Back to monitoring outcomes

Similarity

  • Biodiversity

  • Compostion

Summary

Similarity is a measure of species composition compared to a reference community. It is important to measure composition as well as diversity, because diversity metrics based on species richness and abundance can remain constant as species assemblage changes (Santini et al. 2017, Magurran et al. 2018, Magurran 2021).

Similarity can be used to capture temporal turnover and, in projects with multiple habitats, spatial turnover (beta-diversity) within a project site (Santini et al. 2017, Schindler et al. 2008). Spatial turnover can be linked to topographical and habitat metrics collected simultaneously. With small sample sizes there is a risk that some similarity metrics (e.g. based on Bray-Curtis similarity) may underestimate similarity (Hardersen and La Porta 2023).

Methodology summary

Similarity metrics require data on species richness and abundance which can be obtained using methods for diversity metrics (see Species diversity for methodologies). They are typically calculated for each species group separately (e.g. for birds or plants).

The Bray-Curtis index assesses the number of shared species between two communities and their relative abundances, and can be calculated using the vegdist function in the vegan package in R. The Bray-Curtis index ranges from 0 (no species in common) to 1 (two communities identical).

Ordination methods can be used to visualise species assemblages in different communities. Ordination methods summarise information on species identity and relative abundance into a single value that captures the overall community composition of a site. A set of sites can be plotted on a graph, with sites that are more similar to each other plotted more closely together. A commonly used ordination method is non-metric multidimensional scaling (NMDS), which is based on Bray-Curtis similarity indices (monoMDS function in vegan). With NMDS, community composition in different sites can be assessed and analysis of similarities (anosim function in vegan) is used to test for differences using the vegan package in R.

User guide for vegan.
Information on the vegan package and examples of visualisation, including NMDS.

Metric threshold or direction of change

Changes in species assemblage can be tracked over time, with project specific targets based on assemblages associated with target habitats. Comparisons can also be made to nearby sites with the target habitat type.

  Close

  • Agricultural
  • Forest
  • Grassland
  • Heathland
  • Other
  • Peatland
  • Saltmarsh
  • Wetland

Scale

  • Community

Cost

  • Medium

Tier

  • Tier 1

Technical expertise

  • High

Standardised methodology

  • Partial