Fitting models of multiple hypotheses to partial population data: investigating the causes of cycles in red grouse

Leslie F. New, Jason Matthiopoulos, Stephen Redpath, Stephen T. Buckland

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

There are two postulated causes for the observed periodic fluctuations (cycles) in red grouse (Lagopus lagopus scoticus). The first involves interaction with the parasitic nematode Trichostrongylus tenuis. The second invokes delayed regulation through the effect of male aggressiveness on territoriality. Empirical evidence exists to support both hypotheses, and each hypothesis has been modeled deterministically. However, little effort has gone into looking at the combined effects of the two mechanisms or formally fitting the corresponding models to field data. Here we present a model for red grouse dynamics that includes both parasites and territoriality. To explore the single and combined hypotheses, we specify three versions of this model and fit them to data using Bayesian state-space modeling, a method that allows statistical inference to be performed on mechanistic models such as ours. Output from the three models is then examined to determine their goodness of fit and the biological plausibility of the parameter values required by each to fit the population data. While all three models are capable of emulating the observed cyclic dynamics, only the model including both aggression and parasites does so under consistently realistic parameter values, providing theoretical support for the idea that both mechanisms shape red grouse cycles.

Original languageEnglish
Pages (from-to)399-412
Number of pages14
JournalThe American Naturalist
Volume174
Issue number3
DOIs
Publication statusPublished - Sep 2009

Keywords

  • aggressiveness
  • Bayesian statistics
  • cyclic population dynamics
  • Lagopus lagopus scoticus
  • state-space modeling
  • Trichostrongylus tenuis
  • lagopus-lagopus-scoticus
  • kin-facilitation hypothesis
  • host-parasite system
  • trichostrongylus-tenuis
  • territorial behavior
  • dynamics
  • stability
  • transmission
  • recruitment

Cite this

Fitting models of multiple hypotheses to partial population data : investigating the causes of cycles in red grouse. / New, Leslie F.; Matthiopoulos, Jason; Redpath, Stephen; Buckland, Stephen T.

In: The American Naturalist, Vol. 174, No. 3, 09.2009, p. 399-412.

Research output: Contribution to journalArticle

New, Leslie F. ; Matthiopoulos, Jason ; Redpath, Stephen ; Buckland, Stephen T. / Fitting models of multiple hypotheses to partial population data : investigating the causes of cycles in red grouse. In: The American Naturalist. 2009 ; Vol. 174, No. 3. pp. 399-412.
@article{bf634ca3b4514b5ba902f4182733356e,
title = "Fitting models of multiple hypotheses to partial population data: investigating the causes of cycles in red grouse",
abstract = "There are two postulated causes for the observed periodic fluctuations (cycles) in red grouse (Lagopus lagopus scoticus). The first involves interaction with the parasitic nematode Trichostrongylus tenuis. The second invokes delayed regulation through the effect of male aggressiveness on territoriality. Empirical evidence exists to support both hypotheses, and each hypothesis has been modeled deterministically. However, little effort has gone into looking at the combined effects of the two mechanisms or formally fitting the corresponding models to field data. Here we present a model for red grouse dynamics that includes both parasites and territoriality. To explore the single and combined hypotheses, we specify three versions of this model and fit them to data using Bayesian state-space modeling, a method that allows statistical inference to be performed on mechanistic models such as ours. Output from the three models is then examined to determine their goodness of fit and the biological plausibility of the parameter values required by each to fit the population data. While all three models are capable of emulating the observed cyclic dynamics, only the model including both aggression and parasites does so under consistently realistic parameter values, providing theoretical support for the idea that both mechanisms shape red grouse cycles.",
keywords = "aggressiveness, Bayesian statistics, cyclic population dynamics, Lagopus lagopus scoticus, state-space modeling, Trichostrongylus tenuis, lagopus-lagopus-scoticus, kin-facilitation hypothesis, host-parasite system, trichostrongylus-tenuis, territorial behavior, dynamics, stability, transmission, recruitment",
author = "New, {Leslie F.} and Jason Matthiopoulos and Stephen Redpath and Buckland, {Stephen T.}",
year = "2009",
month = "9",
doi = "10.1086/603625",
language = "English",
volume = "174",
pages = "399--412",
journal = "The American Naturalist",
issn = "0003-0147",
publisher = "UNIV CHICAGO PRESS",
number = "3",

}

TY - JOUR

T1 - Fitting models of multiple hypotheses to partial population data

T2 - investigating the causes of cycles in red grouse

AU - New, Leslie F.

AU - Matthiopoulos, Jason

AU - Redpath, Stephen

AU - Buckland, Stephen T.

PY - 2009/9

Y1 - 2009/9

N2 - There are two postulated causes for the observed periodic fluctuations (cycles) in red grouse (Lagopus lagopus scoticus). The first involves interaction with the parasitic nematode Trichostrongylus tenuis. The second invokes delayed regulation through the effect of male aggressiveness on territoriality. Empirical evidence exists to support both hypotheses, and each hypothesis has been modeled deterministically. However, little effort has gone into looking at the combined effects of the two mechanisms or formally fitting the corresponding models to field data. Here we present a model for red grouse dynamics that includes both parasites and territoriality. To explore the single and combined hypotheses, we specify three versions of this model and fit them to data using Bayesian state-space modeling, a method that allows statistical inference to be performed on mechanistic models such as ours. Output from the three models is then examined to determine their goodness of fit and the biological plausibility of the parameter values required by each to fit the population data. While all three models are capable of emulating the observed cyclic dynamics, only the model including both aggression and parasites does so under consistently realistic parameter values, providing theoretical support for the idea that both mechanisms shape red grouse cycles.

AB - There are two postulated causes for the observed periodic fluctuations (cycles) in red grouse (Lagopus lagopus scoticus). The first involves interaction with the parasitic nematode Trichostrongylus tenuis. The second invokes delayed regulation through the effect of male aggressiveness on territoriality. Empirical evidence exists to support both hypotheses, and each hypothesis has been modeled deterministically. However, little effort has gone into looking at the combined effects of the two mechanisms or formally fitting the corresponding models to field data. Here we present a model for red grouse dynamics that includes both parasites and territoriality. To explore the single and combined hypotheses, we specify three versions of this model and fit them to data using Bayesian state-space modeling, a method that allows statistical inference to be performed on mechanistic models such as ours. Output from the three models is then examined to determine their goodness of fit and the biological plausibility of the parameter values required by each to fit the population data. While all three models are capable of emulating the observed cyclic dynamics, only the model including both aggression and parasites does so under consistently realistic parameter values, providing theoretical support for the idea that both mechanisms shape red grouse cycles.

KW - aggressiveness

KW - Bayesian statistics

KW - cyclic population dynamics

KW - Lagopus lagopus scoticus

KW - state-space modeling

KW - Trichostrongylus tenuis

KW - lagopus-lagopus-scoticus

KW - kin-facilitation hypothesis

KW - host-parasite system

KW - trichostrongylus-tenuis

KW - territorial behavior

KW - dynamics

KW - stability

KW - transmission

KW - recruitment

U2 - 10.1086/603625

DO - 10.1086/603625

M3 - Article

VL - 174

SP - 399

EP - 412

JO - The American Naturalist

JF - The American Naturalist

SN - 0003-0147

IS - 3

ER -