Monitoring relates to the periodic measurement of carbon pools in the project area and in proxy or reference non-project areas. Permanent sample plots, which often are used in the initial carbon inventory (e.g., Box 5-3), are generally considered to be statistically superior means for evaluating changes in forest carbon pools. Methods are well established and tested for determining the number, size, and distribution of permanent plots (i.e., sampling design) in several LULUCF projects to maximize the precision for a given fixed monitoring cost (MacDicken, 1997a; Winrock International, 1999). The use of permanent plots allows for efficient assessments of changes in carbon stocks over time and for cost- and time-efficient verification of the project's reported carbon benefits (MacDicken, 1997a). Moreover, a random selection of permanent plots may be measured only as part of the ongoing monitoring program. In addition, not all of the initial carbon pools need be measured at every interval in some projects; judicious selection of some pools could enable them to serve as indicators that the project is following the expected trajectory. For example, projects that are designed to avoid emissions by arresting deforestation or logging need only establish that no trees are removed or clearings made over the course of the project. In projects that are designed to sequester carbon, changes in vegetation carbon or soil carbon pools need to be re-measured periodically.
Remote sensing can provide a useful means to monitor LULUCF projects (see Chapter 2). A variety of remote data collection technologies are now widely available, ranging from satellite imagery to aerial photographs from low-flying airplanes. A new advance in this area couples dual-camera videography with a pulse laser profiler, data recorders, and differential global positioning system (GPS) mounted on a single-engine airplane (Department of Forestry and Conservation Management, University of Massachusetts, 1999). This system can produce indices of crown density, number of trees per unit area, and tree height; it also can identify the extent of gaps, which will be especially useful for projects that are related to arresting or modifying logging, as well as monitoring for small-scale human disturbance in protected forests.
In some circumstances, models (parameterized for project conditions) can be used to project changes in carbon pools over short time periods for which direct measurements fall below easily detectable levels, followed by direct measurements over longer time intervals to verify model projections (Post et al., 1999; Vine et al., 1999). Process-based models are particularly useful in projecting slowly occurring changes in soil carbon pools (Paustian et al., 1997; Post et al., 1999). Likewise, models exist for plantations and agroforestry systems (e.g., Maclaren, 1996; Schlamadinger and Marland, 1996; Mohren et al., 1999; ICRAF, n.d.) that could be used in conjunction with direct field measurements to estimate changes in carbon pools over shorter time frames.
Other reports in this collection