Interaction assessments in correlated and autocorrelated environments
(2007) 2. p.111-131- Abstract
- Natural food webs are embedded in a variable environment, which causes population densities to fluctuate, despite a potential stable equilibrium. Population interactions as well as the characteristics of the environmental fluctuations determine the resulting population dynamics. Populations sensitive to the same kind of environmental disturbances will show correlated responses in their respective growth rates. Such 'environmental correlation' between species can have profound effects on the populations' dynamics, e.g. generating a positive correlation between the abundances of two competitors, which makes a direct correlation a highly inappropriate measure of population interactions. However, multivariate time series analysis will still... (More)
- Natural food webs are embedded in a variable environment, which causes population densities to fluctuate, despite a potential stable equilibrium. Population interactions as well as the characteristics of the environmental fluctuations determine the resulting population dynamics. Populations sensitive to the same kind of environmental disturbances will show correlated responses in their respective growth rates. Such 'environmental correlation' between species can have profound effects on the populations' dynamics, e.g. generating a positive correlation between the abundances of two competitors, which makes a direct correlation a highly inappropriate measure of population interactions. However, multivariate time series analysis will still identify and quantify population interactions correctly. The picture is more complicated if the environmental fluctuations are correlated over time – environmental autocorrelation causes biases in interaction assessments and possibly falsely identified delayed interactions. We present approximate expressions for the estimation bias, which show that the bias is the weakest when food web dynamics are close to unstable. In the absence of close to unstable dynamics the only way avoid this estimation error is to incorporate the most important environmental drivers as covariates in the time series analysis. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/629236
- author
- Ripa, Jörgen LU and Ives, Anthony R.
- organization
- publishing date
- 2007
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- food web dynamics, environmental stochasticity, multivariate time series analysis, interaction assessment, autocorrelation, correlation, community dynamics
- host publication
- The impact of environmental variability on ecological systems. The Peter Yodzis Fundamental Ecology Series Vol. 2
- editor
- Vasseur, David A. and McCann, Kevin S.
- volume
- 2
- pages
- 111 - 131
- publisher
- Springer
- external identifiers
-
- wos:000248254100006
- ISBN
- 978-1-4020-5850-9
- language
- English
- LU publication?
- yes
- id
- b3844d4d-e429-464c-8485-e7bc0ad042dc (old id 629236)
- date added to LUP
- 2016-04-04 10:54:01
- date last changed
- 2021-01-04 18:16:34
@inbook{b3844d4d-e429-464c-8485-e7bc0ad042dc, abstract = {{Natural food webs are embedded in a variable environment, which causes population densities to fluctuate, despite a potential stable equilibrium. Population interactions as well as the characteristics of the environmental fluctuations determine the resulting population dynamics. Populations sensitive to the same kind of environmental disturbances will show correlated responses in their respective growth rates. Such 'environmental correlation' between species can have profound effects on the populations' dynamics, e.g. generating a positive correlation between the abundances of two competitors, which makes a direct correlation a highly inappropriate measure of population interactions. However, multivariate time series analysis will still identify and quantify population interactions correctly. The picture is more complicated if the environmental fluctuations are correlated over time – environmental autocorrelation causes biases in interaction assessments and possibly falsely identified delayed interactions. We present approximate expressions for the estimation bias, which show that the bias is the weakest when food web dynamics are close to unstable. In the absence of close to unstable dynamics the only way avoid this estimation error is to incorporate the most important environmental drivers as covariates in the time series analysis.}}, author = {{Ripa, Jörgen and Ives, Anthony R.}}, booktitle = {{The impact of environmental variability on ecological systems. The Peter Yodzis Fundamental Ecology Series Vol. 2}}, editor = {{Vasseur, David A. and McCann, Kevin S.}}, isbn = {{978-1-4020-5850-9}}, keywords = {{food web dynamics; environmental stochasticity; multivariate time series analysis; interaction assessment; autocorrelation; correlation; community dynamics}}, language = {{eng}}, pages = {{111--131}}, publisher = {{Springer}}, title = {{Interaction assessments in correlated and autocorrelated environments}}, volume = {{2}}, year = {{2007}}, }