Spatial + Temporal Scope [MGD Sections]

REDD+ Activities [MGD Sections]

Forest Definition [MGD Sections]

Carbon Pools [MGD Sections]

2.3.1   Definition of forest Previous topic Parent topic Child topic Next topic

A forest definition(1) is needed to be able to determine whether deforestation or afforestation or reforestation has taken place, and to define the areas within which degradation and the other REDD+ activities may occur. Definitions can have a significant effect on the estimate of emissions or removals associated with REDD+ activities, and the allocation to each activity. Definitions should be used consistently over time and across REDD+ activities, and the definition used to establish the FREL or FRL should be the same as that used subsequently for MRV. For example exclusions from the forest definition, e.g. for oil palm plantations, should be applied consistently over time.
No single definition of forest has been agreed under the UNFCCC for REDD+ purposes. The annex to decision 12/CP.17 Opens in new window requests Parties to provide the definition of forest used, and if it differs from the definition of forest used in the national GHGI, or in reporting to other international organizations, to explain why, and why the definition used in the construction of FREL and/or FRL was chosen. This indicates an expectation that the forest definition used for REDD+ will be the same as that used for previous reporting on forests or that other reporting will be updated to reflect any new definition.
In considering forest definitions, NFMS may wish to note that, as already identified in Table 3, the GPG2003 defines Forest Land as including all land with woody vegetation consistent with thresholds used to define forest land in the national GHG inventory, sub-divided into managed and unmanaged, and also by ecosystem type as specified in the IPCC Guidelines. It also includes systems with vegetation that currently fall below, but are expected to exceed, the threshold of the forest land category. The Forest Land definition in the 2006GL refers to threshold values. IPCC therefore anticipates that countries will have a forest definition with quantitative thresholds, based on land use since temporary loss of forest cover does not entail transition to another land use provided there is expectation of recovery of threshold values. Threshold values commonly refer to minimum area, percentage crown cover and tree height although other thresholds are possible e.g. referring to minimum width.
The IPCC definition subdivides forests into managed and unmanaged. This is because anthropogenic carbon stock changes and associated greenhouse gas emissions and removals are assumed(2) to occur predominantly on managed land and therefore those on land remaining unmanaged are not reported under the IPCC Guidelines. Reporting is required when unmanaged land is subject to land use conversion(3). According to GPG2003, Managed land may be distinguished from that unmanaged by fulfilling not only the production but also ecological and social functions. The detailed definitions and the national approach to distinguishing between unmanaged and managed land should be described in a transparent manner(4). Given this broad definition of 'managed' it is entirely possible that countries may have little or no land considered unmanaged.
The detailed definition of what is considered managed may differ from country to country, but national definitions should be applied consistently over time otherwise there is risk that apparent changes in emissions or removals will reflect differences in the way definitions are applied, rather than the effect of REDD+ activities. For the same reason the procedures used to assess whether thresholds are met also need to be applied consistently over time, especially where different methods (e.g. ground-based and remote sensing) are being used together. How consistency is achieved could usefully be reported under MRV provisions. Issues include determination of forest boundaries in fragmented landscapes (relevant to minimum area), determination of crown cover(5), and how height is determined, or (where used as a criterion) minimum width.
Countries that do not already have a forest definition may wish to note that for Kyoto Protocol (KP) purposes Forest …is a minimum area of land of 0.05–1.0 hectare with tree crown cover (or equivalent stocking level) of more than 10–30 per cent with trees with the potential to reach a minimum height of 2–5 metres at maturity. A forest may consist either of closed forest formations where trees of various storeys and undergrowth cover a high proportion of the ground or open forest. Young natural stands and all plantations which have yet to reach a crown density of 10–30 per cent or tree height of 2–5 metres are included under forest, as are areas normally forming part of the forest area which are temporarily unstocked as a result of human intervention such as harvesting or natural causes but which are expected to revert to forest(6).
The Cancun Agreements specify that REDD+ mitigation actions should not incentivize conversion of natural forests and therefore the NFMS should be able to distinguish natural forest within land meeting the forest definition. This may require supplementary data on the distribution of forest ecosystems within the country.
National forest definitions need to support reliable classification of forest areas and changes, and hence to estimate carbon stock changes, and associated GHG emissions and removals. In establishing a national forest definition it is important to distinguish ‘forest cover’ from ‘forest land’, which is typically reported by forest inventories and takes account of land-use.
From the forest inventory perspective, and as defined by Food and Agricultural Organization of the United Nations (FAO), forest land may include areas that are temporarily treeless as a result of harvesting or natural disturbance. The same land may be classified as non-forest category by remote-sensing of land cover, and in a forest category from an inventory of forest land. The opposite is also true – the FAO forest definition does not include land that is predominantly agricultural or urban, even if such land has tree cover which may meet the national threshold.
These differences can have a significant effect on the resulting REDD+ estimates and can complicate comparisons with land cover classification approaches, e.g. when losses due to temporary removals of trees followed by regrowth are classified as deforestation according to the national definition, forest land use has been maintained and forest regrowth is expected. This bias can be corrected for by use of auxiliary data, by analysing time series of remotely-sensed data to detect where regrowth is occurring, and by estimating REDD+ activities jointly so that regrowth as well as forest loss is captured. Full tracking of lands affected by REDD+ would require the use of set rules to ensure that lands are correctly categorized and through time.
If in practice information on threshold recovery is not available it may be necessary to base the definition on tree cover, at least until there is sufficient integration of remotely sensed and ground-based data to permit a land-use definition. Clearly the minimum area used in the forest definition can have implications for the spatial resolution of the imagery used to detect forest areas and changes, and may affect the ability to track the identified drivers of changes with different scales, intensity and spatial distribution. Reduction in canopy cover below the minimum does not necessarily entail clearance of the entire area which may require detection at finer resolution, especially with large minimum areas.
Table 5 summarizes the forest definitions used by countries which, by February 2016, had submitted REDD+ reference levels. This table shows that the threshold criteria being used are generally consistent with the forest definition introduced under the UNFCCC for KP purposes. In one case a much larger minimum area unit is used. This was to help capture botanical species and human activity information in a variegated landscape. Its use does not prevent defining forest using smaller minimum areas. The larger the minimum area used, the more land that would otherwise be considered forest will be transferred to other land classifications.
The table also indicates that most countries are not using potential height in their forest definitions adopted for REDD+ purposes. Some countries are excluding various areas that would otherwise meet the forest definition, on the grounds that they are under non-forest land uses. This shows that land use considerations can in fact be used to operationalize the forest definition. In all cases it will be important to apply the forest definition consistently over time and to link it to other definitions in the GHGI to ensure that all significant emissions and removals are reflected in national estimates.

Table 5: Forest definitions adopted by countries having submitted FREL/FRLs by February 2016

Country
Area (ha)
Canopy cover (%)
Height (m)
Exclusions
Brazil (a)
0.5
10
5
Land predominantly under agricultural or urban land use
Chile
0.5
10 / 25
(d)
Self-sown trees of introduced species.
Colombia
1
30
5 (e)
Commercial forest plantations, palm crops and planted trees for agricultural production.
Congo
0.5
30
3
--
Costa Rica
1
30
5
--
Ecuador
1
30
5
--
Ethiopia
0.5
20
2
--
Guyana
1
30
5
--
Indonesia (f)
0.25
30
5
Non-natural forested peat
Malaysia
0.5
30
5
Oil palm and rubber plantations
Mexico
50
10
4 (g)
Lands subject to a land use that is predominantly agricultural or urban.
Paraguay
1
10 / 30 (b)
3 / 5 (b)
Urban areas, grasslands, plantations predominantly agricultural, agroforestry and silvopastoral systems, whose primary purpose is agriculture.
Peru
0.09
(c)
5
--
Vietnam
0.5
10
5 (g)
--
Zambia
0.5
10
5
--
Notes: a) Deforestation for the Amazonia biome is not associated with thresholds, but simply with canopy cover equal to zero. Situations in which forest falls below the thresholds of the FAO definition but still does not have canopy cover equals to zero is characterized as forest degradation. b) First alternative applies in the Western region; second in the Eastern c) Detection of forest depends on classification algorithm at the pixel level d) Tree species required to predominate. Minimum width of 40 m applied e) At the time of identification f) Working definition identifies forest through visual interpretation with a polygon size equivalent to 6.25 ha. g) Definition includes the potential to reach this height.

 (1)
A general discussion of forest definitions is provided by Tomppo et al. 2010. National Forest Inventories: Pathways to common reporting. Springer.
 (2)
See discussion in 2006GL vol 4 page 1.5 Opens in new window
 (3)
 (4)
The 2006 GL say that Managed land is land where human interventions and practices have been applied to perform production, ecological or social functions. All land definitions and classifications should be specified at the national level, described in a transparent manner, and be applied consistently over time.
 (5)
See e.g. Magdon P & C Kleinn 2012. Uncertainties of forest area estimates caused by the minimum crown cover monitoring. Environment Monitoring and Assessment 185(6): 5345-5360.
 (6)
In the Forest Resource Assessment 2010 FAO defines Forest as Land spanning more than 0.5 hectares with trees higher than 5 meters and a canopy cover of more than 10%, or trees able to reach these thresholds in situ. It does not include land that is predominantly under agricultural or urban land use. The area threshold falls within the range in the KP definition and the height threshold is at the upper end of the KP range.