Land Use, Land-Use Change and Forestry

Other reports in this collection

6.3. Adequacy of the IPCC Guidelines for Reporting National Activities under the Kyoto Protocol

This section assesses the adequacy of the Guidelines in relation to activities under Articles 3.3 and 3.4. It begins with a discussion of reporting issues common to ARD, then addresses issues that are specific to deforestation.

6.3.1. Afforestation, Reforestation, and Deforestation under Article 3.3 Generic Issues on Afforestation, Reforestation, and Deforestation

In the Guidelines, stock changes arising from afforestation and reforestation activities are reported under the same category, "Changes in forest and woody biomass stocks." For accounting purposes, there is no distinction between afforestation and reforestation; the same method is used for both. Deforestation can be dealt with separately, according to the principles for accounting of "Forest and Grassland Conversion." Chapter 2 explores a range of possible definitions of ARD. Chapter 3 further explores a wide range of definitional scenarios, including that of the FAO. If Parties choose the IPCC definitions, afforestation and reforestation can be dealt with using the principles in the Guidelines. For discussion only, this section assumes that the principles of afforestation and reforestation are the same as those in the Guidelines.

There would be implications for all three ARD activities if Parties were to adopt an extension beyond the Workbook. Carbon pools are limited to changes in aboveground biomass and soil carbon in the Workbook. Additional carbon pools are discussed in and can be accounted for using the approach laid out in the Reference Manual. According to the rules, modalities, and guidelines agreed to under Articles 3.3 and 5.2, however, Parties may decide that accounting methods for these additional carbon pools may need to be developed and provided in the Workbook.

The calculations in the Workbook do not explicitly link the aboveground biomass and soil carbon pools for a specific activity. To resolve this issue, reporting of ARD for aboveground and below-ground biomass could be geographically explicit (mapped) or linked through modeling, or the accounting methods for these pools could be slightly modified. The same principle would apply to the linking of other carbon pools, such as more complete treatment of surface soil litter, below-ground biomass, and wood products (see Section 6.3.3). Geographically explicit reporting and modification of accounting methods are not expected to be difficult to implement, although data can be expensive to obtain on a large scale, however, even for Parties with a good deal of scientific infrastructure (see Chapter 3). Furthermore, the data would have to be reported at appropriate time intervals in all cases and with consideration of the "since 1990" time clause.

All three ARD activities cause changes in aboveground biomass. Closely linked to the issue of ARD is the definition of forests, which has implications for lands that are included in an inventory (see Chapters 2 and 3). If Parties adopt definitions for forests other than the existing ones, the Guidelines may have to be improved to account for the new definitions. Changing these definitions by forest ecosystem or crown cover is unlikely to have significant implications on the Guidelines; the approach for estimating changes in aboveground biomass is expected to remain similar. If Parties adopted definitions of ARD activities that differed from the IPCC definitions, however, more significant improvements to the Guidelines might be required-depending on how the regeneration and harvest cycle were treated, for example.

If a land-based accounting system (Chapter 2) were to apply, a generic issue for ARD is that there may be a need to separately identify land into ARD categories and report land categorization in 1990, as well as in the commitment period. The optimum size of the landscape unit at which ARD activities may be detected is another consideration (Chapter 3).

If an activity-based accounting system (Chapter 3) were to apply, another solution might be possible. Differentiation among categories can be made at the beginning of a commitment period (in 2008, in the first instance) on the basis of activities implemented prior to 2008. In this system, the impact of an activity on carbon stocks could be tracked and summed per unit area. Carbon stocks would be more difficult to verify under this system, however, than those tracked under a land-based accounting system (see Chapter 2).
For reporting of ARD under both the land-based or activity-based systems, the activities and the areas subject to these activities since 1 January 1990 would have to be identified. The Guidelines do not currently apply this time clause to such land areas. The Guidelines require the use of land areas and ARD activities to estimate changes in carbon stock, however; hence, either a land-based or activity-based system could be used to track aboveground biomass and soil carbon.

To more accurately account for changes in carbon stocks under ARD, a finer level of geographic detail and subcategories may be appropriate. Stocks estimates could be improved by reflecting forest management type, species, soil type, and so forth, as well as by substituting default assumptions with national or region-specific data and assumptions from local sources (IPCC, 1997, Vol. 3, p 5.15).

On the other hand, if changes in stocks associated with ARD activities were reported at aggregated levels, reporting could be made geographically explicit at an appropriate scale, with geo-referencing of stocks according to species, forest type, and age of stand. The same principle applies with regard to verifying that a forest qualifies as ARD land and verifying changes in these stocks. Geo-referencing in itself does not provide transparency. The Guidelines would not be adequate for this purpose. A well-designed national system, together with the application of good practice, may be needed to ensure verifiability and transparency of reported data.

In relation to ARD, examples of data or assumptions that may be reported include the following:

  • Forest inventories at the required level of accuracy (IPCC, 1997, Vol. 3, p. 5.16). New methods based on satellite and aerial photography are also available for large-scale measurements, but only of area. Generic treatment for data quality is covered in the GPG.
  • Conversion factors appropriate to categories of forests (IPCC, 1997, Vol. 3, p. 5.18) for estimating tons of biomass per unit volume of commercial timber and tons of carbon per ton of dry biomass.
  • Biomass densities appropriate to the mix of species in the growing stock of a particular region could be used, based on literature data or field measurements (IPCC, 1997, Vol. 3, p. 5.22). Any default values would have to be age-class dependent, especially because stands established since 1990 will be very young during the commitment period, and expansion factors tend to be significantly higher for young stands.
  • The same principles apply to below-ground biomass (coarse woody roots, etc.) and to forest litter (dead roots, slash, etc.).

All three ARD activities can affect the level of soil carbon. Although the Guidelines can be readily applied to account for ARD, they focus on agricultural land use and management, rather than ARD. For comprehensive carbon accounting, soil carbon components of soil surface litter could be included. Consideration would have to be given to soil depth and rates of change in carbon, as appropriate. In cropped soils, however, most changes occur in the upper 30 cm (Chapter 2), as assumed in the Workbook. In some situations, consideration of deep soil carbon and more types of soil humus, with differing rates of decomposition, might be necessary for ARD activities. Significant changes may occur at 30-50 cm depth-for example, in deep tropical soils (see Chapters 2 and 4), where changes in soil carbon are generally faster than in boreal areas. Consistency in the accounting rules may need to be established, especially if crediting for soil carbon is allowed.

With respect to soil carbon, other refinements to the Guidelines may include the following:

  • Use of soil carbon models or methods with realistic time dependencies that reflect the soil type, moisture, and temperature conditions. In the IPCC default method, changes are assumed to be linear in time, so a straight averaging of past rates of change of land use is used. In the Protocol context, changes in stock resulting from increase or decay of soil carbon could be calculated using a time course (e.g., exponential) appropriate to the region and the commitment period in light of local conditions. This approach will give greater weight to more recent events.
  • Use of improved soil carbon data from surveys, field studies, and long-term agricultural experiments to replace default values.
  • Use of geo-referencing to relate data on base factors, tillage factors, and input factors and initial soil carbon content (IPCC, 1997, Vol. 2, p. 5.25) that is transparently reported with aboveground biomass changes. The base factor refers to the change in soil organic matter that is associated with the conversion of native vegetation to agriculture (IPCC, 1997, Vol. 3, p. 5.47). The term "initial" is used here rather than "native" to take into account that deforestation may be of forest that had been disturbed sometime in the historical past.
  • As discussed in Section 6.3.3, changes in carbon wood products could be considered (Winjum et al., 1998; Brown et al., 1999).

A key question is whether averaging periods for data are appropriate. Changes in biomass are linked to climatic fluctuations, so changes in stocks between 2008 and 2012 could be depressed or boosted by unusual weather patterns. Such fluctuations may affect stock changes over the commitment period for afforestation and reforestation or regrowth after deforestation. Separating the observed stock change that is directly human-induced from the influence of indirect causes and natural variability is generally difficult. One possibility is the use of default data to smooth out the effects of natural variability and extreme climatic events. This use of default data would deviate, however, from the principles of the Guidelines.
Because the Guidelines do not fully address the issue of uncertainties (Lim et al., 1999b), the reporting of ARD activities could be modified to include treatment of uncertainties and methods for error propagation-using, for example, Monte Carlo analyses or standardized calculation methods. Uncertainties in measurements of stocks at two points in time (e.g., 2008 and 2012) may cancel out for systematic errors. Under the gross-net approach of Article 3.3, however, the uncertainties would not cancel out for 1990 and the commitment period: For 1990, the uncertainties would comprise the non-LUCF sectors (gross) only, whereas for the commitment period the uncertainties would be the net of the uncertainties of estimates in non-LUCF and LUCF sectors.

Other reports in this collection

IPCC Homepage