Shopping for Ecological Indices? On the Use of Incidence-Based Species Compositional Similarity Measures

Round 1

Reviewer 1 Report

Very interesting article and I highly recommend that the session of conclusion can be added.

Author Response

We greatly appreciate the comments of the reviewer. As suggested we included a conclusion subtitle to our last paragraph, which was actually a conclusion paragraph.

Reviewer 2 Report

    The paper compares 12 beta-diversity indices to better understand their behavior facing characteristics patterns of dissimilarity.
    Its purpose is clearly empirical and pragmatic: it does not try to derive the mathematical properties of the indices but rather relies on the computation of examples with easy-to-use R code.
    It is quite clear except for the definition of the indices and pleasant to read. I have a few suggestions to improve it.

    1. table S1 must be included in the main text.
    Without it, the paper is cryptic. The indices are defined line 104 by a reference to Supplementary Materials: a reader without an immediate access to the Internet can't understand anything.
    The abstract mentions several indices without defining them, with the same consequence. I suggest to remove their names, line 31, and write something like "several indices should be used cautiously" and explicitly define βsim and βsor.
    In the table, call βsim "Simpson", not only "Turnover component of βsor" for clarity.
    The formula of βjtu is poorly written: either a is at the denominator of the fraction (then, parentheses are missing) or it is added to the fraction: then write it before the fraction to avoid ambiguity, i.e. a + 2min(b,c)/2min(b,c).
    I think the reference of the paper that introduced each index should be in the table.

    2. Proportional diversity is not addressed.
    Jurasinski et al. (2009) oppose differentiation diversity (between two communities, e.g. Jaccard) and proportional diversity (the ratio between gamma and alpha diversity). Whittaker (1960) defines beta diversity first with Jaccard dissimilarity but then as gamma/alpha, so the statement of the first sentence of the paper (lines 43-47) is questionable. The literature about proportional diversity (e.g. Lande 1996, Jost 2007) also claims  Whittaker's heritage.
    I suggest to add a few sentences after the first paragraph, e.g. line 57, to explain that proportional diversity is not addressed here.
    It should be clarified somewhere that the additive partitioning of beta-diversity proposed by Baselga is for nestedness and turnover, without any relation to the additive partitioning of diversity (actually: of entropy) derived by Lande (1996). The best place to do that may be immediately after the previous addition.

    3. I think that Podani & Schmera (2011)'s partitioning, argued for by Podani & Schmera (2016) against Baselga's, should not be ignored because it has a strong background (see fig 1 in Podani & Schmera 2016). I acknowledge that the paper's aim is not to take part of the controversy but I suggest to add this one for a little more exhaustivity.

    References
    Jost, L. (2007). Partitioning diversity into independent alpha and beta components. Ecology, 88(10), 2427–2439. https://doi.org/10.1890/06-1736.1
    Jurasinski, G., Retzer, V., & Beierkuhnlein, C. (2009). Inventory, differentiation, and proportional diversity: A consistent terminology for quantifying species diversity. Oecologia, 159(1), 15–26. https://doi.org/10.1007/s00442-008-1190-z
    Lande, R. (1996). Statistics and partitioning of species diversity, and similarity among multiple communities. Oikos, 76(1), 5–13. https://doi.org/10.2307/354574
    Podani, J., & Schmera, D. (2011). A new conceptual and methodological framework for exploring and explaining pattern in presence—Absence data. Oikos, 120(11), 1625–1638. https://doi.org/10.1111/j.1600-0706.2011.19451.x
    Podani, J., & Schmera, D. (2016). Once again on the components of pairwise beta diversity. Ecological Informatics, 32, 63–68. https://doi.org/10.1016/j.ecoinf.2016.01.002

Author Response

We appreciate the positive and thoughtful feedback. We agree that table S1 should be in the text and thus moved it.

 

Regarding mentioning indices in the abstract, we proceeded as suggested.

 

We agree in that we did not consider proportional diversities, which we clarify in the first paragraph of the introduction. The new phrase now reads: “Notably, Whittaker further defined β-diversity as a quotient of diversities (i.e., γ/α, establishing the basis for its consideration as proportional diversity [9–11]. This particular topic is not addressed in this paper.”

 

We also acknowledge the type of additive partitioning that we refer to and clarify to avoid misunderstandings, as follows: “As such, this additive partitioning of β-diversity is not related with other proposed additive partitioning of diversity which follow a different rationale (i.e, β-diversity = γ/α) [10,18]”

 

We expanded on the reasons of Podani and Schmera’s criticism, as we fully agree that it is important to let the readers know what was shared by them (aside of the controversies), as follows: “Although there are elements of the proposed partitioning approach that have been criticized (e.g., inconsistency with the variation of species replacement and species loss, failure to accurately represent the species replacement and species loss processes, independence of richness difference; lack of connections to any other nestedness indices, overrepresentation of the replacement component due to the scaling difference; Almeida-Neto et al. [19]; Chen and Schmera [20]; Podani and Schmera [21,22]), it has recently gained relevance in the literature due to the information that it can provide on different ecological processes related to changes in species composition among communities, with topics ranging from genetic to biogeographical diversity (e.g., Diniz-Filho et al. [23]; Mouillot et al. [24]; Norhazrina et al. [25]; Ramachandran et al. [26]).”

Reviewer 3 Report

This is a valuable manuscript. However, they missed some important literature related with beta-diversity partitioning.

• The authors should also test the β-3 solution 2min(b,c)/(a+b+c) (see Cardoso et al. 2009; Carvalho et al. 2013).
• Critics to the performance of Baselga’s (2010) method should also be highlighted. Baselga’s (2010) method was found to be inconsistent with the variation of species replacement and species loss. Indeed, when tested against modelled gradients, the partitioning methods proposed by Baselga failed to accurately represent the species replacement and species loss processes that generate community variation.
• The authors should also mention the possibility of using the R-Package BAT (Cardoso et al., 2015).

 

 

 

-Cardoso, P., Borges, P.A.V. & Veech, J.A. (2009). Testing the performance of beta diversity measures based on incidence data: the robustness to undersampling. Diversity and Distributions, 15: 1081-1090.

-Cardoso, P., Rigal, F. & Carvalho, J.C. (2015) BAT - Biodiversity Assessment Tools, an R package for the measurement and estimation of alpha and beta taxon, phylogenetic and functional diversity. Methods in Ecology and Evolution. 6 (2) 232-236.

-Carvalho, J.C., Cardoso, P., Borges, P.A.V., Schmera, D. & Podani, J. (2013). Measuring fractions of beta diversity and their relationships to nestedness: a theoretical and empirical comparison of novel approaches. Oikos, 122: 825–834.

Author Response

We greatly appreciate the positive feedback of the reviewer. We agree in that there are variations to some of the tested indices. However, given the rationale of choosing them, we consider it is best to stick to the “standard” formulas.

We included a note acknowledging the suggested literature, together with the BAT package (although we must confess that it did not run on Linux, only on Windows), as follows: “In order to validate our interpretation of the NMDS grouping, we ran an analysis to fit the interpreted groupings (a vector) to the ordination using the ‘envfit’ function for the package ‘vegan’ (beta calculations can also be ran on BAT [39]).”

Reviewer 4 Report

The manuscript shows an interesting test for several β-diversity indices commonly used. It is not a novelty per se but it further studies the impacts of those indices’ properties on their results (and their similarities). In general, the manuscript is well written and organized (except comment below).

The most important issue with this manuscript is the restricted literature review both in time and scope. There are several new developments for evaluating species diversity (alpha, beta, gamma) in the last ten year or so that are completely absent here. All the profound analysis that considers Hill numbers (as an example) is ignored.

I strongly urge authors to include in their review the following papers (and many others from those authors and others) and adjust their study and results accordingly.

CHAO, A.; CHIU, C. Bridging the variance and diversity decomposition approaches to beta diversity via similarity and differentiation measures.

CHIU, C. H.; JOST, L.; CHAO, A. Phylogenetic beta diversity, similarity, and differentiation measures based on Hill numbers. Ecological Monographs, v. 84, n. 1, p. 21–44, 2014.

SÆTHER, B.; ENGEN, S.; GRØTAN, V. Species diversity and community similarity in fluctuating environments : parametric approaches using species abundance distributions. , p. 721–738, 2013.

Chao, A., Chiu, C.-H., Villéger, S., Sun, I.-F., Thorn, S., Lin, Y.-C., Chiang, J. M. and Sherwin, W. B. (2019). An attribute-diversity approach to functional diversity, functional beta diversity, and related (dis)similarity measures. Ecological Monographs, 89, e01343.

Chao, A. and Colwell, R. K. (2022) Biodiversity: concepts, dimensions, and measures. pp 25-46 in “The Ecological and Societal Consequences of Biodiversity Loss” (coordinated by Loreau, M., Hector, A., and Isbell, F.) John Wiley, New York.

Rother, D. C., Liboni, A. P., Magnago, L. F. S., Chao, A., Chazdon, R. L. and Rodrigues, R. R. (2019). Ecological restoration increases conservation of taxonomic and functional beta diversity of woody plants in a tropical fragmented landscape. Forest Ecology and Management, 451, 117538.

Chao, A. and Ricotta, C. (2019). Quantifying evenness and linking it to diversity, beta diversity, and similarity. Ecology, 100(12), e02852

Chao, A., Chiu, C.-H., Wu, S.-H., Huang, C. L. and Lin, Y.-C. (2019). Comparing two classes of species alpha diversities and their corresponding beta and (dis)similarity measures, with an application to the assessment of a reintroduction program of Formosan sika deer (Cervus nippon taiouanus). Methods in Ecology and Evolution,10, 1286-1297

 

 

L 60 – Authors cite β-diversity properties “symmetry, homogeneity, nestedness, and additivity”, but only nestedness is described. As all properties are tested against, they should be equally described.

 

 

Author Response

We appreciate the positive feedback of the reviewer. We agree fully in that the novelty of the study relies on the comparison among their results. We recognize that there are many other indices and approaches. Yet, with the aim of keeping the paper manageable, we selected the assessed indices based on their use and popularity in the literature, as stated in the manuscript: “The main goal of this study was to evaluate the performance of 12 incidence-based β-diversity indices under different scenarios (see Table 1 for their mathematic formula and interpretation): two classical and widely used measures (i.e., β_j, β_sor), four measures which have been demonstrated to be symmetric, homogeneous, and sensitive to nestedness by Koleff et al. [12] (i.e., β_sim, β_r, β_-2, β_-3), and six recently proposed measures related to the partitioning of β-diversity (i.e., β_sne, β_rich, β_jne, β_jtu, β_rich.s, β_-3.s; reviewed in Baselga and Leprieur [17]).”

Particularly, Hill numbers (with the exception of that equal to S – often referred to as 0q) are abundance-based and this paper focuses on incidence-based metrics.

We agree in that the definition of the cited properties of indices was lacking and included them in the revised version, as follows: “ In 2003, Koleff et al. [12] assessed and synthesized the performance of the large array of β-diversity measures available at that time. Specifically, the authors undertook a comparative analysis of 24 β-diversity measures focusing on four mathematical properties: symmetry (for two quadrats x and y, β (x,y) must be equal to β (y,x)), homogeneity (if the components of the measure are multiplied by the same constant, this should not affect the resulting β-diversity value), nestedness (all the species occurring in the focal quadrat also occur in the neighboring quadrat), and additivity (for three quadrats in the spatial sequence x, y, z, the sum of the values of beta diversity between x and y and between y and z equals the value of beta diversity between x and z).”

We have cited some of the proposed papers. We appreciate the list.

Round 2

Reviewer 2 Report

The authors addressed my remarks. I suggest accepting the paper for publication.

Reviewer 3 Report

Many thanks for considering my suggestions. Now it can be accepted 

Reviewer 4 Report

I congratulate the authors for the revised version of the manuscript. It is an interesting paper.