Next Article in Journal
Soil Carbon and Biochemical Indicators of Soil Quality as Affected by Different Conservation Agricultural and Weed Management Options
Previous Article in Journal
Management Implications at a Protected, Peri-Urban, Salt Lake Ecosystem: The Case of Larnaca’s Salt Lakes (Cyprus)
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Land
Volume 12
Issue 9
10.3390/land12091782
land-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Mapping Slaughterhouse Supply Zones in the Brazilian Amazon with Cattle Transit Records

by
Amintas Brandão Jr.
1,*,
Lisa Rausch
1,
Jacob Munger
1 and
Holly K. Gibbs
1,2
1
Center for Sustainability and the Global Environment (SAGE), University of Wisconsin-Madison, 1710 University Avenue, Madison, WI 53726, USA
2
Department of Geography, University of Wisconsin-Madison, Madison, WI 53706, USA
*
Author to whom correspondence should be addressed.
Land 2023, 12(9), 1782; https://doi.org/10.3390/land12091782
Submission received: 8 August 2023 / Revised: 2 September 2023 / Accepted: 7 September 2023 / Published: 14 September 2023
Browse Figures
Versions Notes

Abstract

:
Mapping slaughterhouse supply zones is crucial for assessing cattle concentration, environmental impact, and promoting sustainable practices. This study combines cattle transit records (GTA in Portuguese) with rural property boundaries (CAR in Portuguese) to map these zones in the Brazilian Amazon. It evaluates the influence of Zero-Deforestation Cattle Agreements (CA) and explores the overlap between CA and non−CA slaughterhouse supply zones. Results reveal that CA slaughterhouse supply zones significantly impact forest preservation and cover a large area equivalent to entire municipalities. Over two-thirds of the study region, including most non-protected areas, falls within these zones. There is a high degree of overlap (95%) with non−CA supply zones, indicating competition for suppliers and limited expansion potential for CA. Direct and indirect suppliers are located nearby, with approximately 80% of indirect suppliers within 100 km of direct suppliers. Consequently, supply zones for both types of suppliers largely overlap. These findings demonstrate that assessing slaughterhouse deforestation risk for the entire supply chain in our study region can be achieved by mapping only the direct suppliers. This research provides valuable insights into cattle concentration, the effectiveness of zero-deforestation commitments, and the need for sustainable practices in the slaughterhouse industry.

1. Introduction

Brazil’s Amazon cattle herd has been growing rapidly in size since the first ranchers arrived in the region in the late 1960s. Over a span of three decades, the cattle population in the Brazilian Amazon experienced a remarkable six-fold increase, rising from 15 million heads in 1985 to a herd size in 2020 that matched the entire United States, totaling 93 million heads [1,2]. One of the consequences of this cattle herd surge was the increase in the total pasture area in the Amazon region, which increased from 23 million hectares (Mha) to 70 Mha over the same period—and covered 80% of the Amazon’s deforested area by 2020 [3]. Since the early 2000s, many governmental and private policies have been implemented in the Brazilian Amazon to reduce deforestation [4,5,6]. A significant reduction of 84% of the yearly deforestation rate was achieved between 2004 and 2008, from 2.78 Mha/year to 0.46 Mha/year (the region’s lowest deforestation rate), but deforestation rates rebounded by roughly 80% from 2013 to 2022, averaging more than 0.80 Mha/year during that time. While soy deforestation has slowed in the Amazon biome [7,8], pasture deforestation has continued to surge, driving the overall increase in yearly rates.
Starting in 2009, Brazil’s Federal Prosecutors (known by the acronym MPF in Portuguese), Greenpeace, and other non-governmental organizations (NGOs) have pressured and encouraged meatpacking companies to sign Zero-Deforestation Cattle Agreements (CAs) to help halt deforestation associated with cattle farming in the Brazilian Amazon. The agreements with the MPF consist of extrajudicial Terms of Adjustment of Conduct (known collectively as the Beef TAC). The Beef TAC emphasizes blocking purchases from suppliers who engage in illegal deforestation in the Legal Amazon as defined by the Brazilian Forest Code (FC) or that have committed other socio-environmental violations of the law such as forced labor, environmental embargoes, and overlap with protected areas [9,10,11,12]. The Greenpeace agreement, known as the Public Livestock Commitment or the “G4”, on the other hand, is a voluntary commitment that was also signed in 2009 by the region’s four largest slaughterhouses [13]. The G4 follows the Beef TAC’s criteria but emphasizes blocking suppliers with any deforestation (zero-deforestation criteria) and only applies in the Amazon biome. By 2016, approximately half of the state and federal slaughterhouses in the Brazilian Amazon, accounting for 70% of the region’s slaughter capacity that year, had signed the CAs [10].
Despite evidence that shows CA slaughterhouses avoid buying from properties with deforestation [11], research suggests the CAs have not resulted in significantly decreased deforestation even after over a decade of implementation [14]. This is partly because half of the major slaughterhouses in the Amazon, representing at least 30% of the slaughter market share, have still not signed onto the CAs [15]. Consequently, many direct suppliers with deforestation can continue to avoid monitoring by selling to a non−CA slaughterhouse, creating competition between the slaughterhouses that do and do not monitor deforestation [14,16,17]. In addition, animals live up to three years and move between multiple properties prior to slaughter, but only direct suppliers are monitored [15,18,19,20,21]. As a result, slaughterhouses leave out the indirect suppliers who rear and fatten the cattle before slaughter and where most of the deforestation occurs [11]. In some cases, this complexity in the supply chain can facilitate laundering, a process in which non-compliant suppliers purposefully modify documentation or transport animals to compliant suppliers before the ultimate sale to a CA slaughterhouse [11,18,21,22]. Expanding the CA to more slaughterhouses and suppliers is crucial to increase its influence in halting deforestation and reducing laundering. However, there are still challenges such as the lack of transparency in the data required to achieve this goal.
Previous studies suggest the potential benefits of combining cattle transit records (GTA in Portuguese) with rural property boundaries (CAR in Portuguese) to enhance information about indirect suppliers [11,15,16,18,19,21,22,23,24,25]. The GTA was developed by the Brazilian government in 2006 to control animal disease spread under the Unified Agricultural Health Care System [26]. It comprises information on the groups of animals being transported and their origins and destinations, while the CAR data contain property boundaries and landowner information, to manage and plan against deforestation in Brazil [27].
Merging GTA and CAR presents an opportunity to map the geographical locations of both direct and indirect suppliers. However, many challenges still need to be addressed in this approach [18,19,22]. Despite its great potential, merging GTA and CAR requires significant computational effort as various distinct programming techniques are necessary for data cleaning and standardization, as well as for organizing the information within a database. Furthermore, GTA data are publicly available only for some states, with some states lacking recent data with public access. Therefore, finding an approach to map the full supply zones for direct and indirect suppliers using only information about direct suppliers, which all slaughterhouses have access to, could be an important step forward for resolving many policy questions and business decisions.
To address this need, we have devised an innovative approach rooted in the concept of slaughterhouse supply zones. Slaughterhouse supply zones refer to geographic regions containing high concentrations of suppliers of a given slaughterhouse [10,14]. While supply zones have been used to analyze slaughterhouse influence, accurately delineating these zones has proven difficult due to data constraints. While other studies have suggested different methods to delineate the zones, including creating buffers around slaughterhouse coordinates [14], estimating maximum purchasing distances [10], and using trip lengths [28], they are mostly best guesses for direct suppliers and do not consider the full supply chain. This has resulted in imprecise zone boundaries, which impedes a comprehensive understanding of the dynamics within slaughterhouse supply chains. Our approach improves these methods and results in the first supply zone maps calibrated with actual supplier data, revealing for the first time the true sizes and shapes of these zones. Though we would expect that supply zones in other regions where supplier information is not available would differ somewhat from those that we can map on the Amazon, we believe that understanding the characteristics of these zones can provide important insights that improve risk analysis that encompasses regions or in contexts where supplier data are not available.
In this study, we conducted a comprehensive mapping of slaughterhouse supply zones, considering both direct and indirect suppliers, utilizing data obtained from GTA and CAR. The mapping process encompassed the states of Pará, Mato Grosso, and Rondônia and covered the period from 2013 to 2018, which is when comprehensive data is available for all three states. Our goal was to evaluate the following research questions: (i) What was the size of the supply zones of direct and indirect suppliers of CA and non−CA slaughterhouses from 2013 to 2018? (ii) What are the characteristics of these zones and suppliers in terms of overlap and distances among them, land use, and deforestation? (iii) How would CA supply zones change if the first layer of indirect suppliers were to be monitored? And finally, (iv) How would they change if more slaughterhouses joined the CA? Our findings provide a comprehensive geographical picture of the cattle supply chain across current and potential CA supply zones for slaughterhouses in the three most important cattle-producing states in the Brazilian Amazon.

2. Materials and Methods

2.1. Study Area

Our study covers Pará, Mato Grosso, and Rondônia, which span 239 Mha, accounting for approximately 48% of Legal Amazon (Figure 1). These states are of particular significance in the context of our research as they represent 80% of the total cattle slaughtered in the Amazon. This area contains portions of three biomes: the Amazon, Cerrado, and Pantanal. The Amazon biome covers 82% of the land, followed by the Cerrado with 15% and the Pantanal with 3%. More than half of the Amazon and Pantanal biomes and 17% of the Cerrado biome are within our study area.

2.2. Processing the GTA-CAR Data

To identify direct and indirect suppliers, we downloaded roughly ten million GTA transactions from the websites of the state animal sanitation offices in Pará, Mato Grosso, and Rondônia, as well as the Brazilian Ministry of Agriculture, Livestock, and Food Supply (MAPA in Portuguese) (Table A1). These GTA transactions took place between January 2013 and December 2018 and moved a total of 233 million heads during this period for various purposes (Table S1). Approximately 25% of the total cattle heads were moved for slaughter, while 75% were for other purposes, with the majority being for fattening (58%) and breeding (14%). Since GTAs do not contain individual information about animals, it is possible that one or more animals have been counted more than once in the total count of moved animals. To assess the scope of our GTA database, we juxtaposed the aggregate of animals designated for slaughter within the analyzed states and time frames against the slaughter information from the Brazilian Institute of Geography and Statistics (IBGE) [29]. The GTA records tallied a sum of 57,181,510 heads for slaughter, whereas the IBGE data indicated 59,197,746 heads—yielding a correspondence rate of 97%.
Next, we downloaded 381,269 distinct rural property boundaries from each state’s environmental agency and two federal systems, the System of Rural Environmental Registry-SICAR and Land Management System-SIGEF (both acronyms in Portuguese) across Pará, Mato Grosso, and Rondônia between 2013 and 2018. The distribution of these properties within the analyzed states unfolded as follows: Pará concentrated 173,584 properties (representing 45% out of 381,269), Mato Grosso had 112,949 properties (30%), and Rondônia featured 94,736 properties (25%).
In the third step, we linked the GTA and CAR properties by applying the matching methods described in [18,22,23], which involve matches based on common landowner and property information found in both databases (full methods described in Appendix A). We identified 439,469 unique properties based on the property and owner attributes in the downloaded GTA transactions [18,22,23]. From this total, we identified 149,491 unique GTA properties (34% of 439,469) associated with unique CAR properties, referred to here as matched properties (Figure 1). The matching rate was higher in Pará and Mato Grosso (50–55% and 45–50%) and lower in Rondônia (25%). The remaining 66% of the unique GTA properties had no match due to limitations in the quality of information available in the GTA and CAR databases.
The matched properties covered 40 Mha, representing 26% of the total area of rural property boundaries within Pará, Mato Grosso, and Rondônia. These matched properties moved 155 million heads, representing 67% of the total cattle volume (233 million heads) traded within our full set of GTA transactions (matched and unmatched). Almost one-third (45 of 155 million heads) of the cattle that originated in the CAR-GTA properties were sold for slaughter, which corresponded to 76% (of 59 million heads) of the total heads slaughtered according to IBGE.

2.3. Mapping the Slaughterhouse Supply Zones

We mapped the slaughterhouse supply zones by identifying the locations of their direct suppliers and the tier−1 indirect suppliers that sell to the direct suppliers. We define “supply zone” as the generalized polygon surrounding the suppliers of most of the cattle that are purchased by a slaughterhouse. Suppliers that are spatially dispersed and contribute minimally to the total cattle received by the slaughterhouse are not considered part of the slaughterhouse supply zone (see Figure S3). By defining these polygons, we identify the zone in which most potential suppliers to the slaughterhouse are likely to be located and, conversely, the remaining regions in which suppliers or important suppliers to the slaughterhouse are unlikely to be found.
We followed four main steps to identify the slaughterhouse supply zones. First, we selected a sample of slaughterhouses. We mapped only the slaughterhouse supply zones of those that slaughtered more than 1000 heads of cattle per year between 2013 and 2018 and had a sanitary inspection code. This threshold was determined because fewer than 1% of the analyzed slaughterhouses sell fewer than 1000 heads per year. This resulted in a sample of 142 slaughterhouses listed by GTA. Approximately half of these slaughterhouses (54%) were part of the Federal Inspection System (SIF), which allows for export to other states and countries. Another 38 slaughterhouses (27%) held licenses from the State Inspection System (SIE), meaning the meat could only be sold within the state of slaughter. The remaining 27 plants (19%) that were not listed under SIF or SIE were classified as “Others”.
We proceeded to select a sample of direct and tier−1 indirect suppliers, with direct suppliers being establishments that sold cattle for slaughter to the identified slaughterhouses between 2013 and 2018, and tier−1 indirect suppliers being those selling to direct suppliers for non-slaughter-related purposes in the same year. While the Brazilian cattle supply chain involves diverse roles, we classified properties based on their “highest” role in the chain (e.g., as direct or tier−1 indirect suppliers) to ensure clarity in our analysis and alignment with supply chain policies like CA, which operate at the property level. In this process, we considered only direct and tier−1 indirect suppliers who sold at least 16 heads of cattle in a single transaction per year, resulting in a total of 97,085 unique matched properties for the entire study period. We chose a threshold of 16 animals, the number required to fill a single cattle truck, to exclude suppliers who sold very small quantities of cattle [18].
The third step was to determine the maximum spatial autocorrelation among the suppliers, weighted by the volume of cattle moved in a year to identify the spatial clusters of suppliers. For this, we used the Incremental Spatial Autocorrelation (ISA) function available in the ArcGIS Pro software version 3.1.3 [30]. For example, to identify the maximum spatial autocorrelation of direct suppliers for a specific slaughterhouse X each year Y, we provided the ISA function with the locations of the direct supplier polygons, along with a field in the attribute table indicating the total cattle sent for slaughter at slaughterhouse X in year Y. The result of the ISA function was the distance in kilometers where the spatial autocorrelation is maximum.
The last step was to draw polygons representing the slaughterhouse supply zones based on the supplier locations and the distances calculated in the previous step. For this, we used the Aggregate Polygons method in ArcGIS Pro [31]. This function combines polygons that are within a specified distance of each other, forming new polygons. In this case, the specified distance was the result of step three. This process was repeated for all direct and tier−1 indirect suppliers of the sample slaughterhouses (n = 142).

2.4. Defining the CA and Non−CA Slaughterhouse Supply Zones

We then identified which slaughterhouse supply zones from Section 2.3 were already being monitored by CA between 2013 and 2018, by checking if they were on CA slaughterhouse lists compiled by [10]. We found 81 slaughterhouses that had signed the CA, representing 57% (out of 142) of all slaughterhouses in our sample; more slaughterhouses signed after our study period. A significant proportion of the CA slaughterhouses (59 out of 142, or 42%) were SIFs, accounting for 77% of the total SIF facilities in the states of Mato Grosso, Pará, and Rondônia. Additionally, seven CA slaughterhouses were classified as SIEs, representing 18% of the total SIE facilities. We then defined four main types of zones described in Table 1.

2.5. Characterizing the Slaughterhouses’ Supply Zones

We quantified a series of characteristics for each slaughterhouse supply zone to provide a comprehensive understanding of the geographic features, land use, deforestation, carbon emissions, and suppliers’ roles. Appendix B has further information on how we computed these characteristics.

2.6. Estimating the Potential Expansion of CA Slaughterhouse Supply Zones

We then investigated three pathways to expand the CA supply zone area: (1) CA slaughterhouses expand monitoring to include tier−1 indirect suppliers, (2) Non−CA slaughterhouses begin monitoring direct suppliers, and (3) Non−CA slaughterhouses begin monitoring both direct and tier−1 indirect suppliers. To do this, we first aggregated all CA direct supply zones onto a single map. Then, we incorporated the corresponding additional supply zones (i.e., tier−1 indirect supply zone for pathway 1) to consider the three potential pathways of expansion.

3. Results

3.1. Spatial Extent of Slaughterhouse Supply Zones between the Years 2013 and 2018

The land across Mato Grosso, Pará, and Rondônia is likely to be influenced by slaughterhouses, especially those in CA. In 2018, their direct and indirect supply zones spanned 170 Mha, which accounts for 71% of the total study area. A large portion of the supply zone area (94% of 170 Mha; or 160 Mha) was within the direct supply zone of the CA slaughterhouses, which have monitoring systems for direct suppliers (Figure 2). As shown in Figure 2, there was a considerable overlap (95%) between the direct supply zones of the CA slaughterhouses and zones not monitored by the CAs.
Our analysis identified a significant degree of proximity between direct suppliers and tier−1 indirect suppliers, for both CA and non−CA slaughterhouses. Approximately 80% of tier−1 indirect suppliers were located within 100 km of direct suppliers, with a median value of 21 km. Additionally, we observed that CA and non−CA slaughterhouses were, on average, within a radius of 39 km from each other. The greatest distance found between a pair of neighboring CA and non−CA slaughterhouses was 170 km, which is a little over four hours of travel time for a truck traveling at 40 km/h. These spatial characteristics promote competition between CA and non−CA slaughterhouses and facilitate the transfer of cattle between direct and tier−1 indirect suppliers.

3.2. CA’s Slaughterhouse Supply Zones Characteristics

3.2.1. Spatial Stability between 2013 and 2018

The coverage of the CA direct supply zone was relatively stable between 2013 and 2018 (Figure 3). We observed that 96% of the total extent corresponded to locations that remained the same for two or more consecutive years. Of these, 71% of the area persisted unchanged for all years. Approximately 25% of the CA direct supply zone persisted between two to five years, while 4% appeared only in one year. In Mato Grosso, 82% of the CA direct supply zone persisted all six years, 16% between two to five years, and only 2% were included in only one year. In Pará, these levels were 63%, 29%, and 8%, respectively. Rondônia was the state with the lowest persistence over time, with half of that state’s CA direct supply zone persisting between two and five years, 46% in all six years, and 3% included in just one year.

3.2.2. Land-Use Characteristics

We observed that practically all the pasture in our study region (99% of 49 Mha) was within CA direct supply zones, though most of this area overlapped with other zone types. Indeed, supply zones from non-monitored suppliers (non−CA direct and tier−1 indirect suppliers, as well as CA tier−1 indirect suppliers) that did not intersect with the CA direct supply zone accounted for less than 1% of the pasture. The concentration of pasture in CA direct supply zones was almost 100% in Mato Grosso, 99% in Rondônia, and 96% in Pará.
In addition, 86% of natural vegetation (90% covered by forest vegetation) situated outside of protected areas (Conservation Units, Indigenous Lands), and Military Areas in 2018 fell within CA direct supply zones. In Mato Grosso, Rondônia, and Pará, this proportion was 98%, 90%, and 71%, respectively. Here again, we observed that 84% of this vegetation was also included inside other types of zones (CA tier−1 indirect, non−CA direct, or indirect). In Mato Grosso, Rondônia, and Pará, this percentage was 94%, 88%, and 70%, respectively.
Likewise, the CA direct supply zone covered nearly all the deforested areas across the three states. We observed that approximately 99% (61 Mha) of the cumulative deforestation by 2007 and 97% (7 Mha) of the yearly deforestation between 2008 and 2018 was within the CA supply zones. Finally, we observed that the median deforestation within the CA supply zones was 0.04 Mha between 2008 and 2018 (Figure S2). The associated CO2 emissions from this deforestation amounted to 12.45 MtCO2e per year.

3.2.3. Suppliers’ Characteristics

The 39,439 CA direct suppliers used to define the slaughterhouses’ supply zones covered 17% (28 Mha of 160 Mha) of the total zone area and 70% of the total GTA properties area. The remaining 83% (132 Mha of 160 Mha) was composed of 120 Mha (75% of 160 Mha) of area not within our GTA properties sample, and 12 Mha of other GTA properties (8% of 160 Mha).
CA direct suppliers have multiple roles in the supply chain. We found that 12% of total cattle moved by CA direct suppliers are slaughtered in CA slaughterhouses, 7% are slaughtered in non−CA slaughterhouses, and 78% are sent to other properties for purposes other than slaughter. Based on medians, we discovered that direct suppliers from SIF slaughterhouses delivered around 11% to 17% of their cattle for slaughter to a CA slaughterhouse, while they sent only 4% to 11% of their animals to non−CA slaughterhouses. CA direct suppliers to state and other smaller slaughterhouses similarly sent the bulk of their cattle (more than 90%) for non-slaughter transactions.
Finally, most deforestation within CA direct supply zones was outside the CA direct suppliers’ properties. Two-thirds (66% of 61 Mha) of the deforestation accumulated by 2007 occurred outside of our GTA-CAR sample, with 24% (15 Mha) on CA direct suppliers, and 10% (6 Mha) on other GTA properties. Similarly, the non-GTA-CAR areas accounted for much of the deforestation from 2008 to 2018 (63% or 5 Mha of 8 Mha), with CA direct suppliers accounting for 13% (1 Mha) and other GTA properties contributing to just over 1 Mha, with another 1 Mha found on CA indirect suppliers.

3.3. Expanding Pathways of CA Slaughterhouse Supply Zones

The spatial distribution of slaughterhouses and of direct and indirect suppliers implies that extending the CA to non−CA slaughterhouses or to tier−1 indirect suppliers would have a negligible effect on the area of CA direct supply zones. As previously reported, there was a substantial overlap (95%) between the accumulated area of CA direct supply zones and the non−CA supply zones. Incorporating all non−CA slaughterhouses in the agreements would have only increased the CA supply zones by 3 Mha. The magnitude of this increase would have varied by state, with Pará experiencing the highest increase at 14%, followed by Rondônia at 7%, and Mato Grosso at a marginal 1%. Likewise, adding only the zones of CA tier−1 indirect suppliers to the CA direct supply zones would have expanded the reach of CA slaughterhouses by 5 across all states combined, with the increase being more pronounced at the state level in Pará (11%), and more modest in Mato Grosso (2%) and Rondônia (<1%).
In addition, expanding the CAs to include additional slaughterhouses could have a modest impact when considering the coverage at the property level. The inclusion of all non−CA slaughterhouses and their direct suppliers into the CA direct supplier’s area would result in a 7% increase in the area covered by properties subject to the CA. However, expanding the agreements to encompass tier−1 indirect suppliers of CA would increase the monitored properties area by 34%, as estimated from GTA data. If all slaughterhouses monitored their direct and tier−1 indirect suppliers, the total increase would be 42% over the property area that was monitored by the current signatories of the CA.
While our findings indicate that expanding the CA would have a negligible effect on the area covered by CA direct supply zones, it’s worth noting that expanding it to non−CA slaughterhouses or their tier−1 indirect suppliers could significantly boost the CA coverage of the overall cattle supply. CA direct suppliers sold 48% of the total slaughter volume to slaughterhouses and received 12% of total cattle moved between properties in our study period between 2013 and 2018. Federally inspected non−CA slaughterhouses along with CA slaughterhouses, slaughtered 84% of cattle in our study area and period; these cattle were sold to slaughter from properties that purchased 21% of all the cattle traded between properties. Thus, the expansion of the CA to all remaining federally inspected slaughterhouses in the region represents an opportunity to significantly increase the influence of the CA. Considering all state-level slaughterhouses raises the slaughter volume further, to 90%, and the links to the total number of cattle moved to 22%. Direct suppliers overall received 25% of the total cattle moved between properties.
Expanding the CA to include monitoring of the properties that sell cattle to direct suppliers (tier−1 indirect suppliers) would increase their coverage of the cattle supply chain overall. The inclusion of tier−1 indirect suppliers to CA slaughterhouses in our study zone and region would have increased their coverage to 29% of the total cattle moved between properties. The inclusion of tier−1 indirect suppliers to all federally inspected slaughterhouses would have covered 54% of cattle moved between properties. Expansion to all state-inspected slaughterhouses would further increase this coverage to 63% of animals moved between properties.

4. Discussion

4.1. Our Supply Zones Are More Narrow and Precise than Previous Estimates

Our delimitation of supply zones, based on registered cattle transactions, produced more precise and smaller zones than previous work. We found that the buffers approach, which is the most widely used approach [14,16,18,22], tended to produce larger slaughterhouse zones when compared to our method—by 34% in total area of all zones combined and 33% per zone (median). The cost-distance approach from Barreto and colleagues [10] also overestimated zone boundaries compared to our results, by including regions covered by rivers and without cattle suppliers, leading to a 12% larger median area of each zone. Policies and studies using more generalized approaches should be aware of the potential overestimation of the area involved.
Our approach is a significant methodological advancement in utilizing GTA traceability data to estimate cattle supply chains and assess the associated risk of deforestation. The method we have developed has the potential for broader application in other states across the country where cattle transactions and property boundary data are available. One of the major achievements of our approach is demonstrating that slaughterhouses and other stakeholders, such as the leather sector, can assess the overall risk within their entire supply chains by mapping only slaughterhouses’ direct suppliers. By utilizing this information alone, they can now comprehensively evaluate the overall risk encompassing their entire supply chain. This accomplishment highlights the clear and effective mapping of the full supply chain, which can provide valuable insights for decision-making and risk-management strategies.

4.2. High Overlap between CA and Non−CA Supply Zones May Stimulate Competition and Laundering

Our findings show a high degree of overlap (95%) between CA direct supply zones and other zone types (CA tier−1 indirect suppliers and non−CA direct and tier−1 indirect suppliers). Although this result is generally consistent with other studies that found a large overlap between zones of slaughterhouses that signed the agreements and zones of slaughterhouses that did not sign [10,16,17,32], our study is the first to find overlap between direct and indirect supplier zones of slaughterhouses that signed and did not sign the agreements. These findings provide fresh knowledge of the possible risk of competition and cattle laundering operations, which limit the efficiency of the agreements in reducing deforestation.
Competition with and leakage to non−CA slaughterhouses and cattle laundering are activities that may occur in isolation or in tandem. Competition often arises when CA slaughterhouses cannot locate enough cattle from complying suppliers within their supply zones to satisfy their demand due to suppliers selling cattle to non−CA slaughterhouses, potentially resulting in a cattle shortfall [17]. Cattle laundering may occur when non-compliant tier−1 indirect suppliers can sell to compliant CA direct suppliers without triggering the CA slaughterhouse’s monitoring systems [11,33]. Leakage may also happen when deforestation from a CA direct supplier moves to non−CA suppliers or to indirect suppliers [17].
Expanding the CA to include additional non−CA slaughterhouses would help mitigate these risks by lowering the economic burden incurred by CA slaughterhouses due to higher buying regulations, as well as limiting the potential for competition and laundering [9,10,11,19]. Without an expansion of CA to more slaughterhouses and more suppliers, we believe there is an imminent risk that cattle suppliers will resort to deforestation to produce cattle. Increased competition and cattle laundering will likely result in CA slaughterhouses being forced to buy cattle coming from deforested farms, which could further decrease the effectiveness of CA in forest conservation.
New alternatives have been discussed to expand CA monitoring to more suppliers and more slaughterhouses. Recently, the MPF has strengthened its partnership with non−governmental agencies and universities to harmonize the methods of analysis and audit of CA slaughterhouses [12]. In addition, new projects have been launched to help slaughterhouses understand the deforestation contamination level of their supply chain. Among these projects, we highlight a few: SeloVerde, Visipec, Beef on Track, and Trase. SeloVerde, launched in the state of Pará, is the first public tool in Brazil with the objective of providing information on compliance with the Forest Code and estimating the contamination of the cattle supply chain [34]. Visipec, is a tool developed in partnership with the National Wildlife Federation (NWF) and the University of Wisconsin-Madison to help the cattle industry estimate the contamination with deforestation of indirect suppliers [35]. It also includes ways to incorporate indirect suppliers in the monitoring based on Best Practices criteria, validated by industry representatives and indirect suppliers [36]. Finally, Beef on Track and Trase have been helping increase the transparency of information regarding the cattle sector at the Brazilian Amazon and Global scales, respectively [12,37]. As these projects are in their early years of implementation in the Brazilian cattle sector, it is still too early to estimate their effectiveness in curbing deforestation and CA expansion. However, strategies such as these have the potential to accelerate the expansion of CAs through social or market pressure created by the information arising from these instruments.

4.3. CA Slaughterhouses’ Direct Supply Zones Are Stable Regions That Concentrate Much of the Unprotected Vegetation

Our results suggest that the supply zones of CA slaughterhouses covered a significant portion of the states of Pará, Mato Grosso, and Rondônia. The CA direct supply zones are stable over time and individually comparable in size to large Amazonian municipalities (Figure S1). This means that the companies and managers of each CA slaughterhouse could potentially influence land use decision-making over very large swaths of land. Moreover, these areas contain virtually all vegetation outside Indigenous Lands, Conservation Units, and Military Areas within the CA direct supply zones. Given that the decision-making regarding deforestation ultimately rests with landowners such as cattle ranchers, which is influenced by CA slaughterhouses’ sourcing criteria, we posit that these slaughterhouses have the potential to impact vegetation regions beyond their direct suppliers [38].
It is also worth noting that areas of natural vegetation outside the reach of CA slaughterhouses could be designated for other purposes, such as the creation of new protected areas or limited-use areas, without affecting cattle production in the states studied. Several studies have shown that Protected Areas are an effective way to contain deforestation in the Amazon [39], while new investments in intensified cattle ranching can expand cattle production without new deforestation in the Amazon [40].

5. Conclusions

Our results have demonstrated the remarkable potential that meatpacking companies hold to influence the fate of the Amazon. We found that CA and non-CA slaughterhouse supply zones combined cover vast areas, including most unprotected land. Current CA supply zones alone encompass 67% of the total area included in the states of Pará, Mato Grosso, and Rondônia (160 Mha), consisting predominantly of pasture and deforested lands. Additionally, we found the large size of the supply zones and proximity between slaughterhouses lead to a 95% overlap between CA and non-CA supply zones, which probably results in increased competition for suppliers. This makes it harder for CA slaughterhouses’ blockage of non-compliant suppliers to be impactful, as those suppliers can easily sell to non-CA slaughterhouses instead. Expanding the CA to all slaughterhouses would help reduce the competition and increase the pressure on producers to avoid deforestation, even if the geographical area directly influenced does not expand significantly. Similarly, monitoring indirect suppliers would not expand the area covered substantially, but would provide a leap forward in helping to reduce deforestation [11,16,18,19,21]. The geographical proximity of direct and indirect suppliers enables a more comprehensive risk assessment across Brazil, and our methodological advancements in accurately mapping supply zones equip stakeholders, such as slaughterhouses, with tools to thoroughly evaluate deforestation risk within their supply chains.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/land12091782/s1, Figure S1: Boxplot showing the supply zone areas per health inspection type, per CA, per state for direct (red) and tier-1 indirect (blue) suppliers.; Figure S2: Boxplot showing the annual deforestation (A) and associated carbon emissions (B) within the CA supply zones.; Figure S3: Example of a supply zone drawn from the direct suppliers of a single slaughterhouse; Table S1: Total number of cattle heads moved for various purposes using GTAs in Pará, Mato Grosso, and Rondônia, for the period from 2013 to 2018.

Author Contributions

Conceptualization, A.B.J.; methodology, A.B.J.; formal analysis, A.B.J. and J.M.; writing—original draft preparation, A.B.J.; writing—review and editing, A.B.J., H.K.G. and L.R.; visualization, A.B.J.; supervision, H.K.G.; funding acquisition, H.K.G. and L.R. All authors have read and agreed to the published version of the manuscript.

Funding

This study was made possible by the Gordon and Betty Moore Foundation 2106-036 and Norway’s International Climate and Forest Initiative 2110-118. The contents of this study are the sole responsibility of the University of Wisconsin Madison and do not necessarily reflect the of the funders.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. The data are not publicly available due to data privacy restrictions.

Acknowledgments

We would also like to thank Malena Candino for reviewing various versions of the paper.

Conflicts of Interest

L.R., J.M. and H.K.G. have an ongoing consulting relationship with the National Wildlife Federation, which is also a partner on the projects supported by the funding sources listed above and on the development of the Visipec cattle traceability tool. The National Wildlife Federation did not provide editorial oversight over the contents of this manuscript.

Appendix A. Mapping Cattle Supply Chains by Linking GTA and Property Records

Our research team has developed a precise method to match GTA and CAR based on the methods from [18,22,23] and data from different sources (Table A1). We have fine-tuned coding rules, informed by extensive testing and the expertise of our team and specialists, to pair matching fields in records. When specific criteria are met (see Table A2 for details), two records are considered a match, and they receive a unique “property ID” indicating they belong to the same property.
It is essential to note that our method differs from typical matching systems that allow for variations (fuzzy matching). Instead, we use strict matching criteria that we have refined over time. This approach ensures accurate matches while minimizing errors. In groups of properties with CAR records (“matched properties”), we correlate spatial and land-use data, including deforestation data from PRODES and information on slaughterhouses. After matching, we organize the data into structured tables in the PostgresSQL database, allowing for various types of analysis while preserving data traceability [23].
Table A1. Datasets used to match the GTA and CAR data.
Table A1. Datasets used to match the GTA and CAR data.
DataSourcePeriodLink to Access the Data
(Accessed Period: 1 January 2018 to 31 March 2019)
GTAPará2013–2018http://www.adepara.pa.gov.br
https://www.indea.mt.gov.br
Mato Grosso
Rondônia http://www.idaron.ro.gov.br
CARPará2013–2018http://car.semas.pa.gov.br
Mato Grosso https://geoportal.sema.mt.gov.br
Rondônia http://car.sedam.ro.gov.br
Table A2. Rules for matching records in GTA and property databases. Source West et al. [23].
Table A2. Rules for matching records in GTA and property databases. Source West et al. [23].
Rule DescriptionNotes
Match on property name, municipality, and CPF.CPF is the Brazilian Tax Identification Number for Individuals.
Match on property name, municipality, and CNPJ.CNPJ is the Brazilian Tax Identification Number for companies.
Match on property identifier, municipality, and similar property name.Examples of property identifiers: Código do estabelecimento (GTA), SICAR number, state CAR number, Código do imóvel (INCRA), etc.
Match on municipality and property boundary.
Match on property name, municipality, and owner name.Applied when the CNPJ number is missing.
Match on property name, municipality, and owner name under different circumstances.Applied when the CNPJ number is available.
Match on municipality, person/company, and “property name containment”.“Property name containment”: when one name contains the other as a substring. Example: BOA VISTA contains the name VISTA
Match on property name, person/company, and neighboring municipality.
Match on person/company, municipality, and property name with numerals removed.
Match on person/company, neighboring property boundary, and property name with numerals removed.
Match on property name and coordinates falling within the property boundary.
Match on property name, municipality, and coordinates.
From the property maps, match on the best GTA property with a similar owner name using a random forest classifier.In addition to the CAR, the matching also included the INCRA and Terra Legal databases. Together, we refer to sources that provide property boundaries as property maps.
From the GTA, match on the best property with pasture with similar owner name using a random-forest classifier.The random forest is trained on familiar attributes like municipality, property name, etc.
From the MT-GTA, match on the best MT-CAR property with a similar owner name using a random-forest classifier.Mato Grosso state (MT).
From the GTA, match RO properties to the CAR on municipality, owner, and similar address information.Rondônia state (RO).
Address information extracted: “LINHA”, “GLEBA”, “LOTE”, “BR”, and “KM”.
Match on CAR number if the geometries are within 1 km of each other.
For CAR records without a geometry, match on CAR number and most similar property name to records with a geometry.
For GTA records that list a CAR number in their notes, match on CAR number and most similar property name.
For CAR records that only provide a geometry, match to JBS traceability data on intersecting coordinates.More information available at:
https://jbs.com.br/en/press/releases-en/blockchain-platform-developed-by-jbs-begins-operation/ (Accessed on 15 February 2019)

Appendix B. Supply Zones Characteristics

Below we provide a series of characteristics that we calculated for each supply zone from this study. The data used here are organized in Table A3. First, we classified the zones based on the health inspection classification (SIF, SIE, Others), agreements (CA or non−CA), and supplier type (direct or tier−1 indirect). Second, we calculated geographical characteristics such as the area in hectares and the overlap with other zones. We computed a set of summary statistics based on the zone’s areas (average, median, and range) and how they vary at the state level. In addition, we calculated the cumulative area of each zone by year and the overlap of these cumulative areas over the years 2013 to 2018. We further analyzed the overlap of the cumulative areas of CA and non−CA zones for the entire period (2013–2018).
Third, we assessed the land use based on Mapbiomas collection 6 data from [41]. We started by reclassifying the original 30+ land-use classes from Mapbiomas for the year 2018 into four main classes: (i) pasture, (ii) natural vegetation (forest or non-forest vegetation); (iii) soybean, (iv) and others. The output here was the total zone percentage and total area in Mha covered by each of the land uses for 2018.
The following characteristics were deforestation and associated carbon emissions. We first calculated deforestation using PRODES data for the Amazon and Cerrado biomes [42]. The PRODES measure the clear-cutting of forest areas. We excluded deforestation below 6.25 hectares from the PRODES Amazon and one hectare from the PRODES Cerrado since these figures are the minimum mapping unit according to their methodologies. Zones within the Pantanal biome were excluded from our assessment since the PRODES Pantanal is still preliminary. The PRODES data were grouped into one category for deforestation through 2007 and a second category for annual deforestation between 2008 and 2018. To quantify carbon emissions from deforestation, we overlaid deforestation data on the above and below-ground biomass carbon per hectare map from [43] to calculate the total carbon dioxide equivalent (CO2) emission in mega (106) metric tons (MtCO2e) using the committed flow method outlined in [40].
Table A3. Datasets used to assess the supply zones’ characteristics.
Table A3. Datasets used to assess the supply zones’ characteristics.
DataSourcePeriodLink to Access the Data
DeforestationPRODES AmazonAccumulated deforestation 2007 and annual deforestation 2008–2018http://terrabrasilis.dpi.inpe.br/en/download-2/
(Accessed on 20 February 2023)
PRODES Cerrado
Land-useMapbiomas collection 6.02013 and 2018https://mapbiomas.org/en/download
(Accessed on 20 February 2023)
Carbon stockSoto-Navarro et al. (2020) [43]2010https://developers.google.com/earth-engine/datasets/catalog/WCMC_biomass_carbon_density_v1_0
(Accessed on 20 February 2023)

References

  1. IBGE. Pesquisa Da Pecuária Municipal (PPM). Available online: https://www.ibge.gov.br/estatisticas/economicas/agricultura-e-pecuaria/9107-producao-da-pecuaria-municipal.html (accessed on 24 July 2023).
  2. USDA. ERS—Sector at a Glance. Available online: https://www.ers.usda.gov/topics/animal-products/cattle-beef/sector-at-a-glance/ (accessed on 25 July 2023).
  3. Mapbiomas MapBiomas Project—Collection 7 of the Annual Series of Land Use and Land Cover Maps of Brazil. Available online: https://mapbiomas.org/en/terms-of-use (accessed on 25 July 2023).
  4. Lambin, E.F.; Gibbs, H.K.; Heilmayr, R.; Carlson, K.M.; Fleck, L.C.; Garrett, R.D.; Le Polain De Waroux, Y.; McDermott, C.L.; McLaughlin, D.; Newton, P.; et al. The Role of Supply-Chain Initiatives in Reducing Deforestation. Nat. Clim. Chang. 2018, 8, 109–116. [Google Scholar] [CrossRef]
  5. Nepstad, D.; Soares-Filho, B.S.; Merry, F.; Lima, A.; Moutinho, P.; Carter, J.; Bowman, M.; Cattanec, A.; Rodrigues, H.; Schwartzman, S.; et al. The End of Deforestation in the Brazilian Amazon. Source Sci. New Ser. 2009, 326, 1350–1351. [Google Scholar] [CrossRef] [PubMed]
  6. Nepstad, D.; McGrath, D.; Stickler, C.; Alencar, A.; Azevedo, A.; Swette, B.; Bezerra, T.; DiGiano, M.; Shimada, J.; Da Motta, R.S.; et al. Slowing Amazon Deforestation through Public Policy and Interventions in Beef and Soy Supply Chains. Science 2014, 344, 1118–1123. [Google Scholar] [CrossRef] [PubMed]
  7. Heilmayr, R.; Rausch, L.L.; Munger, J.; Gibbs, H.K. Brazil’s Amazon Soy Moratorium Reduced Deforestation. Nature food 2020, 1, 801–810. [Google Scholar] [CrossRef] [PubMed]
  8. Gibbs, H.K.; Rausch, L.; Munger, J.; Schelly, I.; Morton, D.C.; Noojipady, P.; Soares-Filho, B.; Barreto, P.; Micol, L.; Walker, N.F. Brazil’s Soy Moratorium: Supply-Chain Governance Is Needed to Avoid Deforestation. Science 2015, 347, 377–378. [Google Scholar] [CrossRef]
  9. Pereira, R.; Rausch, L.L.; Carrara, A.; Gibbs, H.K. Extensive Production Practices and Incomplete Implementation Hinder Brazil’s Zero-Deforestation Cattle Agreements in Pará. Trop. Conserv. Sci. 2020, 13, 1940082920942014. [Google Scholar] [CrossRef]
  10. Barreto, P.; Pereira, R.; Brandão, A., Jr.; Baima, S. Will Meat-Packing Plants Help Halt Deforestation in the Amazon? Imazon and Instituto Centro de Vida: Belém, PA, USA, 2017. [Google Scholar]
  11. Gibbs, H.K.; Munger, J.; L’Roe, J.; Barreto, P.; Pereira, R.; Christie, M.; Amaral, T.; Walker, N.F. Did Ranchers and Slaughterhouses Respond to Zero-Deforestation Agreements in the Brazilian Amazon? Conserv. Lett. 2016, 9, 32–42. [Google Scholar] [CrossRef]
  12. Imaflora; MPF. Protocolo de Monitoramento de Fornecedores de Gado da Amazônia. 2021. Available online: https://www.boinalinha.org/wp-content/uploads/2022/08/Protocolo-Monitoramento-Gado_A4_PORTUGUES_1.1_ALT4.pdf (accessed on 2 May 2023).
  13. Greenpeace. Slaughtering the Amazon. 2009. Available online: https://www.greenpeace.org/usa/wp-content/uploads/legacy/Global/usa/planet3/PDFs/slaughtering-the-amazon-part-1.pdf (accessed on 15 December 2022).
  14. Alix-Garcia, J.; Gibbs, H.K. Forest Conservation Effects of Brazil’s Zero Deforestation Cattle Agreements Undermined by Leakage. Glob. Environ. Chang. 2017, 47, 201–217. [Google Scholar] [CrossRef]
  15. Skidmore, M.E.; Sims, K.M.; Rausch, L.L.; Gibbs, H.K. Sustainable Intensification in the Brazilian Cattle Industry: The Role for Reduced Slaughter Age. Environ. Res. Lett. 2022, 17, 064026. [Google Scholar] [CrossRef]
  16. Levy, S.A.; Cammelli, F.; Munger, J.; Gibbs, H.K.; Garrett, R.D. Deforestation in the Brazilian Amazon Could Be Halved by Scaling up the Implementation of Zero-Deforestation Cattle Commitments. Glob. Environ. Chang. 2023, 80, 102671. [Google Scholar] [CrossRef]
  17. Moffette, F. Forest Conservation and Agriculture: Essays on Land Use Trade-Offs and Spillovers. Ph.D. Thesis, University of Wisconsin Madison, Madison, WI, USA, 2018. [Google Scholar]
  18. Skidmore, M.E.; Moffette, F.; Rausch, L.; Christie, M.; Munger, J.; Gibbs, H.K. Cattle Ranchers and Deforestation in the Brazilian Amazon: Production, Location, and Policies. Glob. Environ. Chang. 2021, 68, 102280. [Google Scholar] [CrossRef]
  19. Rajão, R.; Soares-Filho, B.; Nunes, F.; Börner, J.; Machado, L.; Assis, D.; Oliveira, A.; Pinto, L.; Ribeiro, V.; Rausch, L.; et al. The Rotten Apples of Brazil’s Agribusiness. Science 2020, 369, 246–248. [Google Scholar] [CrossRef] [PubMed]
  20. Carvalho, W.D.; Mustin, K.; Hilário, R.R.; Vasconcelos, I.M.; Eilers, V.; Fearnside, P.M. Deforestation Control in the Brazilian Amazon: A Conservation Struggle Being Lost as Agreements and Regulations Are Subverted and Bypassed. Perspect. Ecol. Conserv. 2019, 17, 122–130. [Google Scholar] [CrossRef]
  21. Klingler, M.; Richards, P.D.; Ossner, R. Cattle Vaccination Records Question the Impact of Recent Zero-Deforestation Agreements in the Amazon. Reg. Environ. Chang. 2018, 18, 33–46. [Google Scholar] [CrossRef]
  22. Skidmore, M.; Moffette, F.; Rausch, L.; Gibbs, H. Characterizing Compliance in Cattle Supply Chains: What Factors Encourage Deforestation-Free Production in the Brazilian Amazon?; 2020. Available online: https://www.evidensia.eco/resources/937/download/ (accessed on 2 June 2022).
  23. West, T.A.P.; Rausch, L.; Munger, J.; Gibbs, H.K. Protected Areas Still Used to Produce Brazil’s Cattle. Conserv. Lett. 2022, 15, e12916. [Google Scholar] [CrossRef]
  24. Slob, B.; Rijk, G.; Piotrowski, M. JBS, Marfrig, and Minerva: Material Financial Risk from Deforestation in Beef Supply Chains. 2020. Available online: https://chainreactionresearch.com/report/jbs-marfrig-and-minerva-material-financial-risk-from-deforestation-in-beef-supply-chains/ (accessed on 2 June 2022).
  25. Proforest. Socio-Environmental Monitoring of the Cattle Sector in Brazil. 2017. Available online: https://www.proforest.net/fileadmin/uploads/proforest/Documents/Publications/bn09_eng_final_web.pdf (accessed on 15 July 2022).
  26. MAPA. Habilitar-Se Para Emissão Da Guia de Trânsito Animal. Available online: https://www.gov.br/pt-br/servicos/habilitar-se-para-emissao-da-guia-de-transito-animal (accessed on 25 July 2023).
  27. SICAR. Sicar—Sistema Nacional de Cadastro Ambiental Rural. Available online: https://www.car.gov.br/#/sobre (accessed on 25 July 2023).
  28. Santos, A.B.; Costa, M.H. Do Large Slaughterhouses Promote Sustainable Intensification of Cattle Ranching in Amazonia and the Cerrado? Sustainability 2018, 10, 3266. [Google Scholar] [CrossRef]
  29. IBGE. Pesquisa Trimestral do Abate de Animais. Available online: https://www.ibge.gov.br/estatisticas/economicas/agricultura-e-pecuaria/9203-pesquisas-trimestrais-do-abate-de-animais.html (accessed on 29 August 2023).
  30. ESRI. Incremental Spatial Autocorrelation (Spatial Statistics)—ArcGIS Pro|Documentation. Available online: https://pro.arcgis.com/en/pro-app/latest/tool-reference/spatial-statistics/incremental-spatial-autocorrelation.htm (accessed on 25 July 2023).
  31. ESRI. Aggregate Polygons (Cartography)—ArcGIS Pro|Documentation. Available online: https://pro.arcgis.com/en/pro-app/latest/tool-reference/cartography/aggregate-polygons.htm (accessed on 25 July 2023).
  32. Alix-Garcia, J.; Rausch, L.L.; L’Roe, J.; Gibbs, H.K.; Munger, J. Avoided Deforestation Linked to Environmental Registration of Properties in the Brazilian Amazon. Conserv. Lett. 2018, 11, e12414. [Google Scholar] [CrossRef]
  33. Carvalho, R.; Rausch, L.; Munger, J.; Gibbs, H.K. The Role of High-Volume Ranches as Cattle Suppliers: Supply Chain Connections and Cattle Production in Mato Grosso. Land 2021, 10, 1098. [Google Scholar] [CrossRef]
  34. ASCOM. SEMAS—Governo Do Pará Apresenta “Selo Verde” a Membros do Parlamento Europeu, Em Belém. Available online: https://www.semas.pa.gov.br/2023/06/22/governo-do-para-apresenta-selo-verde-a-membros-do-parlamento-europeu-em-belem/ (accessed on 25 July 2023).
  35. VISIPEC. VISIPEC—This Website Provides Information on Visipec, a New Tool to Enhance Traceability and Strengthen Deforestation Monitoring in the Brazilian Cattle Sector. Available online: https://www.visipec.com/ (accessed on 25 July 2023).
  36. GTFI. Good Practices—GTFI. Available online: https://gtfi.org.br/good-practices/ (accessed on 25 July 2023).
  37. TRASE. Trase Homepage. Available online: https://www.trase.earth/ (accessed on 25 July 2023).
  38. Borgatti, S.P.; Li, X. On social network analysis in a supply chain context. J. Supply Chain. Manag. 2009, 45, 5–22. [Google Scholar] [CrossRef]
  39. Soares-Filho, B.; Moutinho, P.; Nepstad, D.; Anderson, A.; Rodrigues, H.; Garcia, R.; Dietzsch, L.; Merry, F.; Bowman, M.; Hissa, L.; et al. Role of Brazilian Amazon Protected Areas in Climate Change Mitigation. Proc. Natl. Acad. Sci. USA 2010, 107, 10821–10826. [Google Scholar] [CrossRef]
  40. Brandão, A., Jr.; Rausch, L.; Paz Durán, A.; Costa, C., Jr.; Spawn, S.A.; Gibbs, H.K. Estimating the Potential for Conservation and Farming in the Amazon and Cerrado under Four Policy Scenarios. Sustainability 2020, 12, 1277. [Google Scholar] [CrossRef]
  41. Souza, C.M.; Shimbo, J.Z.; Rosa, M.R.; Parente, L.L.; Alencar, A.A.; Rudorff, B.F.T.; Hasenack, H.; Matsumoto, M.; Ferreira, L.G.; Souza-Filho, P.W.M.; et al. Reconstructing Three Decades of Land Use and Land Cover Changes in Brazilian Biomes with Landsat Archive and Earth Engine. Remote Sens. 2020, 12, 2735. [Google Scholar] [CrossRef]
  42. INPE. Terrabrasilis—Geographic Data Platform. Available online: http://terrabrasilis.dpi.inpe.br/en/home-page/ (accessed on 31 August 2023).
  43. Soto-Navarro, C.; Ravilious, C.; Arnell, A.; De Lamo, X.; Harfoot, M.; Hill, S.L.L.; Wearn, O.R.; Santoro, M.; Bouvet, A.; Mermoz, S.; et al. Mapping Co-Benefits for Carbon Storage and Biodiversity to Inform Conservation Policy and Action. Philos. Trans. R. Soc. B Biol. Sci. 2020, 375, 20190128. [Google Scholar] [CrossRef] [PubMed]
Land 12 01782 g001
Figure 1. Slaughterhouses and rural properties connected to cattle transit records in 2018.
Figure 1. Slaughterhouses and rural properties connected to cattle transit records in 2018.
Land 12 01782 g001
Land 12 01782 g002
Figure 2. Spatial overlap of all slaughterhouse supply.
Figure 2. Spatial overlap of all slaughterhouse supply.
Land 12 01782 g002
Land 12 01782 g003
Figure 3. Persistence of coverage of CA supply zone over time from 2013 to 2018.
Figure 3. Persistence of coverage of CA supply zone over time from 2013 to 2018.
Land 12 01782 g003
Table 1. Slaughterhouse supply zone definitions.
Table 1. Slaughterhouse supply zone definitions.
Supply Zone TypesSuppliers Used to Define the Zone BoundaryZone Definition
CA direct supply zoneCA direct suppliersGeneralized polygon surrounding CA direct suppliers
CA tier−1 indirect supply zoneCA tier−1 indirect supplierGeneralized polygon surrounding CA tier−1 indirect suppliers
Non−CA direct supply zoneNon−CA direct suppliersGeneralized polygon surrounding non−CA direct suppliers
Non−CA tier−1 indirect supply zoneNon−CA tier−1 indirect supplierGeneralized polygon surrounding non−CA tier−1 suppliers
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Brandão Jr., A.; Rausch, L.; Munger, J.; Gibbs, H.K. Mapping Slaughterhouse Supply Zones in the Brazilian Amazon with Cattle Transit Records. Land 2023, 12, 1782. https://doi.org/10.3390/land12091782

AMA Style

Brandão Jr. A, Rausch L, Munger J, Gibbs HK. Mapping Slaughterhouse Supply Zones in the Brazilian Amazon with Cattle Transit Records. Land. 2023; 12(9):1782. https://doi.org/10.3390/land12091782

Chicago/Turabian Style

Brandão Jr., Amintas, Lisa Rausch, Jacob Munger, and Holly K. Gibbs. 2023. "Mapping Slaughterhouse Supply Zones in the Brazilian Amazon with Cattle Transit Records" Land 12, no. 9: 1782. https://doi.org/10.3390/land12091782

APA Style

Brandão Jr., A., Rausch, L., Munger, J., & Gibbs, H. K. (2023). Mapping Slaughterhouse Supply Zones in the Brazilian Amazon with Cattle Transit Records. Land, 12(9), 1782. https://doi.org/10.3390/land12091782

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop