Proba NTree: an AI agent for traceable fertilizer emission data
Proba launches Proba NTree, an AI agent that gives food companies and fertilizer producers traceable fertilizer
Press Release Disclaimer: This is a press release distributed through the XPR Media network. It has not been independently verified by our newsroom.

![]()
Proba launches Proba NTree, an AI agent that gives food companies and fertilizer producers traceable fertilizer emission numbers for supply chain accounting
AMSTERDAM, NETHERLANDS, October 12, 2026 /EINPresswire.com/ — Proba has launched Proba NTree, an emission accounting AI tool for fertilizer emissions in agri-food supply chains, which is built for Scope 3, SBTi FLAG and CSRD reporting, NTree replaces generic emission factors with values matched to the fertilizer product, where it was made and where it is applied. Fertilizer accounts for a significant share of emissions in many food companies’ Scope 3 inventories, but most companies report using generic emission factors. Actual emissions vary with the fertilizer product, the production site and local field conditions. With generic factors, switching to a lower-carbon fertilizer, changing supplier or reducing nitrogen rates often makes no difference to the reported number, and auditors can’t easily check which method was used.
Who NTree is for
Food and beverage companies and retailers can use NTree to calculate fertilizer emissions for the crops they source, set a defensible baseline for SBTi FLAG targets and CSRD reporting, and test which sourcing or farming changes will actually reduce their reported emissions.
Fertilizer producers and distributors can show customers product-specific emission numbers and quantify the difference between low-carbon or inhibitor-treated products and conventional alternatives, using methods their customers’ auditors can trace.
Cooperatives, traders and sustainability advisors can process a full sourcing list in one run and compare scenarios, such as a different product, origin or nitrogen rate, before committing to a change.
How it works
The calculation works in two steps. An AI agent converts the user’s question into the inputs the calculation requires and asks a follow up question when information is missing. Proba’s decision tree selects one scientific method per case across four emission sources and uses only published values. The AI agent does not generate any numbers, so the same input will always produce the same result.
Why this is important
Everyone is adding AI to sustainability, and buyers are right to ask where the numbers come from. In NTree, the AI only asks the questions. Our methodology sets the rules, the decision tree does the calculation, and every number comes with its source. The same question always gets the same answer.
“Ask three experts for a fertilizer emission factor and you get four answers,” said Sijbrand Tieleman, CEO and Co-Founder of Proba. “Our team made those calls once, case by case, and wrote them down. Now any sustainability team can apply that expertise in minutes, and any auditor can check it.” said Sijbrand Tieleman, CEO and Co-Founder of Proba. “A Scope 3 number that doesn’t move when your growers change fertilizer tells you very little about your supply chain. We built NTree so that it does.”
Companies can request a half hour walkthrough of the Proba NTree with the Proba team via https://proba.earth/ntree-ai .
About Proba
Proba quantifies and certifies fertilizer-related emission reductions in agri-food supply chains. Proba’s approach is designed for credible Scope 3 accounting, aligned with SBTi and the GHG Protocol (FLAG guidance), with independent verification and a registry that supports traceability. For more information, visit proba.earth.
Erna Maciulis
Proba
erna@proba.earth
Visit us on social media:
LinkedIn
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
![]()
Media gallery

