Modeling present and future aboveground biomass of evergreen broadleaf forests in Vietnam.
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
Forest reforestation and degradation have both occured over the past five decades in Vietnam. To cope with this problem, the Vietnamese government has established plans and set up strategies to reduce deforestation and forest degradation. It also launched national support for the conservation and sustainable management of forests, as well as enhancement of forest carbon stocks in developing countries (REDD+) programs. A limited number of studies have attempted to establish a database of aboveground biomass of evergreen broadleaf forests in Vietnam, as well as growth models such as height versus diameter at breast height (H-D) models, basal area (G) increment models, and above ground biomass (AGB) increment models. In addition, the information describing the relationships between environmental indicators and tree species distributions was also insufficient, leading to potential failure of reforestation and rehabilitation projects.
This study examined the correlation between environmental indicators and tree species distributions. It also sought to develop H-D models based on the outcomes of grouping tree species into different groups, and model the relationship between G and AGB increments with other environment factors and stand characteristics.
The study utilized data collected from Forest Inventory and Planning Insititution (FIPI) and valiation data from Vietnamese Academy of Forest Sciences (VAFS). It then employed ordination analysis to analyse the correlation between tree species groups and environmental factors. In addition, previous H-D functions applied in past research on tropical forests were used to develop H-D models for this particular study. Validation procedures were used to compare selected H-D models with other H-D models applied in the same forest types. Lastly, a decision tree approach was adopted to select the climatic, soil, and stand variables that were most likely to be useful for the development of G and AGB increment models.
The findings were that there was a correlation between solar radiation, depth to bedrock, clay content, temperature and rainfall with tree species distributions. Nine selected H-D models for nine respective tree species groups were less biased and more precise comprared to two given H-D models in a validation procedure. Finally, both G increment models and AGB increment models were developed, in which these climatic, soil, and stand variables were directly added. The study was intended to contribute valuable data and relevant models for the benefit of forest managers and administrators who could use the results to effectively carry out the process of reforestation, REDD+ projects, and national forest inventories programs at minimal cost in timely and efficient manners.