The response of pine trees to climate in the southern French Alps and the palaeoclimatic potential of stable carbon isotope ratios from treerings

• Main Author: Gagen, M. H.
• Published: Swansea University 2000
• Subjects: 560
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.637018

This thesis is grounded in Quaternary and global change science and considers the application of proxy analogues to reconstruct past climate. Latewood width, wood density and latewood holo cellulose δ13</SUP>C series were compiled from treeline pine trees from the southern French Alps, with the aim of comparing and contrasting the types of climate information preserved in the three proxies and assessing the potential of δ<SUP>13</SUP>C for palaeoclimate reconstruction. Two main research questions were identified. First, what climate parameters might be reconstructed from the annual δ<SUP>13</SUP>C series? Second, how useful is the δ<SUP>13</SUP>C calibration for reconstructing palaeoclimate compared to ringwidth and density series? The calibration stage formed the bulk of the analysis. It was revealed that δ<SUP>13</SUP>C ratios are strongly controlled by growth season moisture stress, in a relationship that appears to have a high degree of temporal stability. The potential exists, therefore, for extracting a strong moisture-stress signal form long δ<SUP>13</SUP>C series from this region. It is argued that, lower frequency signal strength is often dramatically reduced or lost when short ringwidth and density series are standardised to remove growth trends. This thesis presents evidence to suggest that sub- to decadal frequency climate signals can be retained in δ<SUP>13</SUP>C series. It is concluded that there is great potential for reconstructing palaeoclimate with δ<SUP>13</SUP>C treering series. δ<SUP>13</SUP>C ratios are fixed by a simple biological process that is well understood, and are not greatly affected by site and stand characteristics, making the series far less 'noisy' than equivalent growth proxy data.