Identifying and Quantifying Tree-Ring Chronology Variance Artefacts Related to Co-Occurring Changes in Growth Rate
Citation
Anthony M. Fowler, Gretel Boswijk, and Andrew Lorrey "IDENTIFYING AND QUANTIFYING TREE-RING CHRONOLOGY VARIANCE ARTEFACTS RELATED TO CO-OCCURRING CHANGES IN GROWTH RATE," Tree-Ring Research 75(2), 139-151, (7 August 2019). https://doi.org/10.3959/1536-1098-75.2.139Journal
Tree-Ring ResearchAdditional Links
https://www.treeringsociety.org/Abstract
Expectations that a warming world will be associated with more hydro-climatological extremes has motivated research exploring if an associated signal is evident in paleoclimate archives. Tree-ring chronologies are central to this work because of their high temporal resolution, but they are also potentially compromised by variance artefacts associated with the evolving composition of the chronology and with data processing. Here we present two empirical methods to identify and quantify potential artefacts related specifically to temporally varying growth rate (local level, LL): LL-based partitioning analysis and LL-based chronology stripping. The two methods were developed and tested using a multi-site New Zealand kauri (Agathis australis) living-tree data set. Our results show that the methods are complementary in terms of artefact identification and quantification, and that they can provide useful insight into causal processes when used conjointly. Our results also indicate that data pre-processing to remove LL-related artefacts may be sub-optimal, that there may be an optimal standardization that minimizes bias, and that the evolving variance of kauri master chronologies over the last 500 years is not significantly affected by LL-related artefacts.Type
Articletext
Language
enISSN
1536-1098EISSN
2162-4585ae974a485f413a2113503eed53cd6c53
10.3959/1536-1098-75.2.139