Abstract
Functional diversity or trait diversity refers to the diversity of the value and range of species traits, and is a rapidly growing research topic in ecology. Functional diversity is essential to assess ecosystem processes and their responses to environmental stress or disturbance. The higher value the functional diversity is, the more dissimilar the characteristics among species are, and as a result the whole ecosystem can better adapt to environmental changes. Functional diversity is typically quantified by using species abundances and species trait values. Many functional diversity measures have been proposed in the literature. The thesis includes two parts. The first part focuses on modifying Chao et al. (2018) abundance-based functional diversity measures to replicated incidence-based data under quadrat sampling for a single ecosystem. The proposed measures and estimators are formulated based on attribute diversity (a generalization of Hill numbers) in terms of diversity order q 0 and any positive level of threshold of functional distinctiveness. Statistical estimators of the proposed incidence-based measures are proposed and their variances are estimated by a bootstrap method. The second part focuses on the functional diversity decomposition and estimation under multiple communities for both individual-based abundance data and incidence data. Functional (dis)similarity indices are derived as transformations of functional beta diversities. Simulation results are presented to compare the proposed estimators with the conventional empirical diversities; the proposed estimators exhibit substantial improvement in terms of bias and RMSE. Real data sets are used to illustrate the proposed functional diversity measures and related functional dissimilarity indices. To facilitate all computations, online software “FunD” is developed with R language and Shiny package.