A landscape-based analysis of temporal and spatial variation of deforestation and carbon budgets: an approach for multi-project baselines

Para consultar el capítulo impreso véase el libro con la clasificación 333.7516 P7, en SIBE-Campeche

Acceso en línea sin restricciones

Human activities in forests do not always lead to deforestation, but also may affect the forest architecture and species composition, without removing permanently the forest cover. Land uses that can activate forest degradation include extensive and intensive extraction of wood for timber and fuelwood, intensive grazing, and slash-and-burn agriculture. Any of these activities would only be considered as deforestation, if it leads to a permanent or long-term removal of more than 90% of the forest cover (FAO 1997), e.g. if traditional slash-and-burn agriculture with long fallow periods shifts to permanent or short-fallow agriculture. Tracking the change of forest fragments in a case study carried out in Chiapas, Mexico, revealed that of the 213,000 ha of mature forests that were present in 1974, in 1996 79% were converted to other land-covers, such as secondary forest (=forest degradation, 46%), degraded forest (= forest degradation, 24%), and open areas (= deforestation, 9%). During the two decades, three major processes could be distinguished: (i) closed forests were transformed to degraded or secondary forest, (ii) tree and shrub fallow were converted to more permanent open areas, and (iii) open areas present in 1974 did not change to forested classes. As such, closed forests were either incorporated into the slash-and-burn system or degraded due to extraction of forest products and/or grazing of cattle. Traditional slash-and-burn systems with long fallow and short production periods were converted to permanent or short-fallow agriculture. eng

Not all carbon in ecosystems is susceptible to disappear due to human impact, e.g. stable soil carbon will remain present a long time after clearing a forest. Assuming that the level of soil carbon in permanent agriculture or pastureland would be the level of stable Cafter disturbance, the amount of carbon released shortly after a LU/LC class changes to another class (so-called vulnerable C) can thus be estimated, if the amount of SOM in permanent agriculture is known. Assigning the vulnerable C densities to the LU/LC statistics of 1974 and 1996 give an approximation of the amount of C removed or added in this period. The results of this analysis can than be used as a multi-project baseline emission scenario for carbon sequestration projects, extrapolating the past rates of change of carbon stocks into the future for the whole region (trend-based modelling). Multi-project baselines in the forestry sector need to take into account spatial variability in land-use and land-use change dynamics. The major goal for these baselines is to define the level of spatial disaggregation that is necessary or acceptable for the multi-project baseline definition. This in turn will depend on the spatial variability within the multi-project baseline area in terms of biophysical and socio-economic factors that drive the land-use change dynamics and associated GHG emissions. In this case study the density of farmers per km2 and distance to agricultural land are factors closely related to the observed land-cover change and related carbon fluxes. A simple matrix approach is therefore proposed to estimate the percent of vulnerable carbon that is susceptible to be emitted in the future according to 5 classes of farmer density and 2 classes of proximity to developed areas, to be applied as a standardized multi-project baseline approach. The approach is exemplified with a proposal for community forest management and the results are compared with regional and project-based approaches without applying the risk matrix. eng