Deciphering the effect of climate change and separating the influence of confounding factors in sediment core records using additive models
We describe a new approach to modeling sediment core records, one that uses additive models (AMs) incorporating a serial correlation structure to model residual autocorrelation. Species assemblages, for example, are reduced to ordination axis scores that capture major changes in the data through tim...
| Published in: | Limnology and Oceanography |
|---|---|
| Main Authors: | , |
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
Wiley
2009
|
| Subjects: | |
| Online Access: | http://dx.doi.org/10.4319/lo.2009.54.6_part_2.2529 https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.4319%2Flo.2009.54.6_part_2.2529 https://aslopubs.onlinelibrary.wiley.com/doi/pdf/10.4319/lo.2009.54.6_part_2.2529 |
| Summary: | We describe a new approach to modeling sediment core records, one that uses additive models (AMs) incorporating a serial correlation structure to model residual autocorrelation. Species assemblages, for example, are reduced to ordination axis scores that capture major changes in the data through time. Each set of axis scores is then modeled using an AM, where covariates represent forcing variables (e.g., tree‐ring‐inferred temperature or proxies for atmospheric deposition) and/or trend and, where necessary, periodic components. The effect of forcing variables on species composition through time can be determined via the contribution each covariate makes to the fitted model, which can be used to separate effects due to competing forcing variables. We illustrate the approach using data from Kassjön, northern Sweden, and Loch Coire Fionnaraich, northwest Scotland. |
|---|