Helen Bellfield

Newly published maps from Trase show the location, ownership and capacity of facilities such as cattle slaughterhouses in Brazil, palm oil mills in Indonesia and cocoa cooperatives in Côte d'Ivoire. The open-access data is freely available at our 

Demand for asset-level data in agricultural supply chains has boomed in response to regulations such as the EU Corporate Sustainability Reporting Directive (CSRD) and voluntary frameworks including the Taskforce on Nature-related Financial Disclosures (TNFD) that require companies and financial institutions to assess and disclose the impact of their activities on people and the environment, and associated risks and opportunities.

The location of physical assets such as slaughterhouses or mills is vital information for companies seeking to assess and manage environmental and social risks in their operations and value chains, and for financial institutions looking to understand risks in loan books and investment portfolios. This is because risks from environmental (e.g. deforestation) or social (e.g. land conflicts) impacts are frequently specific to a location.

For example, cattle slaughterhouses located near frontiers of deforestation may be more exposed to deforestation, cocoa cooperatives operating in certain areas may be more vulnerable to climate change impacts, and soy processing facilities operating in water-scarce areas may be more vulnerable to disruption.

Meeting the needs of companies and financial institutions

Companies need to map their supply chains back to sourcing regions in order to assess, mitigate and remediate environmental and social impacts. The most precise information would link supplies to individual farms, but farm-level traceability remains a challenge in some supply chains, especially for commodities where there are many intermediaries such as cattle, or smallholder producers such as cocoa. In that case, area-level information can also support company action.

Information on supply chain facilities can make a critical contribution. Facilities such as cooperatives, silos and mills are the first points of aggregation from producers and therefore provide an indication of the likely sourcing region around them. This enables companies to undertake area-level risk assessments of environmental and social impacts in that region, providing some level of granularity and specificity. Even where farm-level traceability is available, an area-level assessment around a facility can provide crucial additional information on risks such as water scarcity or the risks of mixing with products from other farms linked to environmental and social impacts. This is reflected in guidance by the 

OECD-FAO Responsible Agricultural Supply Chains

The ability of financial institutions to assess their exposure to environmental and social impacts through their lending and investments similarly requires them to 

 within the operations and value chains of their clients. This data is needed for due diligence, disclosure and reporting, credit-risk assessments and pricing, and new products such as sustainability linked loans and green bonds.

There are significant challenges in companies and financial institutions accessing asset-level data in agricultural supply chains that this new Trase data release helps to address:

At present, access to data is often limited because large proprietary databases, often owned by consultancies and trade data vendors, require payment to access. Existing open databases often cover only a subset of facilities. Trase helps to address this by providing and consolidating open access to supply chain facilities data for a set of key forest risk commodities. Open data brings significant advantages in bringing sector-wide transparency, enabling peer review and clarity on the accuracy and limitations of the underlying data, and for crowd-sourcing updates.

Existing data often has conflicting entries between government sources, corporate disclosures and crowd-sourced maps. The same facilities can be listed under multiple names across datasets. Verification methods are rarely published and in many cases coordinates are wrong or approximate (e.g. centroids of region or address of headquarters). Trase’s new data collects, consolidates and verifies publicly available information on assets from governments including 

 in Brazil, corporate and industry disclosures, researchers and mapping platforms such as Google Maps or OpenStreetMap.

There are a huge number of supply chain facilities; for instance, there are more than 9,000 soy silos in Brazil. In addition, there are thousands of small or remote facilities, such as on-farm silo bags or village bulking centres, that are often not recorded. Currently, there is often no data on ownership, capacity, licenses and buyers that can support supply chain mapping.

Trase’s initial release of facilities data includes slaughterhouses and meat processing facilities for Brazil (cattle, chickens etc), palm oil mills in Indonesia (integrating UML identifiers) and cocoa cooperatives in Côte d’Ivoire. Rapid progress in artificial intelligence is also enabling more rapid identification of unidentified facilities. We will be adding information on more commodities and countries, including soy silos in Brazil where we are collaborating with 

 on using artificial intelligence to improve identification and verification. Where possible Trase data will also include information on the ownership, capacity, and unique identifiers such as company tax numbers that enable users to connect and triangulate data linked to the facility. This approach follows the success of 

 that has been widely adopted by the palm oil sector in reporting and disclosing, and the 

 in the sugar sector. We will also add information on the supply sheds and associated deforestation risks for each facility.

Trase would like to acknowledge the hard work and expertise of the researchers and data scientists who developed the facilities datasets including: Jason Benedict, Robert Heilmayr, Kimberly Carlson, Ramada Febrian, Valentin Guye, Cécile Renier and the Do Pasto Ao Prato team (Erasmus zu Ermgassen, Andrea Garcia, Finn Mempel).