Jamka, Joanna Magdalena (2019) High resolution integrated reconstruction of holocene climatic and paleoenvironmental changes in the southern tyrrhenian sea: a coda approach. [Tesi di dottorato]

Anteprima

Testo PhDJoannaJamka.pdf Download (12MB) | Anteprima

Abstract

The field of this research is the reconstruction of paleoenvironmental changes used complex methods obtained by Compositional Data Analysis statistical methods (CoDA), so in the result to develop new transfer functions (mathematical methods to obtain data to reconstruct paleoenvironmental changes) due to which we can observe the paleoclimatic and paleoenvironmental changes and volcanic activities during The Late Glacial and Holocene in the Gulf of Naples and Gulf of Salerno in collaboration with IAMC (CNR) and University of Girona (statistical methods). One of the main objectives of the thesis is to elaborate statistical methods under the framework of CoDa (Compositional data analysis: CoDa; Aitchison, 1986) to obtain the SST (sea surface temperatures) from counting of fossil assemblages. Two approaches were considered: 1) The modern analogue technique (CoDaMAT)– the revised method adopts the Aitchison distance, obtained from isometric log-ratio coordinates of relative abundances, as a natural measure of similarity among assemblages. The number of analogues from which obtain the estimates was determined through leave-one-out verification of modern assemblages. Mean distances and local outlier factor are considered to evaluate the quality of palaeoestimates. The method has been tested on Atlantic Ocean and Mediterranean planktonic foraminiferal assemblages to reconstruct past sea surface temperatures (SST). The MAT compares fossil assemblages with modern ones using a distance measure or a similarity coefficient. The palaeoenvironmental estimates are obtained from the environmental parameters measured at the location of the most similar modern assemblages. For each fossil samples the nearest modern ones are found by adopting an appropriate distance (d). 2) Regression-based methods - standard and robust Partial Least Squares Regression (PLSR) and Principal Component Regression (PCR), applied to the log-ratio coordinates of percentage data of Atlantic Ocean and Mediterranean Sea planktonic foraminifera assemblages. Due to the presence of groups, it was preferred to model separately high latitude and mid to low latitude assemblages. The application of the obtained transfer function involves cluster, MANOVA and discriminant analysis applied to the log-ratio transformed fossil assemblage’s composition.

Downloads

Actions (login required)

Modifica documento