Elsevier

Biological Conservation

Volume 177, September 2014, Pages 203-209
Biological Conservation

Roads, deforestation, and the mitigating effect of protected areas in the Amazon

https://doi.org/10.1016/j.biocon.2014.07.004Get rights and content

Highlights

  • ∼95% of deforestation was within 5.5 km of a road or 1.0 km of a navigable river.

  • 35.2% of the Brazilian Amazon was highly accessible by river or road.

  • Accessible unprotected land was 43.6% deforested.

  • Accessible protected land was 10.9% deforested.

  • Protected areas avoided or displaced 34–39 thousand km2 of deforestation.

Abstract

Roads have a major impact on Amazon deforestation. However, the effects of the rapidly growing network of illegal or unofficial roads in the Amazon are usually not considered. We assessed relationships between past deforestation and existing networks of highways, navigable rivers, and all other roads, including more than 190,000 km of unofficial roads. We found that deforestation was much higher near roads and rivers than elsewhere in the Amazon; nearly 95% of all deforestation occurred within 5.5 km of roads or 1 km of rivers. Protected areas near roads and rivers had much lower deforestation (10.9%) than did unprotected areas near roads and rivers (43.6%). If one assumes that existing protected areas halt deforestation, then we estimate that 39,462 km2 of expected forest clearing would have been avoided. However, if one assumes that protected areas merely displace deforestation to other locations, then we estimate that 34,501 km2 of expected clearing would have been displaced elsewhere. We conclude that proximity to transportation networks, particularly the rapidly growing unofficial road network, is a major proximate driver of deforestation in Amazonia and that protected areas are having a strong mitigating effect on that risk.

Introduction

The Brazilian Amazon harbors one third of the world’s tropical forests and has been estimated to sustain 13% of the world’s biota (Lewinsohn and Prado, 2005). Moist, tropical forests in this region span across an area of roughly 4 million km2, 90% of which was once forested (Olson et al., 2001). This region of rich biodiversity is also experiencing some of the world’s highest deforestation rates, averaging 0.52% yr1 (18,857 km2 yr1) through the 1990s and 2000s (INPE, 2009). As of 2009, roughly 18% of forests had been converted to other land uses (Pereira et al., 2010), with an area likely larger than this 18% modified by selective logging, edge effects, surface fires, and hunting (Peres et al., 2006, Souza et al., 2005).

Throughout the tropics, major roads open up areas of forest to settlement and resource extraction (Laurance et al., 2009). In Amazonia, most deforestation has been found to occur in proximity to major roads. Alves (2002) reported that nearly 90% of deforestation occurred within 100 km of major roads. Additional studies, using 50 km as a baseline distance, have accounted for deforestation levels varying from 67% (Asner et al., 2006, Nepstad et al., 2001) to 85% (Chomitz and Thomas, 2001). However, defining regions of deforestation by such large (50 and 100 km) distances from major roads corresponds to 40% and 63% of the Amazon, respectively. Hence, these measures are imprecise and are only marginally predictive of deforestation. Despite this, it is clear that the transportation network will play a significant role in future forest clearing in the region (Fearnside, 2007, Fearnside and Graca, 2006, Kirby et al., 2006, Laurance et al., 2001, Laurance et al., 2002).

In addition to major road networks, a network of unofficial roads, built without any government oversight or incentives, is rapidly growing in the Amazon region (Arima et al., 2005, Asner et al., 2006, Brandão and Souza, 2006, Perz et al., 2007). These roads are generally built to open up forests to exploitative activities such as logging but subsequently lead to new colonization (Veríssimo et al., 1995), forest fragmentation (Arima et al., 2008), ecological degradation (Laurance et al., 2006), and increased fire risk (Cochrane, 2003, Nepstad et al., 2001). Very high annual growth rates (exceeding 40 km of new roads per 10,000 km2 of area) have been reported in some regions (Brandão and Souza, 2006). Navigable rivers provide another potential mode of access to forested regions and further promote deforestation and logging (Peres and Terborgh, 1995, Veríssimo et al., 1998), although they are largely left out of region-wide analyses of deforestation drivers.

Brazil’s extensive network of protected areas (PAs) was established to preserve priority biodiversity conservation areas, establish biodiversity corridors, and protect portions of the 23 Amazonian ecoregions identified by World Wildlife Fund – Brazil (Rylands and Brandon, 2005, Silva, 2005). By 2006, 1.8 million km2, roughly 45% of Amazonian tropical forest, was under some level of protection as federal- or state-managed land or designated as indigenous reserve. Strictly protected areas, whose primary function is to conserve biodiversity, constitute only a modest fraction (19.2%) of the Amazon PA network. Federal and State sustainable use areas allow various levels of resource use and extraction, and make up 30.6% of the network. Indigenous reserves constitute the remainder; large-scale deforestation is prohibited in these lands and hence they also play an important role in protecting forests (Schwartzman and Zimmerman, 2005). The combinations of federal- and state-managed strictly protected areas, federal- and state-managed sustainable use areas, and indigenous lands constitute the 5 types of protection that were examined in this study.

Protected areas in the Amazon fall into two distinct classes; those currently under threat of unwanted forest loss or degradation due to human activities which need to provide an active resistance to development pressure; and those under no current pressure due to their remoteness and inaccessibility, thereby affording a default protection status (Adeney et al., 2009, Barber et al., 2012, Joppa et al., 2008). This dichotomy can confound aggregated assessments of entire PA network performance. A substantial area of protected forest is located far away from transportation networks and not easily accessible, and thus can be deemed “successfully protected” even though it has not been under any direct development pressure. Several studies have shown that the Amazonian PA network as a whole has been successful at resisting development pressure and forest clearing within protected boundaries (Barber et al., 2012, DeFries et al., 2005, Joppa et al., 2008, Nepstad et al., 2006), however there are individual PAs that have not demonstrated similar success (Barber et al., 2012, Pedlowski et al., 2005).

Roads strongly influence threats to protected areas. The protection afforded by reserves against deforestation fires has been assessed using major road networks (Adeney et al., 2009), as has projecting the future effects of conservation efforts (Laurance et al., 2001, Soares-Filho et al., 2006). PAs have been shown to substantially slow the expansion of unofficial roads (Brandão and Souza, 2006), but the effects of different reserve types on patterns and rates of deforestation fostered have not been closely examined. Here we used extensive information on transportation networks to assess the status of accessible protected forests. We assessed relationships between the extended transportation network (including unofficial roads and rivers) and deforestation, and then estimated the amount of deforestation that would occur in protected lands if not for their protected status. PAs impact deforestation by either outright preventing or avoiding it, or by displacing possible deforestation elsewhere into unprotected lands. We estimated the mitigating effect of PAs under both of these scenarios. Our findings have clear implications for managing and conserving Amazonian forests.

Section snippets

Study area and data

We examined the spatial relationships between the road/river transportation networks, deforestation, and protected areas within the Brazilian Amazon – defined here as the roughly 4 million km2 of moist, tropical forest biomes delineated by the WWF Terrestrial Ecoregions (Olson et al., 2001). The network of official roads was sourced from the Instituto Brasileiro de Geografia e Estatística (IBGE) bCIMd dataset (IBGE, 2004) which included 73,553 km of roads in the region. A 22,713 km network of

Highways as deforestation drivers

PRODES land-cover data indicate that 476,925 km2 of forest within the Brazilian Amazon had been cleared up until 2006. Using these data, we assessed regions defined by both 50 and 100 km of highways for comparison with previous studies (Alves, 2002, Asner et al., 2006, Chomitz and Thomas, 2001, Nepstad et al., 2001). These regions encompass 40% and 63% of all land cover classes (Fig. 2a) and account for 85.5% and 95.5% of all cleared forest lands, respectively. As of 2006, there was 32,381 km2 of

Discussion

Protected areas do not exist as isolated islands of pristine forest but are components within a larger matrix of human disturbance and activity (Chadzon et al., 2009, Laurance et al., 2012, Lovejoy, 2006). This surrounding matrix provides context to gauge the resilience of PAs in light of the development pressure they are under (Barber et al., 2012). Many PAs in the Amazon have been identified as having “default protection” due to their remoteness and inaccessibility (Adeney et al., 2009,

Conclusions

We find that 94.9% of all deforestation in the Brazilian Amazon has occurred in a well-defined accessible zone within 5.5 km of some type of roadway or 1.0 km of a navigable river. The overall effect of protected areas is clear: less than 1.5% of all protected forest throughout the region had been cleared by 2006. Protected forests experienced less forest loss than did unprotected lands at all distances from roads and navigable rivers. All protected area types mitigated deforestation risk and had

Acknowledgements

This work was supported by NASA Headquarters under the NASA Earth and Space Science Fellowship Program (NNX08AU94H) with additional support from the Biological Diversity Program (NNX07AF16G) and the Terrestrial Ecology Program (NNX11AB89G) of the NASA Science Mission Directorate’s Earth Science Division. The unofficial road database used in this study was compiled by Júlia Ribeiro and Amintas Brandão of Imazon (Instituto do Homem e Meio Ambiente da Amazônia), Belém, Pará, Brazil through

References (50)

  • D.S. Alves

    Space-time dynamics of deforestation in Brazilian Amazônia

    Int. J. Remote Sens.

    (2002)
  • E. Arima et al.

    Loggers and forest fragmentation: behavioral models of road building in the Amazon basin

    Ann. Assoc. Am. Geogr.

    (2005)
  • E.Y. Arima et al.

    The fragmentation of space in the Amazon basin: emergent road networks

    Photogramm. Eng. Rem. Sens.

    (2008)
  • G.P. Asner et al.

    Selective logging in the Brazilian Amazon

    Science

    (2005)
  • G.P. Asner et al.

    Condition and fate of logged forests in the Brazilian Amazon

    Proc. Natl. Acad. Sci. U.S.A.

    (2006)
  • A.O. Brandão et al.

    Mapping unofficial roads with Landsat images: a new tool to improve the monitoring of the Brazilian Amazon rainforest

    Int. J. Remote Sens.

    (2006)
  • R. Chadzon

    Beyond reserves: a research agenda for conserving biodiversity in human-modified tropical landscapes

    Biotropica

    (2009)
  • Chomitz, K.M., Thomas, T.S., 2001. Geographic patterns of land use and land intensity in the Brazilian Amazon. World...
  • M.A. Cochrane

    Fire science for rainforests

    Nature

    (2003)
  • R. DeFries et al.

    Increasing isolation of protected areas in tropical forests over the past twenty years

    Ecol. Appl.

    (2005)
  • P.M. Fearnside

    Brazil’s Cuiabá-Santarém (BR-163) highway: the environmental cost of paving a soybean corridor through the Amazon

    Environ. Manage.

    (2007)
  • P.M. Fearnside et al.

    BR-319: Brazil’s Manaus-Porto Velho highway and the potential impact of linking the arc of deforestation to central Amazonia

    Environ. Manage.

    (2006)
  • S. Henders et al.

    Forest carbon leakage quantification methods and their suitability for assessing leakage in REDD

    Forests

    (2012)
  • Houghton, R.A., 2005. Tropical deforestation as a source of greenhouse gas emissions. In: Moutinho, P., Schwartzman, S....
  • IBGE, Instituto Brasileiro de Geografia e Estatística, 2004. Projeto Carta Internacional do Mundo ao Milionésimo – CIM....
  • Cited by (391)

    View all citing articles on Scopus
    View full text