Methods for estimating volume, biomass and tree species diversity using field inventory and airborne laser scanning in the tropical forests of Tanzania.

PhD Thesis

Deforestation and forest degradation in the tropical countries have reduced the extent of forest and woodlands, which conserve biodiversity, provide essential resources to people and help in mitigating climate change through carbon sequestration. Forest conservation projects need methods for estimating tree species diversity to effectively generate information necessary for implementing biodiversity management plans, while greenhouse gas reduction programmes such REDD* (Reducing Emissions from Deforestation and Forest Degradation) require robust methods to estimate volume and aboveground biomass (AGB). Such methods are also needed in the context of general forest management planning. The four papers included in this thesis are aimed to test and evaluate methods for estimating volume. AGB. and tree species diversity using field and remotely sensed data in the tropical forests and woodlands of Tanzania. In paper 1. tree models for estimating total, merchantable stem, and branch volume applicable for the entire miombo woodlands of Tanzania were developed. In Paper II. Ill. and IV the potential of airborne laser scanning (AI.S) data for predicting AGB and measures of tree species diversity was tested and evaluated. The results have shown that ALS data can be used for predicting AGB with reasonable accuracy by using both parametric and nonparametric approaches. Effects of plot size on the AGB estimates were investigated and the results indicated that the prediction accuracy of AGB in ALS-assisted inventories improved as the plot size increased. Finally, the results showed that measures of tree species diversity and particularly tree species richness and Shannon diversity index, can potentially be predicted by using ALS data.

The project entitled "Climate Change Impacts. Adaptation, and Mitigation in Tanzania" supported by the government of Norway.