The changelog below will highlight what has changed between this new version and the previous version.
Release notes summary
This release contains updates to Compliance and Data Quality tests, as well as extensive updates to our reference data support.
For Compliance tests, there are two updates:
LGL02 (overlap with concessions) has been removed from this Test Profile, as it did not reflect well the EUDR requirements.
Additionally, there were some changes around the deforestation tests: as there were uncertainties about the interpretation of DEF01 (deforestation based on public maps) and DEF04 (deforestation based on commercial maps), we decided to cap DEF01 to high risk if DEF04 scores are available. This way, it is possible to identify quickly which results to prioritise (DEF04). For countries where we do not have coverage of DEF04, DEF01 continues to generate the same results. As a result, DEF02 was adjusted to reflect this change, giving the same results as DEF01.
For the Data Quality tests, there were multiple changes as well.
Change in thresholds for GER05 (overlap with built up areas). In the past, this test has produced a higher rate of erroneous results, claiming that farm plots were at risk even though overlap with built up areas was negligible. As the underlying reference data has a relatively low accuracy (ca. 74%), we have adjusted the thresholds to account for this. In addition, the thresholds are now favoring small farm plots by having more lenient thresholds compared to larger farm plots. This is reflecting the reality of many small farm plots being in the vicinity of built up areas (such as villages, buildings, roads, etc.).
Changes to the test GEP01 (farm plot is extremely small). As size itself does not correlate with risk, we have decided to have this test only identify extreme outliers, and generate a binary outcome (low or critical risk).
ATT03 (farmer ID is missing) is now capped at medium risk, not producing critical risks anymore. This is to better align with EUDR requirements, where a farmer ID is not essential.
There were also extensive updates to reference data support, including launching new public deforestation maps (DEF01) for 14 countries (see list below in the table). We are also continuing to extend our commercial deforestation map coverage by adding new countries, and updating existing countries. Finally, we added one new country to the LGL03 (overlaps with indigenous people and local communities).
Changelog
Test | Test- Version | Changes | Background |
DEF01 | v1.02 | The risk thresholds now depend on the availability of DEF04 results (overlap with deforestation, commercial maps). | Farm plot overlaps with deforested area If commercial deforestation maps are available (i.e., DEF04 results), DEF01 scores maximum high risk and not critical risk anymore. This helps to prioritise the results from the commercial deforestation maps, as they are more reliable than public maps. When no commercial deforestation maps are available, DEF01 still continues to generate critical risks to alert users of potential deforestation risks. |
DEF01 | v1.02 | The public deforestation maps for the following countries have been updated:
| Farm plot overlaps with deforested area Previously the DEF01 reference data used Primary Humid Tropical Forest and Hansen tree cover loss as the input datasets to detect deforestation. This approach has been updated with new public datasets: JRC Global Forest Cover 2020 v2 for the basemap, and Tropical Moist Forest deforestation and Hansen loss year datasets for deforestation. The map provides deforestation up to 2023, and is at 30m resolution. |
DEF02 | v1.02 | As for DEF01, the risk thresholds for this test now depend on the availability of DEF04 results. | One farm plot with deforestation has contaminated all farm plots of this farmer ID DEF02 changes in accordance to DEF01 (see above), depending on whether or not commercial deforestation maps are available: If they are available, DEF02 scores maximum high risk. If they are not available, DEF02 continues to score up to critical risk. |
DEF04 | v1.00 | Reference data for additional countries have been added for the commercial deforestation maps.
The following countries have been added for this release:
The following countries received updates to their existing maps:
|
|
GER05 | v1.02 | New thresholds have been implemented. The new thresholds take the farm size into account and are more lenient towards smaller farms. | Farm plot overlaps with built up area We noticed that GER05 classified acceptable farm plots at an unusual rate. Investigations showed that this is due to the reference data with a limited accuracy and resolution. As farm plots are often in the immediate vicinity of built up areas (e.g., houses, sheds, roads), this test was particularly vulnerable to these limitations of the reference data. The new thresholds are more lenient, particularly towards smaller farm plots. |
GEP01 | v1.01 | The severity threshold for this test no longer includes medium and high risks. This test is now limited to a low and critical score severity threshold (> and ≤0.001 Ha). | Farm plot is extremely small As farm plot size is not inherently a measure of risk, only for extremely small farm plots, this test now reflects this: medium and high risk categories are removed to only alert for extremely small farm plots that are very unlikely to be true. |
GEP03 | v1.03 | New country exceptions for EC. | Farm plot is extremely large Based on newer data and analyses, we have updated the country exceptions for EC. The new thresholds for EC are stricter. |
ATT03 | v1.01 | This test is now limited to a low and medium risk severity threshold. The severity threshold for this test no longer includes the critical risk. | Farmer ID missing As farmer IDs are not strictly necessary for EUDR compliance, the maximum risk for this test has been reduced to medium risk. |
LGL02 | - | This test has been removed entirely. | Farm plot overlaps with concession areas After review, it was determined that this test does not align well with the actual EUDR risk, leading to confusing results and action steps. It was therefore decided to remove this test from the EUDR Test Profile. |
LGL03 | v1.00 | New country support has been added for:
| Farm plot overlaps with an indigenous people or local communities territory We expanded our database of Indigenous People and Local Community territories to include Nicaragua. In addition to sourcing from deforestation-free farm plots, EUDR also requires that agriculture is sourced legally, and adheres to the principle of free, prior and informed consent (FPIC), including as set out in the UN Declaration on the Rights of Indigenous Peoples. This includes indigenous and local communities having control and inclusion with decisions about land use, including agriculture. Farm plots overlapping with these areas produce a Medium risk, and we provide guidance on the specific local legal frameworks that are important to ensure your due diligence documentation is covered. |