Estimating environmental effects on population dynamics: consequences of observation error
(2009) In Oikos 118(5). p.675680 Abstract
 Within the paradigm of population dynamics a central task is to identify environmental factors affecting population change and to estimate the strength of these effects. We here investigate the impact of observation errors in measurements of population densities on estimates of environmental effects. Adding observation errors may change the autocorrelation of a population time series with potential consequences for estimates of effects of autocorrelated environmental covariates. Using Monte Carlo simulations, we compare the performance of maximum likelihood estimates from three stochastic versions of the Gompertz model (loglinear first order autoregressive model), assuming 1) process error only, 2) observation error only, and 3) both... (More)
 Within the paradigm of population dynamics a central task is to identify environmental factors affecting population change and to estimate the strength of these effects. We here investigate the impact of observation errors in measurements of population densities on estimates of environmental effects. Adding observation errors may change the autocorrelation of a population time series with potential consequences for estimates of effects of autocorrelated environmental covariates. Using Monte Carlo simulations, we compare the performance of maximum likelihood estimates from three stochastic versions of the Gompertz model (loglinear first order autoregressive model), assuming 1) process error only, 2) observation error only, and 3) both process and observation error (the linear statespace model on logscale). We also simulated population dynamics using the Ricker model, and evaluated the corresponding maximum likelihood estimates for process error models. When there is observation error in the data and the considered environmental variable is strongly autocorrelated, its estimated effect is likely to be biased when using process error models. The environmental effect is overestimated when the sign of the autocorrelations of the intrinsic dynamics and the environment are the same and underestimated when the signs differ. With nonautocorrelated environmental covariates, process error models produce fairly exact point estimates as well as reliable confidence intervals for environmental effects. In all scenarios, observation error models produce unbiased estimates with reasonable precision, but confidence intervals derived from the likelihood profiles are far too optimistic if there is process error present. The safest approach is to use statespace models in presence of observation error. These are factors worthwhile to consider when interpreting earlier empirical results on population time series, and in future studies, we recommend choosing carefully the modelling approach with respect to intrinsic population dynamics and covariate autocorrelation. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/record/1426326
 author
 Linden, Andreas and Knape, Jonas ^{LU}
 organization
 publishing date
 2009
 type
 Contribution to journal
 publication status
 published
 subject
 in
 Oikos
 volume
 118
 issue
 5
 pages
 675  680
 publisher
 WileyBlackwell
 external identifiers

 wos:000265711500004
 scopus:65449165039
 ISSN
 16000706
 DOI
 10.1111/j.16000706.2008.17250.x
 language
 English
 LU publication?
 yes
 id
 9837a013200240ce93f00bf266da9173 (old id 1426326)
 date added to LUP
 20090629 15:39:47
 date last changed
 20180107 06:24:54
@article{9837a013200240ce93f00bf266da9173, abstract = {Within the paradigm of population dynamics a central task is to identify environmental factors affecting population change and to estimate the strength of these effects. We here investigate the impact of observation errors in measurements of population densities on estimates of environmental effects. Adding observation errors may change the autocorrelation of a population time series with potential consequences for estimates of effects of autocorrelated environmental covariates. Using Monte Carlo simulations, we compare the performance of maximum likelihood estimates from three stochastic versions of the Gompertz model (loglinear first order autoregressive model), assuming 1) process error only, 2) observation error only, and 3) both process and observation error (the linear statespace model on logscale). We also simulated population dynamics using the Ricker model, and evaluated the corresponding maximum likelihood estimates for process error models. When there is observation error in the data and the considered environmental variable is strongly autocorrelated, its estimated effect is likely to be biased when using process error models. The environmental effect is overestimated when the sign of the autocorrelations of the intrinsic dynamics and the environment are the same and underestimated when the signs differ. With nonautocorrelated environmental covariates, process error models produce fairly exact point estimates as well as reliable confidence intervals for environmental effects. In all scenarios, observation error models produce unbiased estimates with reasonable precision, but confidence intervals derived from the likelihood profiles are far too optimistic if there is process error present. The safest approach is to use statespace models in presence of observation error. These are factors worthwhile to consider when interpreting earlier empirical results on population time series, and in future studies, we recommend choosing carefully the modelling approach with respect to intrinsic population dynamics and covariate autocorrelation.}, author = {Linden, Andreas and Knape, Jonas}, issn = {16000706}, language = {eng}, number = {5}, pages = {675680}, publisher = {WileyBlackwell}, series = {Oikos}, title = {Estimating environmental effects on population dynamics: consequences of observation error}, url = {http://dx.doi.org/10.1111/j.16000706.2008.17250.x}, volume = {118}, year = {2009}, }