Keywords
redd+, deforestation, forest degradation, bayesian networks, participatory research
Start Date
1-7-2012 12:00 AM
Abstract
Understanding local land-use trajectories is of vital importance for designing and implementing REDD+ projects. But such understanding is often marred by difficulties to relate to the effects of broad-scale underlying causes on local land-use responses. In this research, we used participatory data collection methods to calibrate Bayesian networks in order to study land-use change at the household and village level. We extracted historic land-use change trajectories from focus group discussions to identify the underlying causes of major land-use transitions with a focus on changes in forest cover and forest quality. These insights were used to construct influence diagrams consisting of the underlying and proximate drivers of change at different scales and for all sites. We studied six villages in mainland Southeast Asia, two in Laos, Vietnam, and China, respectively. The results suggest that changes in forest land use were shaped by a similar set of influencing factors, but with fundamental differences in the significance and importance of specific causal structures across the six villages of the three countries. REDD+ projects ought to consider such variation with a flexible implementation structure that accounts for site-specific dynamics.
Participatory Bayesian networks reveal site-specific causes of land-use trajectories in Southeast Asia
Understanding local land-use trajectories is of vital importance for designing and implementing REDD+ projects. But such understanding is often marred by difficulties to relate to the effects of broad-scale underlying causes on local land-use responses. In this research, we used participatory data collection methods to calibrate Bayesian networks in order to study land-use change at the household and village level. We extracted historic land-use change trajectories from focus group discussions to identify the underlying causes of major land-use transitions with a focus on changes in forest cover and forest quality. These insights were used to construct influence diagrams consisting of the underlying and proximate drivers of change at different scales and for all sites. We studied six villages in mainland Southeast Asia, two in Laos, Vietnam, and China, respectively. The results suggest that changes in forest land use were shaped by a similar set of influencing factors, but with fundamental differences in the significance and importance of specific causal structures across the six villages of the three countries. REDD+ projects ought to consider such variation with a flexible implementation structure that accounts for site-specific dynamics.
