FAO’s 2025 Global Forest Resources Assessment shows about 64% of North Korea’s land area as forest. The 2020 edition’s country dashboard showed about 50% for 2020. Both dashboards identify their country assessments as desk studies. To measure change, compare the years within one edition and read the supporting data notes before interpreting the difference between the two dashboard headlines.
A forest-area estimate and a satellite tree-cover layer describe different features of the landscape. FAO’s forest category combines vegetation thresholds with land use. A satellite layer can map the amount of canopy visible or a detected loss of tree cover. Start with the definition to understand why the figures can differ.
- FAO’s headline forest shares refer to land area and use the assessment’s definition.
- Satellite tree-cover results depend on the layer, baseline and canopy threshold.
- A detected loss of trees and a permanent change of land use require different evidence.
FAO — DPRK forest assessment, 2025 edition
Choose a reading route
What FAO counts as forest · Worked example: explain why two percentages can differ · Frequently asked questions
Chapters 1–4 · Read the data
Chapters 5–7 · Work through a comparison
Chapter 1 of 7
What FAO counts as forest
FAO’s published 2020 terms set minimum conditions for area, tree height and canopy cover, including land with trees capable of reaching the thresholds. They also consider predominant land use. This matters because trees can grow in places classified as agricultural or urban under the reporting framework.
An orchard can be a useful example. Its trees may be visible in a canopy image. The land can still be classified by its agricultural use in a forest assessment. A temporarily unstocked forest area can work the other way. It can remain in the forest category under specified conditions even when mature canopy is temporarily absent.
When citing a percentage, look at the denominator. A share of land area and a share of forest area are different measures. The country dashboard shows both kinds of percentage for different indicators. State the denominator in the caption, especially when comparing forest cover with the fraction of forest in protected areas.
FAO — forest terms and definitions
Chapter 2 of 7
How a satellite canopy threshold changes the result
Canopy cover describes how much of a surface is covered by the crowns of trees when viewed from above. A map can use a threshold to decide which places to include. At a higher threshold, a lightly wooded pixel may drop out of the displayed area. At a lower threshold, it may be included.
The Global Forest Watch documentation includes a canopy-density setting for displaying tree-cover loss. It uses a baseline density layer in the selection. A loss result therefore depends partly on the baseline and the threshold chosen. Record those settings with any area figure so another reader can repeat the query.
A pixel can contain a mixture of trees, bare ground and other vegetation. The map class represents the data at its resolution. A close-up photo or field survey can supply additional detail for a specific site. Neither kind of source needs to answer every question.
World Resources Institute — tree-cover loss methods
Contents · Previous chapter · Next chapter
Chapter 3 of 7
Why the two dashboard headlines need context
The 2020 and 2025 dashboards are separate assessment editions. The visible percentages refer to different reference years. Assessments can revise earlier figures as sources and methods change. To study the trend, open one edition’s series and compare the same indicator across reference years.
If you want to study revisions, use a different comparison. Choose a single reference year and find its value in both editions. Then read the notes about sources and estimation. This exercise tells you how the published estimate changed for the same year. It does not establish the size of any physical change during that year.
The phrase desk study is also useful context. It identifies the type of country assessment displayed. Look for the documentation that explains inputs and estimates before describing it as though someone made a new field count of trees across the country.
FAO — DPRK forest assessment, 2020 edition
Contents · Previous chapter · Next chapter
Chapter 4 of 7
Tree-cover loss, deforestation and recovery
A tree-cover loss layer records a detected change under its methodology. Fire, harvesting and permanent conversion can all remove trees. Those events can have different outcomes for later land use. To describe a specific loss as permanent conversion, look for evidence of what the land became and whether trees recovered.
Use a sequence of images when studying a site. Start with the baseline. Then look at the year when loss was detected and at subsequent years. Compare the dates and image quality before interpreting a bare or dark patch. A second source can help identify whether the area was converted, replanted or left to regenerate.
World Resources Institute — tree-cover loss methods
Contents · Previous chapter · Next chapter
Chapter 5 of 7
Worked example: explain why two percentages can differ
Suppose a reader brings a forest-area figure from FAO and a canopy figure from a map. Put the two values on separate rows. Add the year, the area covered and the definition. Then identify any threshold or land-use rule. This makes the comparison concrete.
If the question is how much forest area changed, use a consistent FAO series and state its edition. If the question is where a satellite detected a change in canopy, use the mapped layer and give its settings. Both can support an article when you link each claim to the measure that answers it.
FAO — DPRK forest assessment, 2025 edition
Contents · Previous chapter · Next chapter
Chapter 6 of 7
Read a forest map at the scale of its data
The University of Maryland tree-cover loss layer documented by Global Forest Watch uses Landsat-based observations at about 30-metre resolution. A cell covers a patch of ground. A coloured cell represents what the dataset detected in that patch under its method. A dot at the centre of the cell cannot identify an individual tree.
The World Resources Institute explains that its standard reporting uses a canopy baseline tied to the year 2000. Keep that reference year with the threshold in any summary. A threshold determines which areas are eligible for a particular loss analysis. The year of detected loss is another field. Mixing those dates can turn a statement about how the dataset was constructed into an apparent observation about later canopy.
In a classroom comparison, draw three cells with sparse, medium and dense canopy. Choose a threshold and mark which cells would be included. Then change the threshold. The forest has stayed the same in the drawing. The map result changes because the rule changed. This exercise explains one source of difference before you examine country totals.
World Resources Institute — tree-cover loss methods
Contents · Previous chapter · Next chapter
Chapter 7 of 7
How to compare figures from different editions
Download the relevant FAO table and record its edition, country and reference years. If both editions show a value for 2020, place those values in the same row. This is a comparison of published estimates for a fixed year. If you are studying change from 2020 to 2025, take both values from a consistent edition and check its methods.
Read what the table says about units too. Forest area can be shown in thousands of hectares. A dashboard can also show a rounded percentage. A percentage chart and a table in hectares describe related quantities if the denominator and coverage agree. The rounded chart may have less precision than the underlying table.
Suppose you want to explain a change near a mountain reserve. Start with a map of the reserve, then look for dated imagery or survey material that covers it. The country dashboard supplies national context. A local claim needs a local source. This is especially important in North Korea, where a map of one landscape can be easier to find than a consistent field record for the whole country.
A useful final note describes why you selected each source. One may provide a country estimate, another may show where canopy change was detected, and a third may describe what followed at a site. The combination helps you distinguish the extent of forest, the location of change and its possible cause.
For example, a map of canopy loss within the Mount Myohyang reserve would need the reserve boundary and the selected observation dates. A national percentage cannot locate those changes.
FAO — DPRK forest assessment, 2025 edition
Contents · Previous chapter · Questions
Frequently asked questions
What forest share does the 2025 FAO dashboard show?
It displays about 64% of land area for North Korea.
Can the 2020 and 2025 headline percentages prove how much forest grew?
A trend calculation should compare the years in one edition and use its supporting methods and data notes.
Is tree-cover loss automatically permanent deforestation?
The layer measures tree-cover change. Evidence of later land use and recovery helps establish whether conversion was permanent.
Sources and next reading
Sources
- FAO — DPRK forest assessment, 2025 edition
- FAO — DPRK forest assessment, 2020 edition
- FAO — forest terms and definitions
- World Resources Institute — tree-cover loss methods