Comments on separating components of beta diversity

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Beta diversity, species replacement and nestedness are often examined through pairwise comparisons of sites based on presence-absence data, and the relative importance of these ecological phenomena is evaluated by operations with dissimilarity coefficients. An example is the nestedness resultant dissimilarity (NRD) procedure recently proposed by Baselga (2010, Global Ecology andBiogeography 19: 134-143) to disentangle the nestedness fraction of beta diversity from species replacement. In our view, the component terms in this measure are not scaled uniformly and the nestedness fraction cannot be quantified properly without giving clear definitions for its measurement. We suggest to distinguish among three additive fractions of the species set of two sites: number of species shared (overlap), species replacement (=spatial turnover) and richness difference. Then, absolute beta diversity is obtained as a composite of the second two fractions (known as βWB), while nestedness is derived from the first and the third. To express beta diversity and nestedness in a relativized form, the respective sums are divided by the total number of species. These allow defining a new index to measure the fraction of beta diversity which is shared by nestedness as well, and is calculated as relativized richness difference with the condition that the two sites being compared have at least one species in common. It is called diversity-nestedness intersection coefficient (F). Baselga's nestedness resultant dissimilarity and the diversity-nestedness intersection coefficient are compared graphically using artificial and actual examples. These functions follow a mathematical relationship for perfectly nested data, otherwise their results are divergent. Discrepancy increases when beta diversity is large, especially if richness differences override species replacement effects in shaping presence-absence data structures. An advantage of F is its compatibility with a general theoretical and methodological framework for revealing pattern in presence-absence data matrices.

Original languageEnglish
Pages (from-to)153-160
Number of pages8
JournalCommunity Ecology
Volume12
Issue number2
DOIs
Publication statusPublished - Dec 1 2011

Fingerprint

nestedness
replacement
ecological phenomena
interspecific variation
species diversity
turnover
ecology

Keywords

  • Beta diversity
  • Nestedness
  • Presence-absence data
  • Richness difference
  • Species replacement

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Ecology

Cite this

Comments on separating components of beta diversity. / Schmera, D.; Podaní, J.

In: Community Ecology, Vol. 12, No. 2, 01.12.2011, p. 153-160.

Research output: Contribution to journalArticle

@article{ceedb59c105d4953adfbce32c7ebab97,
title = "Comments on separating components of beta diversity",
abstract = "Beta diversity, species replacement and nestedness are often examined through pairwise comparisons of sites based on presence-absence data, and the relative importance of these ecological phenomena is evaluated by operations with dissimilarity coefficients. An example is the nestedness resultant dissimilarity (NRD) procedure recently proposed by Baselga (2010, Global Ecology andBiogeography 19: 134-143) to disentangle the nestedness fraction of beta diversity from species replacement. In our view, the component terms in this measure are not scaled uniformly and the nestedness fraction cannot be quantified properly without giving clear definitions for its measurement. We suggest to distinguish among three additive fractions of the species set of two sites: number of species shared (overlap), species replacement (=spatial turnover) and richness difference. Then, absolute beta diversity is obtained as a composite of the second two fractions (known as βWB), while nestedness is derived from the first and the third. To express beta diversity and nestedness in a relativized form, the respective sums are divided by the total number of species. These allow defining a new index to measure the fraction of beta diversity which is shared by nestedness as well, and is calculated as relativized richness difference with the condition that the two sites being compared have at least one species in common. It is called diversity-nestedness intersection coefficient (F). Baselga's nestedness resultant dissimilarity and the diversity-nestedness intersection coefficient are compared graphically using artificial and actual examples. These functions follow a mathematical relationship for perfectly nested data, otherwise their results are divergent. Discrepancy increases when beta diversity is large, especially if richness differences override species replacement effects in shaping presence-absence data structures. An advantage of F is its compatibility with a general theoretical and methodological framework for revealing pattern in presence-absence data matrices.",
keywords = "Beta diversity, Nestedness, Presence-absence data, Richness difference, Species replacement",
author = "D. Schmera and J. Podan{\'i}",
year = "2011",
month = "12",
day = "1",
doi = "10.1556/ComEc.12.2011.2.2",
language = "English",
volume = "12",
pages = "153--160",
journal = "Community Ecology",
issn = "1585-8553",
publisher = "Akademiai Kiado",
number = "2",

}

TY - JOUR

T1 - Comments on separating components of beta diversity

AU - Schmera, D.

AU - Podaní, J.

PY - 2011/12/1

Y1 - 2011/12/1

N2 - Beta diversity, species replacement and nestedness are often examined through pairwise comparisons of sites based on presence-absence data, and the relative importance of these ecological phenomena is evaluated by operations with dissimilarity coefficients. An example is the nestedness resultant dissimilarity (NRD) procedure recently proposed by Baselga (2010, Global Ecology andBiogeography 19: 134-143) to disentangle the nestedness fraction of beta diversity from species replacement. In our view, the component terms in this measure are not scaled uniformly and the nestedness fraction cannot be quantified properly without giving clear definitions for its measurement. We suggest to distinguish among three additive fractions of the species set of two sites: number of species shared (overlap), species replacement (=spatial turnover) and richness difference. Then, absolute beta diversity is obtained as a composite of the second two fractions (known as βWB), while nestedness is derived from the first and the third. To express beta diversity and nestedness in a relativized form, the respective sums are divided by the total number of species. These allow defining a new index to measure the fraction of beta diversity which is shared by nestedness as well, and is calculated as relativized richness difference with the condition that the two sites being compared have at least one species in common. It is called diversity-nestedness intersection coefficient (F). Baselga's nestedness resultant dissimilarity and the diversity-nestedness intersection coefficient are compared graphically using artificial and actual examples. These functions follow a mathematical relationship for perfectly nested data, otherwise their results are divergent. Discrepancy increases when beta diversity is large, especially if richness differences override species replacement effects in shaping presence-absence data structures. An advantage of F is its compatibility with a general theoretical and methodological framework for revealing pattern in presence-absence data matrices.

AB - Beta diversity, species replacement and nestedness are often examined through pairwise comparisons of sites based on presence-absence data, and the relative importance of these ecological phenomena is evaluated by operations with dissimilarity coefficients. An example is the nestedness resultant dissimilarity (NRD) procedure recently proposed by Baselga (2010, Global Ecology andBiogeography 19: 134-143) to disentangle the nestedness fraction of beta diversity from species replacement. In our view, the component terms in this measure are not scaled uniformly and the nestedness fraction cannot be quantified properly without giving clear definitions for its measurement. We suggest to distinguish among three additive fractions of the species set of two sites: number of species shared (overlap), species replacement (=spatial turnover) and richness difference. Then, absolute beta diversity is obtained as a composite of the second two fractions (known as βWB), while nestedness is derived from the first and the third. To express beta diversity and nestedness in a relativized form, the respective sums are divided by the total number of species. These allow defining a new index to measure the fraction of beta diversity which is shared by nestedness as well, and is calculated as relativized richness difference with the condition that the two sites being compared have at least one species in common. It is called diversity-nestedness intersection coefficient (F). Baselga's nestedness resultant dissimilarity and the diversity-nestedness intersection coefficient are compared graphically using artificial and actual examples. These functions follow a mathematical relationship for perfectly nested data, otherwise their results are divergent. Discrepancy increases when beta diversity is large, especially if richness differences override species replacement effects in shaping presence-absence data structures. An advantage of F is its compatibility with a general theoretical and methodological framework for revealing pattern in presence-absence data matrices.

KW - Beta diversity

KW - Nestedness

KW - Presence-absence data

KW - Richness difference

KW - Species replacement

UR - http://www.scopus.com/inward/record.url?scp=81755184417&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=81755184417&partnerID=8YFLogxK

U2 - 10.1556/ComEc.12.2011.2.2

DO - 10.1556/ComEc.12.2011.2.2

M3 - Article

AN - SCOPUS:81755184417

VL - 12

SP - 153

EP - 160

JO - Community Ecology

JF - Community Ecology

SN - 1585-8553

IS - 2

ER -