More Accurate Deforestation Assessments with TRACT’s New Thresholding Feature
European Union Deforestation Regulation (EUDR) compliance requires agricultural products entering or exiting the EU market to meet strict due diligence standards. Stakeholders in the supply chain must gather geolocation data for their supply chains and provide evidence that their operations are deforestation-free. This is where TRACT comes in.
Understanding the Deforestation Assessment Workflow
To determine the deforestation status at the plot level, TRACT employs a multi-step workflow. This assessment process utilizes various datasets, including forest cover and loss data, to evaluate whether a specific area is compliant with EUDR requirements.
The workflow consists of three key assessments:
Forest Baseline: Establishing the original, 2020 forest conditions before any agricultural activity.
Forest Loss: Evaluating changes in forest cover over time to identify any deforestation.
Visual Assessment: Conducting manual reviews to validate the findings from the data analysis.
Deforestation Alerts: Notifying of any possible deforestation events to identify cases that require further investigation.
These assessments lead to a confidence level regarding the deforestation status, helping businesses make informed decisions.
The Challenge of False Positives
One significant challenge in assessing deforestation status is the possibility of false positives. A false positive occurs when the assessment indicates deforestation where there is none, which can lead to unnecessary compliance issues for businesses. This can occur due to various factors, including:
Data Limitations: Inaccuracies in the dataset for some geographies or land cover can misrepresent forest conditions.
Variations in Land Use: Changes in land use that are not related to deforestation, such as replanting or natural regeneration, can be incorrectly flagged.
What makes TRACT’s approach unique is the movable threshold for detecting Tree Cover Loss. This threshold can be adjusted to reflect different real-world conditions and the limitations of geospatial data.
When mixing vector data (like polygons) with raster data (pixel-based maps), overlaps may not accurately reflect reality. For instance, if a pixel lands on the edge of a farm, it could lead to a false positive. If not identified, this may wrongly flag many farms as non-compliant.
TRACT employs both area-based and percentage-based thresholds. These adjustable thresholds allow users to quickly identify polygons that are more likely to result in false positives versus genuine deforestation. When the threshold is raised from 0% or 0 hectares, the number of polygons categorized as requiring further checks increases, but the number of polygons with detected deforestation decreases. This creates a shortlist of areas where false positives are most likely, allowing businesses to focus their efforts where they’re most urgently required.
When comparing the TRACT approach against the base data layers for deforestation, we’eforestation, we’ve seen a decrease in false positives by 55%. This means more deforestation-free products can enter the market.
Why Accuracy Matters
Accurate deforestation assessments are crucial for compliance with the EUDR. False positives can lead to costly investigations, unnecessary penalties, and damage to a company's reputation. By using TRACT’s platform, businesses can confidently demonstrate their commitment to sustainable practices and meet regulatory requirements.
Want to dive deeper into our deforestation methodology?
Download the full document to understand how TRACT addresses false positives.