Advanced

Interaction assessments in correlated and autocorrelated environments

Ripa, Jörgen LU and Ives, Anthony R. (2007) In The impact of environmental variability on ecological systems. The Peter Yodzis Fundamental Ecology Series Vol. 2 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:
author
organization
publishing date
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
in
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
2007-11-28 12:36:04
date last changed
2016-04-16 08:30:54
@misc{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.},
  editor       = {Vasseur, David A. and McCann, Kevin S.},
  isbn         = {978-1-4020-5850-9},
  keyword      = {food web dynamics,environmental stochasticity,multivariate time series analysis,interaction assessment,autocorrelation,correlation,community dynamics},
  language     = {eng},
  pages        = {111--131},
  publisher    = {ARRAY(0x93f1948)},
  series       = {The impact of environmental variability on ecological systems. The Peter Yodzis Fundamental Ecology Series Vol. 2},
  title        = {Interaction assessments in correlated and autocorrelated environments},
  volume       = {2},
  year         = {2007},
}