moments.Under mean conditions, this solution givesa value for of0.557. Variation of is shown to be controlledprimarily by the ratio of the mean ofc'w3 (where c'is relevant to the entity of interest andw' is vertical component of windvelocity) to the correlationcoefficient between the entity concentrationand vertical component of wind velocity.     

Trajectories of deforestation,coffee expansion and displacement of shifting cultivation in the Central Highlands of Vietnam

Production of commodities for global markets is an increasingly important factor of tropical deforestation, taking over smallholders subsistence farming. Measures to reduce deforestation and convert shifting cultivation systems towards permanent crops have recently been strengthened in several countries. But these changes have variable environmental and social impacts, including on ethnic minorities. In Vietnam, although a forest transition – i.e. shift from shrinking to expanding forest cover – occurred at the national scale, deforestation fronts and agricultural colonization for commodity crops – a.o. coffee – still dominated the Central Highlands plateaus. Previous studies suggested that the dominant land use changes in that region were on the one hand the acquisition and conversion of agricultural lands to perennial crops for external markets by capital-endowed Kinh households – the majority ethnic group in Vietnam – and on the other hand the corresponding displacement of poor households of ethnic minorities relying on shifting cultivation towards the forest margins. This study tested this hypothesis by using remote sensing to analyze land use and cover changes and deforestation trajectories in the coffee-growing area in Dak Lak and Dak Nong provinces over 2000–2010. Land use changes were linked with socioeconomic dynamics using secondary statistics and spatial modelling. Net deforestation reached ?0.31% y^?1 of the total area between 2000 and 2010. Deforestation was indeed mainly directly caused by shifting cultivation for annual crops, but this was partly driven indirectly by expansion of coffee and other perennial crops over agricultural lands. Displacement of shifting cultivation into the forest margins, pushed by market crops expansion, was the spatial manifestation of the marginalization of local ethnic minorities and poor migrants, pushed by capital-endowed migrants. This marginalization is a long-standing process rooted in the colonization and development strategy for the highlands followed since colonial times. Over the late 2000s, rapid deforestation was strongly reducing the benefits of national-scale forest recovery, and might shift the country back to net losses of natural forest. Implications for policies that may affect deforestation are discussed.     

Effects of “realistic” land-cover change on a greenhouse-warmed African climate

The primary goal of this investigation is to focus on a realistic scenario for simulating impacts on regional African climate of future deforestation in a greenhouse-warmed world. Combined effects of plausible land-cover change and greenhouse warming are assessed by time-slice simulations with an atmospheric general circulation model (AGCM) for the middle of the twenty first century. Three time-slice integrations have been performed with the ARPEGE-Climat AGCM incorporating a zooming technique to achieve a resolution of about 100 km over Africa. A control run for the current climate is forced by observed climatological sea surface temperatures (SSTs) and the observed vegetation distribution is specified from a new vegetation database, in order to improve the geographical distribution and properties of the vegetation cover. Future SST changes are derived from a transient coupled atmosphere–ocean simulation for scenario B2 of the International Panel on Climate Change (IPCC). Future vegetation changes are specified from a simulation of scenario B2 with the Integrated Model to Assess the Global Environment (IMAGE) developed at the National Institute of Public Health and the Environment in the Netherlands (RIVM). The results show that land surface processes can locally modulate greenhouse warming effects for African climate, with reductions of surface transpiration and small increases of surface temperature. Deforestation of tropical Africa has overall only a marginal effect on precipitation because of a compensatory increase in moisture convergence. Energy budget analyses show that increases in surface temperature are produced both by increases of greenhouse gases (GHG) concentration from the increase in downward atmospheric longwave radiation, and by African tropical deforestation from the resulting reduction in transpiration. This study indicates that realistic land-use changes, though of smaller amplitude than greenhouse gas forcing, may have a small regional effect in projections of future climate.     

Deforestation dynamics and drivers in different forest types in Latin America: Three decades of studies (1980–2010)

Over the last decades there have been a considerable number of deforestation studies in Latin America reporting lower rates compared with other regions; although these studies are either regional or local and do not allow the comparison of the intraregional variability present among countries or forest types. Here, we present the results obtained from a systematic review of 369 articles (published from 1990 to 2014) about deforestation rates for 17 countries and forest types (tropical lowland, tropical montane, tropical and subtropical dry, subtropical temperate and mixed, and Atlantic forests). Drivers identified as direct or indirect causes of deforestation in the literature were also analysed. With an overall annual deforestation rate of −1.14 (±0.092 SE) in the region, we compared the rates per forest type and country. The results indicate that there is a high variability of forest loss rates among countries and forest types. In general, Chile and Argentina presented the highest deforestation rates (−3.28 and −2.31 yearly average, respectively), followed by Ecuador and Paraguay (−2.19 and −1.89 yearly average, respectively). Atlantic forests (−1.62) and tropical montane forests (−1.55) presented the highest deforestation rates for the region. In particular, tropical lowland forests in Ecuador (−2.42) and tropical dry forests in Mexico (−2.88) and Argentina (−2.20) were the most affected. In most countries, the access to markets and agricultural and forest activities are the main causes of deforestation; however, the causes vary according to the forest types. Deforestation measurements focused at different scales and on different forest types will help governments to improve their reports for international initiatives, such as reducing emissions from deforestation and forest degradation (REDD+) but, more importantly, for developing local policies for the sustainable management of forests and for reducing the deforestation in Latin America.     

Possible climatic impacts of land cover transformations,with particular emphasis on tropical deforestation

The climatic impact of albedo changes associated with land-surface alterations has been examined. The total surface global albedo change resulting from major land-cover transformations (i.e. deforestation, desertification, irrigation, dam-building, urbanization) has been recalculated, modifying the estimates of Sagan et al., (1979). Tropical deforestation (11.1 million ha yr^-1, or 0.6% yr^-1, Lanly, 1982) ranks as a major cause of albedo change, although uncertainties in the areal extent of desertification could conceivably render this latter process of similar significance. The maximum total global albedo change over the last 30 yr for the various processes lies between 0.000 33 and 0.000 64, corresponding to a global temperature decrease of between 0.06 K and 0.09 K (scaled from the 1-D radiative convective model of Hansen et al., 1981), which falls well below the interannual and longer period variability.An upper bound to the impact of tropical deforestation was obtained by concentrating all vegetation change into a single region. The magnitude of this modification is equivalent to 35–50 yr of global deforestation at the current rate, but centered on the Brazilian Amazon. The climatic consequences of such tropical deforestation were simulated, using the GISS GCM (Hansen et al., 1983). In the simulation, a total area of 4.94 × 10⁶ km² of tropical moist forest was removed and replaced by a grass/crop cover. Although surface albedo increased from 0.11 to 0.19, the effect upon surface temperature was negligible. However, other climate parameters were altered. Rainfall decreased by 0.5–0.7 mm day^-1 and both evapotranspiration and total cloud cover were reduced. The absence of a temperature decrease in spite of the increased surface albedo arises because the reduction in evapotranspiration has offset the effects of radiative cooling. The decrease in cloud cover also counteracts the increase in surface albedo. These locally significant changes had no major impact on regional (Hadley or Walker cells) or the global circulation patterns.We conclude that the albedo changes induced by current levels of tropical deforestation appear to have a negligibly small effect on the global climate.     

Quantifying the Impact of Global Climate Change on Potential Natural Vegetation

Impacts of climate change on vegetation are often summarized in biome maps, representing the potential natural vegetation class for each cell of a grid under current and changed climate. The amount of change between two biome maps is usually measured by the fraction of cells that change class, or by the kappa statistic. Neither measure takes account of varying structural and floristic dissimilarity among biomes. An attribute-based measure of dissimilarity (V) between vegetation classes is therefore introduced. V is based on (a) the relative importance of different plant life forms (e.g. tree, grass) in each class, and (b) a series of attributes (e.g. evergreen-deciduous, tropical-nontropical) of each life form with a weight for each attribute. V is implemented here for the most used biome model, BIOME 1 (Prentice, I. C. et al., 1992). Multidimensional scaling of pairwise V values verifies that the suggested importance values and attribute weights lead to a reasonable pattern of dissimilarities among biomes. Dissimilarity between two maps (V) is obtained by area-weighted averaging of V over the model grid. Using V, present global biome distribution from climatology is compared with anomaly-based scenarios for a doubling of atmospheric CO2 concentration (2 × CO2), and for extreme glacial and interglacial conditions. All scenarios are obtained from equilibrium simulations with an atmospheric general circulation model coupled to a mixed-layer ocean model. The 2 × CO2 simulations are the widely used OSU and GFDL runs from the 1980's, representing models with low and high climate sensitivity, respectively. The palaeoclimate simulations were made with CCM1, with sensitivity similar to GFDL. V values for the comparisons of 2 × CO2 with present climate are similar to values for the comparisons of the last interglacial and mid-Holocene with present climate. However, the two simulated 2 × CO2 cases are much more like each other than they are to the simulated interglacial cases. The largest V values were between the last glacial maximum and all other cases, including the present. These examples illustrate the potential of V in comparing the impacts of different climate change scenarios, and the possibility of calibrating climate change impacts against a palaeoclimatic benchmark.     

Spatially complex land change: The Indirect effect of Brazil's agricultural sector on land use in Amazonia

Soybean farming has brought economic development to parts of South America, as well as environmental hopes and concerns. A substantial hope resides in the decoupling of Brazil's agricultural sector from deforestation in the Amazon region, in which case expansive agriculture need not imply forest degradation. However, concerns have also been voiced about the potential indirect effects of agriculture. This article addresses these indirect effects for the case of the Brazilian Amazon since 2002. Our work finds that as much as thirty-two percent of deforestation, or the loss of more than 30,000 km² of Amazon forest, is attributable, indirectly, to Brazil's soybean sector. However, we also observe that the magnitude of the indirect impact of the agriculture sector on forest loss in the Amazon has declined markedly since 2006. We also find a shift in the underlying causes of indirect land use change in the Amazon, and suggest that land appreciation in agricultural regions has supplanted farm expansions as a source of indirect land use change. Our results are broadly congruent with recent work recognizing the success of policy changes in mitigating the impact of soybean expansion on forest loss in the Amazon. However, they also caution that the soybean sector may continue to incentivize land clearings through its impact on regional land markets.     

On Modeling Dry Deposition of Long-Lived and Chemically Reactive Species over Heterogeneous Terrain

An explicit multi-layer subgrid-scheme was developed for ameso-/-scale model to consider subgrid-scale surface heterogeneity, dry deposition, biogenic and anthropogenic emission of trace gases. Since dry deposition measurements of highly reactive trace species are scarce we try to evaluate this scheme by heuristic principles. The results of simulations conducted for a 5×5 km² resolution with and without thisscheme are evaluated by using results of a model run with 1×1 km²resolution, which is taken as a `grand thruth' and which has the same resolution as the subgrid. The explict multi-layer subgrid scheme provides a similar distribution of dry deposition fluxes as the much more computationally expensive simulation with the 1×1 km² resolution.Dry deposition fluxes determined from observations give evidence that the explicit multi-layer subgrid scheme which does not require a constant flux approximation for a layer of several decameters leads to an improvement in determining the exchange between the atmosphere and the ground.Results of simulations with a microscale model show that the inhomogeneity at forest edges leads to an increase of the turbulent transports of up to a factor 4 compared to horizontally homogeneous terrain, which is assumed to be the conditions of the subgrid cells (and which is usually the assumption for the entire grid cell in mesoscale models). Inhomogeneity inside an extended stand of trees causes an overall increase of 5–10% withhigh local extremes, i.e. such an inhomogeneity results to an underestimation of dry deposition in meso-/-scale models. The effects are most pronounced for a wind direction perpendicular to the forest edge.     

Subnational institutions and power of landholders drive illegal deforestation in a major commodity production frontier

Deforestation is a main threat to the biosphere due to its contribution to biodiversity loss, carbon emissions, and land degradation. Most deforestation is illegal and continues unabated, representing around half of the total deforestation in the tropics and subtropics. Quantifying illegal deforestation is challenging, let alone assessing the social and institutional processes underlying its occurrence. We tackle this challenge by quantifying the relative influence of individual (i.e., landholders’ power, landholding size) and contextual (i.e., subnational institutions, agricultural suitability) factors on the type and size of illegal deforestation in the Argentine Dry Chaco, a major commodity production frontier and global deforestation hotspot. We build a Bayesian network fed with data of 244 illegal deforestation events, obtained from journalistic articles, grey literature, key informant interviews, and geospatial analyses. The results reveal that more powerful landholders were associated with larger illegal deforestation events. Policy simulations suggest that higher concentration of land in the hands of powerful landholders and more flexible subnational forest regulations would escalate illegal deforestation. This points to the need for a smart policy mix that integrates across economic, agricultural, and environmental sectors to halt illegal deforestation at commodity production frontiers. A land tenure reform can facilitate forest protection, while incentives to land-use diversification and the criminal prosecution of illegal deforestation are critical to shift landholder behavior towards more balanced production and conservation outcomes.