Ferrogrão railway line with terminal in Matupá will split Xingu lands in half

William Costa, Juliana Davis, Amanda Oliveira, Fabrício Fernandes,Raoni Rajão, Britaldo Silveira Soares Filho

The EF-170 railway, Ferrogrão, is one of the infrastructure projects that aims to consolidate routes for the outflow of agricultural commodities through Brazil’s Arco Norte ports.

On a line running parallel to the BR-163 highway, this railway is emerging as an alternative to reduce the cost of transporting agricultural commodities and improve the competitiveness of production in the region, which is why the current government considers the project a priority. However, there is some controversy regarding the economic viability of the project [1]  and its social and environmental impacts on the Xingu and Tapajós river basins, as these regions are considered critical for forest conservation, maintenance of the water regime and the protection of indigenous peoples. Furthermore, environmental impact studies for the project only cover an area equivalent to a 10-km strip on either side of the railway [2], even though there is clear evidence that the impact extends well beyond this arbitrary boundary [3][4]. In a previous study, we identified the municipalities in the region that would benefit from the reduction in transportation costs following construction of the Ferrogrão line and we mentioned how regional transport infrastructure ventures can have synergistic and cumulative impacts and promote and/or accelerate processes of change in land use[5].

In this vein, we draw attention to the partially paved state highway MT-322, formerly BR-080, which links the municipality of Novo Mundo in Mato Grosso to the border with the state of Goiás. As part of the Amazon integration plan, the road was built in the 1970s and its main ancillary function is currently the outflow of agricultural production from the region, connecting BR-163 (known as Cuiabá- Santarém) to BR-158 (Fig. 1). Since its implementation, MT-322 has been a source of conflicts and disputes, since it cuts across the Xingu River in a region historically inhabited by indigenous populations. In 1971, following construction of the highway, a portion of the territory in the Xingu area was excluded from the proposed boundaries of the Xingu Indigenous Park (XIP), splitting up some of the Kayapó population that lived there [6]. It was not until 1984, after the so-called “ferry war”[7], that the territory was demarcated as the Capoto-Jarina Indigenous Land (IL). Since then, its population has been responsible for managing the crossing, charging a fee to all non-indigenous people who use the ferry[8]. This 80-km stretch of highway, which passes through the XIP perimeter and within the Capoto-Jarina IL, is in use despite not having an environmental licence or having been the subject of environmental impact studies on the indigenous populations in these territories[9].

Figure 1 – Land use and road infrastructure around the Xingu Indigenous Park and Capoto-Jarina Indigenous Land – MZ (MILITARY ZONE), IL (indigenous land), SP (state park), ECST (ecological station) and BIORE (biological reserve).

Over the last decade, there has been increased pressure from local producers to pave the route and build a bridge over the Xingu River, expediting the outflow of products from the region[10]. In addition, the lack of paving along the route is detrimental to vehicle traffic, primarily in rainy periods, when quagmires prevent the progress of traffic. Nevertheless, although the matter of paving the road has split opinion among indigenous people 8, the proposal to build a bridge over the Xingu River has been rejected by the indigenous associations, who assert that replacing the ferry with a bridge would deprive them of control over third parties accessing their territories, leading to an uncontrolled increase in the flow of vehicles on the highway and consequently an increase in human pressures on the territories.

At this juncture, this analysis complements the aforementioned study 5, focusing on the synergistic and cumulative impacts of building the Ferrogrão line along with other undertakings planned in the region around the railway. In particular, we are examining how implementation of the railway, as well as a transshipment terminal in Matupá, plus asphalting of the MT-322 section along with construction of the bridge over the Xingu River, could increase the flow of transportation in the region, leading to deforestation and the resulting social and environmental impacts. Neglecting the impact beyond the 10-km buffer has had serious implications on compliance with the right of prior consultation of the indigenous peoples in the Xingu and Tapajós basins[11].

To estimate the impact of the Ferrogrão line on MT-322 and therefore on the XIP and Capoto-Jarina IL, we have used the OtimizaINFRA model, which simulates transport logistics in Brazil. As its core data, the model uses templates with the origin and destination of the transported products, infrastructure maps and freight costs for the different modes of transport, among other assumptions 5. The model simulates the transportation routes based on the lowest cumulative cost between the origins (producer municipalities in the case of soybean) and destinations (export units of the Federal Revenue Service (FRS), such as ports). To evaluate the economic impact of implementing new infrastructures on transportation costs, simulations with and without the planned projects are compared. In addition to this response, the model also outputs cargo movements and thus vehicle movements along the entire transport infrastructure. For this study, we simulated the flow of soybean transportation, taking into account:

  1. Current scenario: the current infrastructure and origin-destination template for 2018.
  2. Ferrogrão scenario: implementation of the Ferrogrão line with only two terminals, the first in Sinop/MT and the last in Miritituba, Itaituba/PA; reproduction of the transported loads, using the origin-destination template for 2018; and optimisation of the destination (using the port with the lowest access cost).
  3. Ferrogrão + Works scenario: implementation of the Ferrogrão line with three terminals, the first in Sinop/MT, an intermediate one in Matupá/MT and the last one in Miritituba, Itaituba/PA; reproduction of the transported loads using the origin-destination template for 2018 and optimisation of the destination (using the port with the lowest access cost); and works on the MT-322, including asphalting of the route and construction of a bridge at the current ferry site. For this last scenario, the SimAmazonia deforestation model was also run [12] incorporating the effect of increased vehicle traffic on the MT-322 in a scenario of weak environmental governance[13].

As a result, OtimizaINFRA estimates that there is currently daily movement of 125 and 32 soybean trucks on BR-163 and BR-158, respectively, the two main routes providing access to the northern ports. This corresponds to an annual load of 1.4 million tonnes and 345,000 tonnes of soybean respectively. However, the model does not consider soybean transportation movements on the MT-322, as traffic from the north-eastern municipalities in the state of Mato Gross feeds into the BR-158 which leads to the port of Belém (Fig. 2).

In the scenario in which Ferrogrão is implemented, the railway is a cheaper alternative than using the BR-163. Once the railway is available, the flow of soybean transportation trucks on the highway is reduced by 81%, for a total of 24 trucks per day, or 260,000 tonnes per year, whereas the railway has a flow of 435 carriages per day, for a total of 16 million tonnes per year (Fig. 3). Once again, there were no significant soybean transportation movements on the MT-322; however, the nearby routes, MT-430, MT-431 and MT-437, do receive some flow of cargo. Figure 3 shows the logistical catchment areas (areas of influence) of the ports, demonstrating that, even though Ferrogrão reduces the cost of accessing the Miritituba terminal or the port of Santarém, in the north-eastern portion of the state of Mato Grosso, it is still cheaper to take cargo to the port of Belém via the BR-158.

Figure 2 – Flow of soybean transportation in the current scenario.

The third scenario, which includes the transshipment point in Matupá and the works on the MT-322, indicates that this intermediate station would attract a larger amount of cargo to the railway, further reducing traffic on the BR-163. Ferrogrão would increase its daily movement to 480 carriages or 17.5 million tonnes per year, and traffic on the BR-163 would be reduced to just 24,000 tonnes annually, or the equivalent of 2 trucks per day with a capacity of 30 tonnes. The asphalting work and construction of the bridge would also alter the dynamics of the transportation flow in the region, as the municipalities highlighted in Figure 4 would have reduced transportation costs as a result of these works and would become part of the Santarém port catchment area. The improved quality of the road and elimination of the ferry cost would make the MT-322 an advantageous route for accessing the Ferrogrão line. As such, the daily traffic on the section of road crossing through Xingu Indigenous Territory (TIX) and Capoto-Jarina IL would increase to 174 soybean trucks or 1.9 million tonnes per year.

There are plans for another railway, FICO (Ferrovia de Integração Centro Oeste [Central West Integration Railway]), in Mato Grosso for the outflow of agricultural production, which is currently in the study phase (Fig.4). We also simulated the last scenario with the inclusion of this project to ascertain its role in the state’s logistics, should it become a reality. As in the previous cases, optimisation of the destination was taken into account (using the port with the lowest access cost). However, no difference was identified in the dynamics of the soybean flows already described. This is because the said railway would not provide any advantages to the municipalities of Mato Grosso in accessing Santarém and Belém, the ports with the lowest cumulative cost.

Figure 3 – Flow of soybean transportation in the Ferrogrão scenario.

Despite the gains from the reduced transportation costs (Fig. 6), the heavy flow of vehicles due to construction of the bridge over the Xingu River (Fig. 4), intended to connect the soybean-producing municipalities to the east of the XIP, will not only increase pressure due to the conversion of forests to agricultural land, it will accelerate the invasion and subsequent deforestation of the region’s indigenous lands. Using the SimAmazonia model (13), it is estimated that in a scenario of weak environmental governance[14] which is currently the case, more than 230,000 hectares would be felled by 2035 just within the indigenous lands in the east of Mato Grosso, with more than half of that deforestation occurring in the Xingu Indigenous Park alone (Fig. 5). It is equally concerning that deforestation within the logistical basin of the Ferrogrão line in Mato Grosso (Fig. 4) would reach 65% by 2035. Furthermore, the economic impact of the Ferrogrão line goes well beyond the cannibalisation of the BR-163, which would become just a regional route. Were the Matupá transshipment terminal to also be implemented, the economic losses resulting from CO2 emissions caused by deforestation would be in the region of US$ 1 billion (US$ 10/tonne of CO2) in the Indigenous Lands alone. This is on top of the reduction in the annual volume of rainfall, which has already fallen by 48% in some regions, leading to reduced agricultural productivity[15] and energy production at Belo Monte, which could fall to 25% of maximum capacity[16], as well as the incalculable loss of environmental services[17] and the rich sociobiodiversity in the region 6[18].


Figure 5 – Deforestation today and in 2035, as simulated by SimAmazonia.

In short, the Matupá terminal could potentially set in motion a process that would in fact split the contiguous blocks of indigenous lands of the Xingu Park and Capoto-Jarina IL due to occupation of their lands along the MT-322. Therefore, any analysis of the environmental impact of the Ferrogrão should take into account the entire area of influence of the undertaking, and not only the 10-km area on either side of the line. Moreover, this major socio-environmental and economic impact should not be outsourced by the undertaking; on the contrary, it should be considered in the analysis of its economic viability and, consequently, its potential return for Brazilian society[19].

Figure 6 – Current cost of soybean transportation and cost in the Ferrogrão + Works scenario for municipalities in east Xingu.



  1. Assunção J, Bragança A, Araújo R (2020) Resumo para política pública. Os impactos ambientais da Ferrogrão: uma avaliação ex-ante dos riscos de desmatamento. Rio de Janeiro: Climate Policy Initiative. Disponível em: https://www.climatepolicyinitiative.org/wp-content/uploads/2020/03/PB_Os-impactos-ambientais-da- Ferrograo- 1.pdf.
  2. Lideranças indígenas do povo Kayapó Menkragnoti e do povo Munduruku (2021) Carta conjunta Kayapó e Munduruku ao Tribunal de Contas da União (TCU). Disponível em: https://ox.socioambiental.org/sites/default/files/ficha-tecnica//node/142/edit/2021-03/carta-mdk-kyp.pdf.
  3. Sonter LJ, Herrera D, Barrett DJ, Galford GL, Moran CJ, Soares-Filho BS (2017) Mining drives extensive deforestation in the Brazilian Amazon. Nature Communications, 8:1013.
  4. Rede Xingu+ (2018) Carta a Associação Brasileira das Indústrias de Óleo Vegetal (Abiove), Associação
    Brasileira de Proteína Animal (ABPA), Associação Nacional dos Exportadores de Cereais (Anec), Banco do Brasil
    (BB), Banco Nacional do Desenvolvimento (BNDES), China Construction Bank (CCB), Estação da Luz Participações (EDLP), Embaixada da China, Produzir Conservar Incluir (PCI-MT), Rabobank, Sicredi, The
    Consumers Goods Forum (CGF) e Tropical Forest Alliance (TFA). Disponível em: https://www.socioambiental.org/sites/blog.socioambiental.org/files/blog/pdfs/carta_do_encontro_tematic o_xingu_sobre_os_impactos_socioambientais_da_ferrograo_e_direito_de_consulta.pdf..
  5. Leles W, Davis J, Ribeiro A, Soares-Filho BS (2020) Amazônia do futuro: o que esperar dos impactos socioambientais da Ferrogrão?. CSR, Policy brief. Disponível em: https://csr.ufmg.br/csr/wp- content/uploads/2020/11/Ferrograo_policy-brief_.pdf
  6. Instituto Socioambiental (2011) Almanaque Socioambiental Parque Indígena do Xingu: 50 anos. São Paulo: Instituto Socioambiental. Disponível em: https://www.socioambiental.org/sites/blog.socioambiental.org/files/publicacoes/10380_0.pdf.
  7. Lea V, Ferreira MKL (1985) A guerra no Xingu: Cronologia. In: Ricado CA (Ed.). Povos Indígenas no Brasil 1984. São Paulo: Cedi. p 246-258
  8. Oliveira ES (2017) A terra (vivida) em movimento: nomeação de lugares e a luta Mẽtyktire-Mẽbêngôkre (Kayapó). Dissertação (Mestrado em Antropologia Social). Universidade de Brasília
  9. Rede Xingu+. Sumário. Disponível em: https://xingumais.org.br/obra/mt-322-br-080-trecho-matupa-mt- km- 372. Acesso em 20 de abril de 2021
  10. The slow crossing, toll prices and irregular ferry hours have been causes of discontent among users of the road[footnote]Patroni L (2021) Atoleiros e indignação: logística precária na MT-322 dificulta escoamento da safra em MT. Canal Rural Mato Grosso, 10 de fevereiro de 2021. Disponível em: https://blogs.canalrural.com.br/canalruralmatogrosso/2021/02/10/atoleiros-e-indignacao-logistica-precaria- na-mt- 322-dificulta-escoamento-da-safra-em-mt/.
  11. Ministério Público Federal (2021) MPF reafirma direito à consulta prévia, livre e informada de povos indígenas e comunidades tradicionais atingidos pela Ferrogrão. Procuradoria Geral da República, 24 de maio de 2021. Disponível em: http://www.mpf.mp.br/pgr/noticias-pgr/ferrograo-mpf-reafirma-direito-a-consulta- previa- livre-e-informada-de-povos-indigenas-e-comunidades-tradicionais-atingidos.
  12. Soares-Filho BS, Nepstad D, Curran L, Voll E, Cerqueira G, Garcia RA, Ramos CA, Mcdonald A, Lefebvre P, Schlesdinger P (2006) Modeling conservation in the Amazon basin. Nature, 440:520-523.
  13. Rochedo PRR, Soares-Filho BS, Schaeffer R, Viola E, Szklo A, Lucena AFP, Koberle A, Davis JL, Rajão R, Rathmann R (2018) The threat of political bargaining to climate mitigation in Brazil. Nature Climate Change, 8: 695–698
  14. Rochedo PRR, Soares-Filho BS, Schaeffer R, Viola E, Szklo A, Lucena AFP, Koberle A, Davis JL, Rajão R, Rathmann R (2018) The threat of political bargaining to climate mitigation in Brazil. Nature Climate Change, 8: 695–698.
  15. Leite-Filho AT, Soares-Filho B, Davis JL, Abrhão GM, Börner J (2021) Deforestation reduces rainfall and agricultural revenues in the Brazilian Amazon. Nature Communications, 12.
  16. Stickler CM, Coe MT, Costa MH, Nepstad DC, McGrath DG, Dias LC, Rodrigues HO, Soares-Filho BS (2013) Dependence of hydropower energy generation on forests in the Amazon Basin at local and regional scales. Proc Nat Acad Sci USA 110:9601–9606
  17. Strand J, Soares-Filho B, Costa HM, Oliveira U, Ribeiro SC, Pires GF, Oliveira A, Rajão R, May P, Hoff R, Siikamäki J, Motta RS, Toman M (2018) Spatially explicit valuation of the Brazilian Amazon Forest’s Ecosystem Services. Nature Sustainability, 1(11): 657-664.
  18. Heckenberger Michael J, Christian Russell J, Toney Joshua R and Schmidt Morgan J (2007) The legacy of cultural landscapes in the Brazilian Amazon: implications for biodiversity. Phil. Trans. R. Soc., B362197–208.
  19. Rajão, Raoni, Fernandes Júnior, José Leomar, Melo, Lidiane (2021) Grandes obras de infraestrutura e o risco de corrupção e inviabilidade econômica: uma análise exploratória. Tribunal de Contas da União.