Using species combinations in indicator value analyses

Summary 1. Indicator species are often determined using an analysis of the relationship between the species occurrence or abundance values from a set of sites and the classification of the same sites into site groups (habitat types, community types, disturbance states, etc.). It may happen, however, that a particular site group has no indicator species even if its sites have a community composition that is clearly distinct from the sites of other site groups. This motivates an exploration of the indicator value of not only individual species but also species combinations. 2. Here, we present a novel statistical approach to determine indicators of site groups using species data. Unlike traditional indicator value analysis, we allow indicators to be species combinations in addition to single species. We require that all the species forming the combination must occur in the site to use the combination as an indicator. We present a simple algorithm that identifies the set of indicators (each one being either a single species or a species combination) that show high positive predictive value for the target site group. Moreover, we demonstrate the use of the percentage of sites of the site group where at least one of its valid indicators occurs to determine whether the group can be reliably predicted throughout its range. 3. Using a simulation study, we show that if two species are not strongly correlated and their frequency in the data set is larger than the frequency of sites belonging to the site group, the joint occurrence of the two species has higher positive predictive value for the site group than the two species taken independently. 4. We illustrate the proposed method by determining which combinations of vascular plants can be used as indicators for 29 shrubland and forest vegetation types of New Zealand. 5. The proposed methodology extends traditional indicator value analyses and will be useful to develop multispecies ecological or environmental indicators. Further, it will allow newly surveyed sites to be reliably assigned to previously defined vegetation types.

[1]  P. Murtaugh,et al.  The Statistical Evaluation of Ecological Indicators , 1996 .

[2]  Anne Chao,et al.  A novel statistical method for classifying habitat generalists and specialists. , 2011, Ecology.

[3]  R. Brooks,et al.  Scaling up the value of bioindicators. , 1998, Trends in ecology & evolution.

[4]  Xavier Font,et al.  Numerical reproduction of traditional classifications and automatic vegetation identification. , 2009 .

[5]  S. Miller,et al.  Discriminatory power of different arthropod data sets for the biological monitoring of anthropogenic disturbance in tropical forests , 2004, Biodiversity & Conservation.

[6]  M. McGeoch.  The selection, testing and application of terrestrial insects as bioindicators , 2007 .

[7]  P. Legendre,et al.  Associations between species and groups of sites: indices and statistical inference. , 2009, Ecology.

[8]  R. Freckleton,et al.  An objective, niche‐based approach to indicator species selection , 2012 .

[9]  J. Bakker.  Increasing the utility of Indicator Species Analysis , 2008 .

[10]  Gerald J. Niemi,et al.  Application of Ecological Indicators , 2004 .

[11]  M. McGeoch,et al.  The verification and application of bioindicators: a case study of dung beetles in a savanna ecosystem , 2002 .

[12]  Bahjat F. Qaqish,et al.  A family of multivariate binary distributions for simulating correlated binary variables with specified marginal means and correlations , 2003 .

[13]  Miquel De Cáceres,et al.  Updating vegetation classifications: an example with New Zealand's woody vegetation , 2013 .

[14]  P. Legendre.  Species associations: the Kendall coefficient of concordance revisited , 2005 .

[15]  M. O. Hill,et al.  TWINSPAN: a FORTRAN program of arranging multivariate data in an ordered two way table by classification of individual and attributes , 1979 .

[16]  Susan K. Wiser,et al.  New Zealand's forest and shrubland communities: a quantitative classification based on a nationally representative plot network. , 2011 .

[17]  Vincent Carignan,et al.  Selecting Indicator Species to Monitor Ecological Integrity: A Review , 2002, Environmental monitoring and assessment.

[18]  M. McGlone,et al.  The origin of the indigenous grasslands of southeastern South Island in relation to pre-human woody ecosystems. , 2001 .

[19]  Martin Kočí,et al.  Formalized reproduction of an expert-based phytosociological classification: A case study of subalpine tall-forb vegetation , 2003 .

[20]  P. Legendre,et al.  SPECIES ASSEMBLAGES AND INDICATOR SPECIES:THE NEED FOR A FLEXIBLE ASYMMETRICAL APPROACH , 1997 .

[21]  R. Swihart,et al.  Improving selection of indicator species when detection is imperfect , 2012 .

[22]  W. Willner,et al.  Effects of different fidelity measures and contexts on the determination of diagnostic species , 2009 .

[23]  Milan Chytrý,et al.  Determination of diagnostic species with statistical fidelity measures , 2002 .

[24]  Xavier Font,et al.  The management of vegetation classifications with fuzzy clustering , 2010 .

[25]  Stephan Hennekens,et al.  Assignment of relevés to pre-defined classes by supervised clustering of plant communities using a new composite index , 2008 .

[26]  János Podani,et al.  Detecting indicator species: Some extensions of the IndVal measure , 2010 .

[27]  Werner Ulrich,et al.  The empirical Bayes approach as a tool to identify non-random species associations , 2010, Oecologia.

[28]  P. Wardle.  Vegetation of New Zealand , 1991 .

[29]  A. Merenlender,et al.  Faunal indicator taxa selection for monitoring ecosystem health , 2000 .

[30]  Helge Bruelheide,et al.  A new measure of fidelity and its application to defining species groups , 2000 .

[31]  R. King,et al.  A new method for detecting and interpreting biodiversity and ecological community thresholds , 2010 .

[32]  S. Pignatti,et al.  Reflections on the phytosociological approach and the epistemological basis of vegetation science , 1980, Vegetatio.

[33]  Miquel De Cáceres,et al.  Improving indicator species analysis by combining groups of sites , 2010 .

[34]  Miquel De Cáceres,et al.  Towards consistency in vegetation classification , 2012 .