Webinar "Developing the 2020 forest boundary map for EUDR compliance” organized by the EUDR Forestry Network and the Agroforestry policy research network on May 29, 2026, aims to provide information on the scientific basis, implementation methods, and professional guidance for localities in standardizing map data to meet the requirements for exporting agricultural and forest products to the European Union (EU) market.
Geographical requirements of EUDR and the necessity of a national map
The European Union deforestation regulation (EUDR) sets three conditions for imported goods: no deforestation, legal production according to the laws of the country of origin, and a Due Diligence Statement (DDS).
Dr. Truong Tat Do (Department of Forestry and Forest Protection) explained that if Vietnam does not proactively develop and publish the 2020 forest boundary map, the EU will use available global forest cover data sources such as JRC or Global Forest Watch. However, these international maps often use low or medium-resolution satellite images and lack field verification, easily leading to errors in due diligence results (e.g., misidentifying agroforestry models or areas outside planning as deforested areas). Proactively publishing a national map aims to protect the legitimate interests of exporting agricultural and forest product sectors.
Construction method and results of multi-source data classification
Assoc. Prof. Dr. Pham Van Duan (Institute of Ecology and Environment – IFEE) stated that the project's goal is to develop a forest map of Vietnam as of December 31, 2020, that meets international criteria. This activity is based on the principle of inheriting official Vietnamese data, updating it with remote sensing images, independently comparing it with international data, and verifying it locally. Notably, the forest boundary map for EUDR compliance uses the FAO (International) forest definition criteria, operating relatively independently of the current forest status map (HTR) used for regular state management.
The technical process is implemented through 6 steps, including standardizing the 2020 base map, analyzing remote sensing changes (2015–2020), spatial filtering according to forestry land scope, comparing with global data GFC2020v3/JRC, transferring to localities for review, and integrating onto WebGIS.
Based on the initial multi-source cross-referencing results covering a total area of 4,281,294 ha, the data is divided into 3 management classifications:
- Green zones account for 54.2% (2,321,103 plots): National, remote sensing, and international data sources unanimously confirm the presence of forests, considered safe.
- Yellow zones account for 45% (1,926,829 plots): Minor discrepancies in status, tree species, or planning are recorded; product lines here are outside the scope of EUDR.
- Red zones account for 0.8% (33,362 plots): Contradictions exist where stored data indicates forest but remote sensing actually shows agricultural cultivation (coffee, rubber, timber).
As of the reporting time, 34 out of 34 relevant provinces and cities have provided new administrative boundaries. Among them, 19 provinces have submitted feedback data, and 15 provinces have not yet responded. Some practical difficulties noted include a lack of original data in some localities, inconsistent understanding of the EUDR map across different levels, and uneven technical capacity among grassroots officials.
Professional guidelines for local in-office review
To support local forest rangers, M.Sc. Pham Ngoc Hai shared the data processing procedure. The main implementation tool is QGIS software integrated with the specialized Forest Plots Checker plugin, which automates the viewing and cross-checking of each forest plot.
For plots categorized as suspicious or in risk areas, forest rangers are instructed to apply two solutions for verifying ground history:
- Google Earth (Send2GE Plugin): Use the time slider (Historical Imagery) to access high-resolution satellite images at time points before and after December 31, 2020, to determine the actual status.
- CCDC Landsat (on Google Earth Engine platform): Analyze satellite image series from 1985 via the NDFI canopy openness index to precisely pinpoint the time of any forest reduction changes.
When errors are identified, local technical staff apply the Split Features technique if only a part of the forest plot is affected, or the Vertex Tool to perfectly match Sentinel-2/Planet satellite images. Field verification is only conducted when in-office analysis methods cannot provide a definitive conclusion.
System completion roadmap
The Department of Forestry and Forest Protection has outlined a roadmap for database completion in two phases:
- Current Phase (Version 1.0): Continue reviewing data from provinces that have not yet responded, update additional changes occurring before June 15, and proceed to integrate all data onto the official WebGIS system (eudr.vnforest.gov.vn) by June 30, 2026, for the initial map publication.
- Next Phase (Version 2.0 or 3.0): Localities will continue to allocate resources for more in-depth review, focusing on yellow zone areas to improve the quality and accuracy of the map.
The development of the 2020 Forest Boundary Map is an essential technical preparatory step, laying the foundation for Vietnamese agricultural and forest products to meet international trade standards transparently and clearly.
No thoughts shared yet. Be the first to share your thoughts!