Does a one point sample adequately characterize the lake environment for paleoenvironmental calibration studies? |
| |
| Authors: | J Bunbury K Gajewski |
| |
| Institution: | (1) Department of Geography, Laboratory for Paleoclimatology and Climatology, University of Ottawa, Ottawa, ON, Canada, K1N 6N5 |
| |
| Abstract: | A major goal of paleolimnological studies is the quantification of past environmental conditions. This is accomplished by
computing transfer functions relating organism assemblages to environmental factors. The environmental data are typically
comprised of a point sample of water chemistry and other environmental parameters that are collected at the same time as a
surface sediment sample. We explore whether the year of sampling of the environmental variables affects the parameterization
of organism-environment relations, in particular chironomids, ostracodes and diatoms. Canonical correspondence analyses revealed
that the year of sampling is of secondary importance when relating the organism assemblages to environmental variables, but
only with the major explanatory variables. A chironomid-inferred bottom water temperature partial least squares transfer function
revealed similar performance statistics between the years. Taxon optima and tolerances were computed for both years using
weighted averaging, and the results are comparable. A paired t-test computed on the proxy-inferred bottom water temperature values indicated that the results between the 2 years are not
statistically different. The results of this study provide guarded optimism that the methodology of estimating transfer functions
as currently applied is not entirely determined by the particular year when the data were collected, although more case studies
are needed. |
| |
| Keywords: | Paleolimnology Calibration sets Transfer functions Canonical correspondence analysis Limnology Southwest Yukon |
| 本文献已被 SpringerLink 等数据库收录！ |
|
