Skip to content

taxonomic diversity and distinctness - taxondive() in vegan package

6 messages · Jari Oksanen, Gian Maria Niccolò Benucci

Original
Jari Oksanen
#
On 16/02/2012, at 18:39 PM, Gian Maria Niccol? Benucci wrote:

            

      
      
      
    

        
Dear Giain Maria Niccol? Benucci,

The ?taxondive help page reads:

     Clarke & Warwick (1998, 2001) suggested several alternative
     indices of taxonomic diversity or distinctness. Two basic indices
     are called taxonomic diversity (Delta) and distinctness (Delta^*):

             Delta = (sum sum_{i<j} omega_{ij} x_i x_j)/(n (n-1) / 2)       
       Delta^* = (sum sum_{i<j} omega_{ij} x_i x_j)/(sum sum_{i<j} x_i x_j) 
      
     The equations give the index value for a single site, and
     summation goes over species i and j. Here omega are taxonomic
     distances among taxa, and x are species abundances, and n is the
     total abundance for a site.  With presence/absence data both
     indices reduce to the same index Delta^+

The intention of this passage was to give equations for Delta and Delta^*, and explanation of Delta^+ (and the pdf version has nicer formatting of equations). If this is not enough to explain the methodology of calculation, please consult the original publications referred to in the help page and the R code of taxondive().

            

      
      
      
    

        
It is a long time since I read these papers, but I think C&W suggested plotting Delta^+ values (not Delta^*), and that plot is produced by plot() command for taxondive() results in vegan, like displayed in the examples of the help page of taxondive. However, the labels cannot be displayed with the standard plot() command: you have to do that yourself. Look at the model in vegan:::plot.taxondive to see how simple it is.

Cheers, Jari Oksanen

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        
      

    

    
4 days later

    

      
      

        
      

    

    
1 day later

    

      
      

        


  

        

      Jari Oksanen
    

    

      
      #
    

  


  

      
On 21/02/2012, at 13:38 PM, Gian Maria Niccol? Benucci wrote:

        

            

      
      
      
    

        
When I implemented step lengths, I followed Clarke & Warwick (1999): probably they explain the scaling, and I would read that reference if I needed to explain how it works.

The docs say that the step length is relative to proportional decrease in the number of classes. Does this apply to your genus and class ranks?

Cheers, Jari Oksanen