The relative importance of climate and habitat in determining the distributions of species at different spatial scales

a case study with ground beetles in Great Britain

Phillipa K Gillingham, Stephen C F Palmer, Brian Huntley, William E Kunin, Joseph D Chipperfield, Chris D Thomas

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

Experimental studies have shown that many species show preferences for different climatic conditions, or may die in unsuitable conditions. Climate envelope models have been used frequently in recent years to predict the presence and absence of species at large spatial scales. However, many authors have postulated that the distributions of species at smaller spatial scales are determined by factors such as habitat availability and biotic interactions. Climatic effects are often assumed by modellers to be unimportant at fine resolutions, but few studies have actually tested this. We sampled the distributions of 20 beetle species of the family Carabidae across three study sites by pitfall trapping, and at the national scale from monitoring data. Statistical models were constructed to determine which of two sets of environmental variables (temperature or broad habitat type) best accounted for the observed data at the three sites and at the national (Great Britain) scale. High-resolution temperature variables frequently produced better models (as determined by AIC) than habitat features when modelling the distributions of species at a local scale, within the three study sites. Conversely, habitat was always a better predictor than temperature when describing species distributions at a coarse scale within Great Britain. Northerly species were most likely to occur in cool micro-sites within the study sites, whereas southerly species were most likely to occur in warm micro-sites. Effects of microclimate were not limited to species at the edges of their distribution, and fine-resolution temperature surfaces should therefore ideally be utilised when undertaking climate-envelope modelling.

Original languageEnglish
Pages (from-to)831-838
Number of pages8
JournalEcography
Volume35
Issue number9
DOIs
Publication statusPublished - Sep 2012

Keywords

  • land-cover
  • bioclimatic models
  • range
  • assemblages
  • coleoptera
  • heterogeneity
  • conservation
  • communities
  • adaptation
  • resolution

Cite this

The relative importance of climate and habitat in determining the distributions of species at different spatial scales : a case study with ground beetles in Great Britain. / Gillingham, Phillipa K; Palmer, Stephen C F; Huntley, Brian; Kunin, William E; Chipperfield, Joseph D; Thomas, Chris D.

In: Ecography, Vol. 35, No. 9, 09.2012, p. 831-838.

Research output: Contribution to journalArticle

Gillingham, Phillipa K ; Palmer, Stephen C F ; Huntley, Brian ; Kunin, William E ; Chipperfield, Joseph D ; Thomas, Chris D. / The relative importance of climate and habitat in determining the distributions of species at different spatial scales : a case study with ground beetles in Great Britain. In: Ecography. 2012 ; Vol. 35, No. 9. pp. 831-838.
@article{50f3c00184a14743a4f1b1473a7f2d78,
title = "The relative importance of climate and habitat in determining the distributions of species at different spatial scales: a case study with ground beetles in Great Britain",
abstract = "Experimental studies have shown that many species show preferences for different climatic conditions, or may die in unsuitable conditions. Climate envelope models have been used frequently in recent years to predict the presence and absence of species at large spatial scales. However, many authors have postulated that the distributions of species at smaller spatial scales are determined by factors such as habitat availability and biotic interactions. Climatic effects are often assumed by modellers to be unimportant at fine resolutions, but few studies have actually tested this. We sampled the distributions of 20 beetle species of the family Carabidae across three study sites by pitfall trapping, and at the national scale from monitoring data. Statistical models were constructed to determine which of two sets of environmental variables (temperature or broad habitat type) best accounted for the observed data at the three sites and at the national (Great Britain) scale. High-resolution temperature variables frequently produced better models (as determined by AIC) than habitat features when modelling the distributions of species at a local scale, within the three study sites. Conversely, habitat was always a better predictor than temperature when describing species distributions at a coarse scale within Great Britain. Northerly species were most likely to occur in cool micro-sites within the study sites, whereas southerly species were most likely to occur in warm micro-sites. Effects of microclimate were not limited to species at the edges of their distribution, and fine-resolution temperature surfaces should therefore ideally be utilised when undertaking climate-envelope modelling.",
keywords = "land-cover, bioclimatic models, range, assemblages, coleoptera, heterogeneity, conservation, communities, adaptation, resolution",
author = "Gillingham, {Phillipa K} and Palmer, {Stephen C F} and Brian Huntley and Kunin, {William E} and Chipperfield, {Joseph D} and Thomas, {Chris D}",
year = "2012",
month = "9",
doi = "10.1111/j.1600-0587.2011.07434.x",
language = "English",
volume = "35",
pages = "831--838",
journal = "Ecography",
issn = "0906-7590",
publisher = "Wiley",
number = "9",

}

TY - JOUR

T1 - The relative importance of climate and habitat in determining the distributions of species at different spatial scales

T2 - a case study with ground beetles in Great Britain

AU - Gillingham, Phillipa K

AU - Palmer, Stephen C F

AU - Huntley, Brian

AU - Kunin, William E

AU - Chipperfield, Joseph D

AU - Thomas, Chris D

PY - 2012/9

Y1 - 2012/9

N2 - Experimental studies have shown that many species show preferences for different climatic conditions, or may die in unsuitable conditions. Climate envelope models have been used frequently in recent years to predict the presence and absence of species at large spatial scales. However, many authors have postulated that the distributions of species at smaller spatial scales are determined by factors such as habitat availability and biotic interactions. Climatic effects are often assumed by modellers to be unimportant at fine resolutions, but few studies have actually tested this. We sampled the distributions of 20 beetle species of the family Carabidae across three study sites by pitfall trapping, and at the national scale from monitoring data. Statistical models were constructed to determine which of two sets of environmental variables (temperature or broad habitat type) best accounted for the observed data at the three sites and at the national (Great Britain) scale. High-resolution temperature variables frequently produced better models (as determined by AIC) than habitat features when modelling the distributions of species at a local scale, within the three study sites. Conversely, habitat was always a better predictor than temperature when describing species distributions at a coarse scale within Great Britain. Northerly species were most likely to occur in cool micro-sites within the study sites, whereas southerly species were most likely to occur in warm micro-sites. Effects of microclimate were not limited to species at the edges of their distribution, and fine-resolution temperature surfaces should therefore ideally be utilised when undertaking climate-envelope modelling.

AB - Experimental studies have shown that many species show preferences for different climatic conditions, or may die in unsuitable conditions. Climate envelope models have been used frequently in recent years to predict the presence and absence of species at large spatial scales. However, many authors have postulated that the distributions of species at smaller spatial scales are determined by factors such as habitat availability and biotic interactions. Climatic effects are often assumed by modellers to be unimportant at fine resolutions, but few studies have actually tested this. We sampled the distributions of 20 beetle species of the family Carabidae across three study sites by pitfall trapping, and at the national scale from monitoring data. Statistical models were constructed to determine which of two sets of environmental variables (temperature or broad habitat type) best accounted for the observed data at the three sites and at the national (Great Britain) scale. High-resolution temperature variables frequently produced better models (as determined by AIC) than habitat features when modelling the distributions of species at a local scale, within the three study sites. Conversely, habitat was always a better predictor than temperature when describing species distributions at a coarse scale within Great Britain. Northerly species were most likely to occur in cool micro-sites within the study sites, whereas southerly species were most likely to occur in warm micro-sites. Effects of microclimate were not limited to species at the edges of their distribution, and fine-resolution temperature surfaces should therefore ideally be utilised when undertaking climate-envelope modelling.

KW - land-cover

KW - bioclimatic models

KW - range

KW - assemblages

KW - coleoptera

KW - heterogeneity

KW - conservation

KW - communities

KW - adaptation

KW - resolution

U2 - 10.1111/j.1600-0587.2011.07434.x

DO - 10.1111/j.1600-0587.2011.07434.x

M3 - Article

VL - 35

SP - 831

EP - 838

JO - Ecography

JF - Ecography

SN - 0906-7590

IS - 9

ER -