In implementing the Kyoto Protocol, the Parties are likely to make a series
of policy decisions on definitions, accounting procedures, and methods for accounting
for carbon stocks and changes in stocks. This chapter addresses the core definitional,
accounting, and methodological issues that may be relevant to the interpretation
of several Articles in the Protocol. It introduces many issues that cut across
other chapters in this report.
This chapter has four main sections. Section 2.2 deals
with definitional issues; Section 2.3 with accounting
issues; Section 2.4 with methods available for measuring
and verifying carbon stocks, emissions, and sinks; and Section
2.5 with sustainability issues.
Definitions and Their Implications
In a given land area, a "full" carbon accounting system would consist of a
complete accounting for changes in carbon stocks across all carbon pools in
a given time period. In principle, applying full carbon accounting to all land
in each country would yield the net carbon exchange between terrestrial ecosystems
and the atmosphere. However, the Kyoto Protocol specifies, among other things,
that attention be restricted to land areas subject to "direct human-induced"
(Article 3.3) or human-induced (Article 3.4) activities.
The Parties to the United Nations Framework Convention on Climate Change (UNFCCC)
may wish to make decisions about the definitions of terms used in Articles 3.3
and 3.4-in particular, "land-use change," "forestry activities," "human-induced"
and "direct human-induced," "afforestation," "reforestation," "deforestation,"
and "carbon stocks"-because these definitions may affect carbon accounting under
the Protocol. These definitions are critical to the accounting of sources and
sinks under the Protocol: They determine the scope of human activities that
are accounted under Article 3.3 and which may be eligible under other Articles
(such as Article 3.4) of the Protocol.
Definitions that already exist to serve other purposes will not necessarily
suffice for the particular needs of Articles 3.3 and 3.4; therefore, this Special
Report provides a series of definitional scenarios to illustrate the implications
of possible sets of definitions. The basis of the definitional scenarios is
described in this chapter; the scenarios are presented and analyzed in Chapter
The revised Guidelines for National Greenhouse Gas Inventories-hereafter referred
to as the IPCC Guidelines (IPCC, 1997)-describe a methodology for a comprehensive
approach to measuring "anthropogenic emissions by sources and removals by sinks
of greenhouse gases" that is feasible to implement in most nations. These guidelines
include suggested definitions for many important terms and can form a starting
point for reporting compliance with the Kyoto Protocol. Article 5.2 of the Kyoto
Protocol and a decision (Decision 2/CP.3) at the Third Conference of Parties
(COP) refers to the IPCC Guidelines; Article 5.2 notes that where the IPCC Guidelines
are not used "appropriate adjustments shall be applied according to methodologies
agreed upon by the" first COP.
Most definitions of "forest" used at the national and regional levels are based
on land-use status or a minimum threshold of canopy cover and/or tree height.
Definitions based solely on either land use or canopy cover have limitations
for carbon accounting. Land use, for example, may be defined in terms of administrative
or cultural purposes and may bear little relation to the actual vegetation or
amount of carbon on a site. Similarly, there are times (e.g., during regeneration
after disturbance or harvesting) when the canopy cover of forests may be negligible.
Deforestation can be defined (e.g., in the IPCC Guidelines) as the conversion
of forest to non-forest; reforestation and afforestation are the conversion
of non-forested lands to forest. Issues arise with respect to the timing, sequencing,
and causes of these activities. However, some operational forestry definitions
of reforestation include the regeneration of trees after harvesting. When such
definitions of reforestation are used but the accompanying definition of deforestation
does not include disturbance or harvesting, asymmetric accounting occurs. Under
such definitions of reforestation, many forestry management practices would
create large areas of lands under Article 3.3.
Definitions of ARD that are based only on canopy cover may allow activities
that lead to significant carbon fluxes to remain unaccounted. For instance,
definitions that are based on a single threshold for canopy cover alone may
allow changes in carbon stocks to remain unaccounted under Article 3.3. If a
high threshold for canopy cover (e.g., 70-percent canopy cover) is used in the
definition of a forest, many areas of sparse forest and woodland could be cleared
without the losses in carbon being counted under Article 3.3. If a low threshold
is set (e.g., 10-percent canopy cover), dense forest could be heavily degraded
and significant amounts of carbon released and the actions would not be designated
as deforestation. This problem could be addressed by using national, regional,
or biome-specific thresholds-for example, a low canopy cover for savannas and
a high canopy cover for moist forests-or by definitions based on degradation
or aggradation criteria.
Definitions that are based strictly on a canopy cover threshold may lead to
an interpretation of conventional harvest/ regeneration cycles as deforestation
followed by reforestation. Such interpretations lead to significantly greater
areas of forest being included within lands under Article 3.3. These interpretations
also exacerbate other issues that relate to the measurement of carbon stocks
(e.g., which pools?) over time (e.g., how long and when?) and the attribution
of change to human activity. Additional qualifiers that are based on the timing
of activities or growth potential (see Chapter 3) can be
introduced to the definition of a forest to prevent this problem.
Afforestation is usually defined as the establishment of forest on land that
has been unforested for a long period (variously described as "in historical
times," "many decades," etc.). The precise period that distinguishes afforestation
from reforestation is not critical to lands under Article 3.3, provided that
afforestation and reforestation are treated identically in the Protocol, as
they are in the IPCC Guidelines.
Carbon content in vegetation (or woody component) is an alternative basis for
the threshold for the definition of a forest, but it has similar limitations
as canopy-cover criteria. Such definitions relate more directly, however, to
carbon stocks and fluxes.
There are definitional issues concerning the boundary between natural phenomena
and direct human-induced activities-for example, when significant forest losses
occur as a result of natural or human-induced fires or disturbances such as
pest outbreaks. Distinguishing between natural and anthropogenic factors that
influence the vulnerability of the land to disturbance will be difficult. In
cases involving non-Kyoto lands, where fires or pest outbreaks trigger land-use
change, the consequences are similar to deforestation. If vegetation cover is
allowed to recover, with no change in land use, such disturbances would not
lead to a long-term change in carbon stocks. In this context, a key issue for
policymakers to decide is whether accounting should count neither the loss nor
subsequent sequestration of carbon (symmetric accounting as in the IPCC Guidelines);
count both the loss and subsequent sequestration of carbon (symmetric accounting
that would increase monitoring requirements); count only the loss of carbon
(asymmetric accounting); or count only the subsequent sequestration (asymmetric
In cases involving forested Kyoto lands, a key issue for policymakers to decide
is whether accounting should count neither the loss nor subsequent sequestration
of carbon (symmetric accounting)-which, however, essentially changes the area
burned to non-Kyoto land and leaves several accounting issues to be resolved;
count both the loss and the subsequent sequestration of carbon (symmetric accounting);
count only the loss of carbon (asymmetric accounting), which creates similar
problems to counting neither the loss nor subsequent sequestration of carbon;
or count only the subsequent sequestration (asymmetric accounting), which may
create an incentive to burn forests.
This chapter provides a logical schema for examining a set of alternative forest
definitions that are based on combinations of land use and variable canopy-cover
thresholds. Together with appropriate ARD definitions, these forest definitions
form a set of definitional scenarios to aid in the implementation of the Protocol.
Each of these scenarios leads to a logical sequence of decisions to be made
by policymakers, and each decision has implications for the accounting of carbon,
especially under Article 3.3. These definitional scenarios are analyzed in Chapter
Article 3.4 refers to additional activities that do not explicitly fall under
ARD. These activities potentially include all forms of land management, including
activities in both the agricultural sector and the forestry sector. In the broadest
sense, these activities could incorporate all lands under Article 3.3, as well
as all lands that are (or will be) managed by humans for the production of food,
fiber, forage, and other purposes.
When inclusion of additional activities is considered, it is possible to interpret
"activity" broadly (e.g., cropland management) or narrowly (e.g., change in
tillage method, fertilization, or cover crops). The broad definition may lend
itself more readily to land-based carbon stock change methods of measuring the
net impact of the activity. The narrow interpretation of "activity" may lend
itself to activity-based measures in some instances, where the area of practice
reported is multiplied by research-derived estimates of changes in emissions
or removals in sinks associated with the activity. In other instances, current
scientific knowledge of carbon flux rates associated with particular activities
is very limited, and these combinations could produce very high uncertainty
Narrow definitions may be useful if a small number of discrete activities are
added, whereas a broad interpretation can provide a comprehensive framework
for full carbon accounting as discussed in the IPCC Guidelines. Under either
definitional interpretation, it is possible to choose a land-based or activity-based
method of accounting for carbon impacts. Those combined choices will have major
impacts on the accuracy, transparency, and verifiability of reported amounts.