A1463
Title: Finite mixture modeling for the analysis of spatiotemporal aspects in dendrochronology
Authors: Volodymyr Melnykov - The University of Alabama (United States) [presenting]
Lingge Wang - University of Alabama (United States)
Abstract: The objective of dendrochronology is to study environmental changes as well as date events and archaeological artifacts based on the sequences of annual rings extracted from tree trunks. Dendrochronology provides a unique yet scientifically sound approach to the analysis of events that occurred hundreds of years ago. The analysis of tree ring data is not trivial due to the presence of temporal relationships as well as the uncertainty associated with the location and lifetime of many trees. A finite mixture model is proposed to address the heterogeneity in the tree ring data. Unlike existing approaches discussed in the current literature, the proposed model takes into account both sources of uncertainty. The developed model is applied to the analysis of data publicly available from the international tree ring data bank, with promising results.
