The role of soybean production as an underlying driver of deforestation in the South American Chaco
Introduction
Agricultural expansion into tropical forests continues to be a main driver of global environmental change (Aide et al., 2013, Hansen et al., 2013), resulting in major carbon emissions (Baccini et al., 2012, Carlson et al., 2013), regional climate change (Butt et al., 2011, Silverio et al., 2015), widespread degradation of ecosystem services, and massive biodiversity loss (Metzger et al., 2006, Sala et al., 2000). Halting or slowing down tropical forest loss have thus become international priorities, and understanding what drives deforestation is essential to do so (Geist and Lambin, 2002). While the world's moist tropical forests have been in focus, the question of what drives deforestation in tropical dry forests and savannas remains understudied (Blackie et al., 2014, Lehmann, 2010, Parr et al., 2014) although deforestation there has been rampant recently, for example in Chaco in Argentina (Vallejos et al., 2015), the Cerrado in Brazil (Espirito-Santos et al., 2016), the Miombo woodlands (Mayes et al., 2015) or in Myanmar (Wohlfart et al., 2014).
This is particularly true for the tropical dry forests and savannas of South America. Rising agricultural commodity prices, an increasing integration of South American countries into world markets, favorable climatic conditions, the availability of cheap land and labor, as well as increasing forest protection in the Amazon all have contributed to turning South America’s tropical dry forests and savannas into major agricultural expansion frontiers, particularly regarding soybean and beef production (Garrett et al., 2013, Lambin et al., 2013, Parr et al., 2014). This has triggered widespread forest loss (Hansen et al., 2013), for example in the Cerrado in Brazil (Garcia and Ballester, 2016, Grecchi et al., 2014, LAPIG, 2014), the Chiquitania in Bolivia (Müller et al., 2012), or the Chaco in Argentina and Paraguay (Caldas et al., 2015, Kuemmerle et al., 2017). Given the speed of forest loss in these regions, understanding what drives these losses is important.
While the expansion of soybean cultivation and cattle ranching are both important proximate drivers of forest loss in South America’s dry forests (Gasparri et al., 2013, Graesser et al., 2015, Volante et al., 2016), only a handful of studies quantitatively examined the relative importance of these drivers (Aide et al., 2013, Barona et al., 2010, Garcia and Ballester, 2016, Macedo et al., 2012, Rosa et al., 2015). Likewise, the underlying mechanisms that lead to soybean or cattle ranching expansion are not fully understood: there is an increasing recognition that soybean and cattle ranching form a linked system (Gasparri and le Polain de Waroux, 2014), with farmers engaging in both, with the goal to decrease their dependency on global prices and to reinvest revenues more efficiently (Goldfarb and Zoomers, 2013). For example, during a period of decreasing global soybean prices, it may be more profitable for farmers to invest in the acquisition of forests to expand cattle ranching, whereas under high soybean prices, pastures can be converted into croplands. Likewise, actors may clear land for pastures as a hedge against tightening deforestation regulations, with the ultimate goal of converting pastures to soybean production later (Baumann et al., 2016, le Polain de Waroux et al., 2016). Moreover, profits from pasture lands are generally less vulnerable to drought years compared to soybean cultivation (Gasparri et al., 2013). The empirical evidence for such linkages, however, remains weak.
Even where farmers do not engage in both soybean and cattle ranching, soybean expansion in one region may displace cattle ranching to another, such as in the Brazilian Amazon (Cohn et al., 2016, Gollnow and Lakes, 2014, Morton et al., 2006), from the Argentine Pampas to the Argentine Chaco (Gasparri et al., 2013), or from the Uruguayan Pampas and the Paraguayan Atlantic Forest region to the Paraguayan Chaco (Bertello, 2008, Gonzales, 2013, World Wildlife Fund, 2015). In all these cases cattle ranching appears to be the main proximate driver of deforestation, while soybean expansion may be ultimately underlying forest loss. This would mean that policies and conservation strategies have the potential to fall short if they only target cattle ranching and neglect the role of soybean as an underlying driver.
The Chaco, the largest remaining continuous stretch of tropical dry forest in South America (Portillo-Quintero and Sanchez-Azofeifa, 2010), is an interesting region to learn more about the direct and indirect role of soybeans as a driver of forest loss. Since the 1980s, technological innovation, rising global prices, favorable farming conditions, and the opening of regional land markets to internationally operating agribusinesses triggered rapid deforestation and agricultural expansion in the region, mainly for cattle ranching and soybean cultivation (Mastrangelo and Gavin, 2012, Zak et al., 2008, Zak et al., 2004). Because the Chaco extends into Argentina, Bolivia, and Paraguay, its cross-border setting provides a unique natural experiment to analyze drivers of deforestation under different policy settings, socio-economic conditions, and conservation paradigms.
While soybean cultivation and cattle ranching are both important proximate drivers of deforestation in the Chaco, their relative importance varies across the region. For example, in Argentina cattle ranching is the dominant proximate driver of deforestation (Volante et al., 2012), as many forest areas are being cleared for croplands, particularly soybeans (Baumann et al., 2017a). In contrast, in the Paraguayan Chaco almost all deforestation is to establish pastures, making it the dominant proximate driver there (Caldas et al., 2015, Carr, 2004), much like to northwestern Argentina (Volante et al., 2016). The relative importance of proximate drivers of deforestation also remains unclear because most studies to date rely on satellite imagery to map forest loss, typically for short time periods or small study regions. Existing work has often focused on assessing the spatial determinants of deforestation patterns (Gasparri et al., 2015, Killeen et al., 2008, Volante et al., 2016, Zak et al., 2008), rather than assessing the underlying drivers of these forest losses. An assessment of the role of soybean as a direct and indirect driver of deforestation for the Chaco as a whole is missing.
Consistent, high temporal- and spatial-resolution time series of deforestation in the Chaco (Hansen et al., 2013, Vallejos et al., 2015) provide new opportunities to analyze drivers of deforestation across national boundaries. Analyzing drivers of land-use change in a panel regression framework using time series data is a powerful way to detect possible causal relations (Butsic et al., 2015, Gibbs et al., 2016, Meyfroidt, 2015) for several reasons. First, panel regressions allow linking observed deforestation to changes in potential underlying drivers (e.g., market prices), while controlling for location factors that may govern the spatial patterns of change (e.g., soils, rainfall patterns, ownership, distance to markets). Second, time series models allow for detecting the impact of sudden changes in underlying drivers (e.g., policy interventions, new legislation, or currency devaluation), which may be hard to detect when focusing on longer time periods (Cameron and Trivedi, 2005). Finally, panel models are useful tools for discovering time lags in time series (Tao and Yu, 2012), and may thus reveal such time lagged responses of deforestation rates to underlying drivers. Yet, we know of no study that has assessed land-use change in the Chaco using such a statistical setup.
Our overarching goal was to quantify the drivers of forest loss for the Chaco as a whole, and to examine the role of soybean production as a direct and indirect driver of these losses. Specifically, we asked:
- (1)
What was the role of soybean and cattle ranching as proximate, direct drivers of deforestation in the Chaco between 2001 and 2012, and did the importance of these drivers vary among countries?
- (2)
How was soybean expansion outside the Chaco related to deforestation inside the Chaco, and were there time-lags in this relationship?
To answer these research questions, we tested the following hypotheses: H1 Soybean cultivation is significantly related to deforestation at the district level. H2 The national soybean area of a given country is significantly related to deforestation in the Chaco, regardless of proximate drivers of deforestation, suggesting displacement within countries. H3 Soybean production in neighboring Chaco countries is significantly related to deforestation in the Chaco part of a given country, suggesting displacement across borders. H4 There are time lags in the relationship between soybean cultivation outside, and deforestation inside the Chaco, suggesting links between cattle ranching and soybean actors.
Section snippets
Study area
Spanning 1.1 Million km2, the Chaco is the second largest forest region in South America (Bucher and Huszar, 1999). The Chaco stretches from northern Argentina to southeastern Bolivia and northwestern Paraguay (Fig. 1a). Topography is mostly flat with elevation varying between 100 and 500 m. The climate is semi-arid with a dry season from April to September. Mean annual rainfall varies from 450 mm to 1200 mm and decreases from east to west, resulting in a division into the wet Chaco (900–1200 mm)
Results
Deforestation in the Chaco was widespread between 2001 and 2012, with a total of 7.8 million ha (Mha) deforested. Yet, deforestation rates varied substantially, with higher rates during 2003–2005 (on average ∼600,000 ha of forest loss per year), 2007–2008 (∼920,000 ha/year) and 2010–2011 (∼890,000 ha/year), and somewhat lower deforestation rates between 2001 and 2002 (∼290,000 ha/year) and 2005–2006 (∼540,000 ha/year).
Soybean area increased between 2001 and 2012 from 2.3 to 5.2 Mha in the Chaco
Discussion
The tropical dry forests of Latin America are a global hotspot of deforestation, but our understanding of what drives deforestation, particularly regarding displacement effects within and among countries, remains unclear. We assessed the relative importance of the expansion of cattle ranching and soybean cultivation as direct and indirect drivers of deforestation in the South American Chaco between 2001 and 2012 using a panel regression approach. We found that cattle ranching was the most
Acknowledgements
We gratefully acknowledge support for this research by the German Research Foundation (DFG, KU 2458/5-1), by the Federal Ministry of Science and Education (BMBF, PASANOA 031B0034A), by the Einstein Foundation Berlin and the Argentine National Agricultural Technology Institute (INTA, PNNAT 1128052). We are grateful for comments made by the two anonymous reviewers, whose comments greatly improved the manuscript.
References (94)
- et al.
Sustainable management of the Gran Chaco of South America: ecological promise and economic constraints
J. Environ. Manage.
(1999) - et al.
Conservation and conflict in the Democratic Republic of Congo: the impacts of warfare, mining, and protected areas on deforestation
Biol. Conserv.
(2015) - et al.
Land tenure and agricultural expansion in Latin America: the role of indigenous peoples' and local communities' forest rights
Global Environ. Change-Human Policy Dimension.
(2015) - et al.
Patterns and processes of pasture to crop conversion in Brazil: evidence from Mato Grosso State
Land Use Policy
(2016) - et al.
Deforestation and fragmentation of Chaco dry forest in NW Argentina (1972–2007)
For. Ecol. Manage.
(2009) - et al.
Linkages between soybean and neotropical deforestation: coupling and transient decoupling dynamics in a multi-decadal analysis
Global Environ. Change
(2013) - et al.
Determinants of the spatial distribution of cultivated land in the North Argentine dry Chaco in a multi-decadal study
J. Arid Environ.
(2015) Misspecification tests and their uses in econometrics
J. Stat. Plann. Inference
(1996)- et al.
Policy change, land use, and agriculture: the case of soy production and cattle ranching in Brazil, 2001–2012
Appl. Geogr.
(2014) - et al.
Land use and land cover changes in the Brazilian Cerrado: a multidisciplinary approach to assess the impacts of agricultural expansion
Appl. Geogr.
(2014)
Estimating the world's potentially available cropland using a bottom-up approach
Global Environ. Change
Modifying Argentina: GM soy and socio-environmental change
Geoforum
Forest cover change in Miombo woodlands: modeling land cover of African dry tropical forests with linear spectral mixture analysis
Remote Sens. Environ.
The vulnerability of ecosystem services to land use change
Agric. Ecosyst. Environ.
Tropical grassy biomes: misunderstood, neglected, and under threat
Trends Ecol. Evol.
Extent and conservation of tropical dry forests in the Americas
Biol. Conserv.
Deforestation dynamics and policy changes in Bolivia's post-neoliberal era
Land Use Policy
Exchange rates, soybean supply response, and deforestation in South America
Global Environ. Change-Human Policy Dimension.
Spatially complex land change: the indirect effect of Brazil's agricultural sector on land use in Amazonia
Global Environ. Change-Human Policy Dimension.
The spatial time lag in panel data models
Econ. Lett.
Transformation dynamics of the natural cover in the dry Chaco ecoregion: a plot level geo-database from 1976 to 2012
J. Arid Environ.
Ecosystem functional changes associated with land clearing in NW Argentina
Agric. Ecosyst. Environ.
Agricultural expansion in the Semiarid Chaco: poorly selective contagious advance
Land Use Policy
Do subtropical seasonal forests in the Gran Chaco, Argentina, have a future?
Biol. Conserv.
Deforestation and reforestation of latin america and the caribbean (2001–2010)
Biotropica
Statistical confirmation of indirect land use change in the Brazilian Amazon
Environ. Res. Lett.
Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps
Nature Clim. Change
Econometric Analysis of Panel Data
The role of pasture and soybean in deforestation of the Brazilian Amazon
Environ. Res. Lett.
Land-use competition in the south american chaco
Carbon emissions from agricultural expansion and intensification in the Chaco
Global Change Biol.
Deforestation and cattle expansion in the Paraguayan Chaco 1987–2012
Regi. Environ. Change
Cada vez más productores argentinos van a sembrar soja a Uruguay
La Nacion
Tropical Dry Forests. The State of Global Knowledge and Recommendations for Future Research, CIFOR Discussion Paper No. 2
A simple test for heteroscedasticity and random coefficient variation
Econometrica
Chaco and caatinga − south american arid savannas
Evidence that deforestation affects the onset of the rainy season in Rondonia, Brazil
J. Geophys. Res-Atmos.
Liveweight gains of steers at different stocking rates on monospecific Gatton panic and Estrella grass pastures in the Chaco Central region of Paraguay
Trop. Grasslands
Land-cover change in the paraguayan chaco: 2000–2011
J. Land Use Sci.
Microeconometrics: Methods and Applications
Deforestation and cattle ranching drive rapid range expansion of capybara in the Gran Chaco ecosystem
Global Change Biol.
Carbon emissions from forest conversion by Kalimantan oil palm plantations
Nat. Clim. Change
Proximate population factors and deforestation in tropical agricultural frontiers
Popul. Environ.
Dynamics of indirect land-use change: empirical evidence from Brazil
J. Environ. Econ. Manage.
Understanding patterns of land-cover change in the Brazilian Cerrado from 2000 to 2015
Philos. Trans. R. Soc. B-Biol. Sci.
FAOSTAT Database [Online]. Food and Agriculture Organization of the United Nations
Present and potential area for agricultural use in the arid Chaco of Paraguay
Cited by (167)
Subsidizing extensive cattle production in the European Union has major implications for global agricultural trade and climate change
2024, Journal of Cleaner ProductionWoody cover and pasture within the surrounding matrix drive puma (Puma concolor) occupancy in agroecosystems of the Argentine Dry Chaco
2023, Journal for Nature Conservation