Plant-animal interactions in random environments: Habitat-stage elasticity, seed predators, and hurricanes

Carol C Horvitz, Shripad Tuljapurkar, John B. Pascarella

Research output: Contribution to journalArticle

40 Citations (Scopus)

Abstract

When environments change stochastically, the question arises how to evaluate the effects of a plant-animal interaction on the fitness of the plant, where plant fitness is measured by the stochastic growth rate. We develop the concept of habitat-stage elasticity, EβS, which gives the proportional sensitivity of the stochastic growth rate to perturbations of stage transition rates in each state (β) of the habitat. We employ it to understand why a specialist gall-making seed predator has relatively low impact on the fitness of a subtropical shrub. The plant lives in a forest characterized by patchy, recurrent disturbances caused by hurricanes. Both predation rate and plant demography vary with canopy openness. In the most closed-canopy state, the seed predator destroys 90% of the fruits, and demographic quality, the dominant eigenvalue of that state's matrix, is low, while in the most open-canopy state, predation is negligible and demographic quality high. The seed predator is locally extirpated by strong hurricanes, recolonization taking several years. The effect of the predator on the stochastic growth rate is negligible at both low and high hurricane frequency. Its effect peaks (6%) at an intermediate hurricane frequency. The stochastic growth rate varies in its sensitivity to the predator in different states of the habitat due to a product of two factors: the frequency of the state in the environment and the contribution of fecundity to its elasticity. The latter factor encapsulates the expected sequence of future states of the habitat. In our system, the contribution of fecundity to elasticity of the darkest state increases with hurricane frequency, even though the probability of encountering that state decreases, because today's dark habitats are more likely to become lighter ones. The contribution of fecundity to the two lightest states does not vary with hurricane frequency. In contrast, its contribution in intermediate states at intermediate hurricane frequencies is most dynamic, since uncolonized states may become colonized states and insensitive states may become sensitive states. The effects of an animal on plant fitness is determined by the disturbance regime in addition to its impact on vital rates in each environmental state.

Original languageEnglish
Pages (from-to)3312-3322
Number of pages11
JournalEcology
Volume86
Issue number12
DOIs
StatePublished - Dec 1 2005

Fingerprint

hurricanes
elasticity (mechanics)
hurricane
elasticity
predator
seed
predators
animal
habitat
habitats
seeds
fitness
animals
fecundity
canopy
demographic statistics
predation
disturbance
gall
eigenvalue

Keywords

  • Ardisia
  • Effects of animals on plant fitness
  • Environment-specific elasticity
  • Gall-making moths
  • Habitat elasticity
  • Periploca
  • Stochastic growth rate
  • Stochastic sequence

ASJC Scopus subject areas

  • Ecology

Cite this

Plant-animal interactions in random environments : Habitat-stage elasticity, seed predators, and hurricanes. / Horvitz, Carol C; Tuljapurkar, Shripad; Pascarella, John B.

In: Ecology, Vol. 86, No. 12, 01.12.2005, p. 3312-3322.

Research output: Contribution to journalArticle

Horvitz, Carol C ; Tuljapurkar, Shripad ; Pascarella, John B. / Plant-animal interactions in random environments : Habitat-stage elasticity, seed predators, and hurricanes. In: Ecology. 2005 ; Vol. 86, No. 12. pp. 3312-3322.
@article{8b36040c27ab47a880d3eac23df26f39,
title = "Plant-animal interactions in random environments: Habitat-stage elasticity, seed predators, and hurricanes",
abstract = "When environments change stochastically, the question arises how to evaluate the effects of a plant-animal interaction on the fitness of the plant, where plant fitness is measured by the stochastic growth rate. We develop the concept of habitat-stage elasticity, EβS, which gives the proportional sensitivity of the stochastic growth rate to perturbations of stage transition rates in each state (β) of the habitat. We employ it to understand why a specialist gall-making seed predator has relatively low impact on the fitness of a subtropical shrub. The plant lives in a forest characterized by patchy, recurrent disturbances caused by hurricanes. Both predation rate and plant demography vary with canopy openness. In the most closed-canopy state, the seed predator destroys 90{\%} of the fruits, and demographic quality, the dominant eigenvalue of that state's matrix, is low, while in the most open-canopy state, predation is negligible and demographic quality high. The seed predator is locally extirpated by strong hurricanes, recolonization taking several years. The effect of the predator on the stochastic growth rate is negligible at both low and high hurricane frequency. Its effect peaks (6{\%}) at an intermediate hurricane frequency. The stochastic growth rate varies in its sensitivity to the predator in different states of the habitat due to a product of two factors: the frequency of the state in the environment and the contribution of fecundity to its elasticity. The latter factor encapsulates the expected sequence of future states of the habitat. In our system, the contribution of fecundity to elasticity of the darkest state increases with hurricane frequency, even though the probability of encountering that state decreases, because today's dark habitats are more likely to become lighter ones. The contribution of fecundity to the two lightest states does not vary with hurricane frequency. In contrast, its contribution in intermediate states at intermediate hurricane frequencies is most dynamic, since uncolonized states may become colonized states and insensitive states may become sensitive states. The effects of an animal on plant fitness is determined by the disturbance regime in addition to its impact on vital rates in each environmental state.",
keywords = "Ardisia, Effects of animals on plant fitness, Environment-specific elasticity, Gall-making moths, Habitat elasticity, Periploca, Stochastic growth rate, Stochastic sequence",
author = "Horvitz, {Carol C} and Shripad Tuljapurkar and Pascarella, {John B.}",
year = "2005",
month = "12",
day = "1",
doi = "10.1890/04-1187",
language = "English",
volume = "86",
pages = "3312--3322",
journal = "Ecology",
issn = "0012-9658",
publisher = "Ecological Society of America",
number = "12",

}

TY - JOUR

T1 - Plant-animal interactions in random environments

T2 - Habitat-stage elasticity, seed predators, and hurricanes

AU - Horvitz, Carol C

AU - Tuljapurkar, Shripad

AU - Pascarella, John B.

PY - 2005/12/1

Y1 - 2005/12/1

N2 - When environments change stochastically, the question arises how to evaluate the effects of a plant-animal interaction on the fitness of the plant, where plant fitness is measured by the stochastic growth rate. We develop the concept of habitat-stage elasticity, EβS, which gives the proportional sensitivity of the stochastic growth rate to perturbations of stage transition rates in each state (β) of the habitat. We employ it to understand why a specialist gall-making seed predator has relatively low impact on the fitness of a subtropical shrub. The plant lives in a forest characterized by patchy, recurrent disturbances caused by hurricanes. Both predation rate and plant demography vary with canopy openness. In the most closed-canopy state, the seed predator destroys 90% of the fruits, and demographic quality, the dominant eigenvalue of that state's matrix, is low, while in the most open-canopy state, predation is negligible and demographic quality high. The seed predator is locally extirpated by strong hurricanes, recolonization taking several years. The effect of the predator on the stochastic growth rate is negligible at both low and high hurricane frequency. Its effect peaks (6%) at an intermediate hurricane frequency. The stochastic growth rate varies in its sensitivity to the predator in different states of the habitat due to a product of two factors: the frequency of the state in the environment and the contribution of fecundity to its elasticity. The latter factor encapsulates the expected sequence of future states of the habitat. In our system, the contribution of fecundity to elasticity of the darkest state increases with hurricane frequency, even though the probability of encountering that state decreases, because today's dark habitats are more likely to become lighter ones. The contribution of fecundity to the two lightest states does not vary with hurricane frequency. In contrast, its contribution in intermediate states at intermediate hurricane frequencies is most dynamic, since uncolonized states may become colonized states and insensitive states may become sensitive states. The effects of an animal on plant fitness is determined by the disturbance regime in addition to its impact on vital rates in each environmental state.

AB - When environments change stochastically, the question arises how to evaluate the effects of a plant-animal interaction on the fitness of the plant, where plant fitness is measured by the stochastic growth rate. We develop the concept of habitat-stage elasticity, EβS, which gives the proportional sensitivity of the stochastic growth rate to perturbations of stage transition rates in each state (β) of the habitat. We employ it to understand why a specialist gall-making seed predator has relatively low impact on the fitness of a subtropical shrub. The plant lives in a forest characterized by patchy, recurrent disturbances caused by hurricanes. Both predation rate and plant demography vary with canopy openness. In the most closed-canopy state, the seed predator destroys 90% of the fruits, and demographic quality, the dominant eigenvalue of that state's matrix, is low, while in the most open-canopy state, predation is negligible and demographic quality high. The seed predator is locally extirpated by strong hurricanes, recolonization taking several years. The effect of the predator on the stochastic growth rate is negligible at both low and high hurricane frequency. Its effect peaks (6%) at an intermediate hurricane frequency. The stochastic growth rate varies in its sensitivity to the predator in different states of the habitat due to a product of two factors: the frequency of the state in the environment and the contribution of fecundity to its elasticity. The latter factor encapsulates the expected sequence of future states of the habitat. In our system, the contribution of fecundity to elasticity of the darkest state increases with hurricane frequency, even though the probability of encountering that state decreases, because today's dark habitats are more likely to become lighter ones. The contribution of fecundity to the two lightest states does not vary with hurricane frequency. In contrast, its contribution in intermediate states at intermediate hurricane frequencies is most dynamic, since uncolonized states may become colonized states and insensitive states may become sensitive states. The effects of an animal on plant fitness is determined by the disturbance regime in addition to its impact on vital rates in each environmental state.

KW - Ardisia

KW - Effects of animals on plant fitness

KW - Environment-specific elasticity

KW - Gall-making moths

KW - Habitat elasticity

KW - Periploca

KW - Stochastic growth rate

KW - Stochastic sequence

UR - http://www.scopus.com/inward/record.url?scp=30644457334&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30644457334&partnerID=8YFLogxK

U2 - 10.1890/04-1187

DO - 10.1890/04-1187

M3 - Article

AN - SCOPUS:30644457334

VL - 86

SP - 3312

EP - 3322

JO - Ecology

JF - Ecology

SN - 0012-9658

IS - 12

ER -