Back to monitoring outcomes

Species diversity (alpha, beta, gamma)

  • Biodiversity

  • Compostion

Summary

Species diversity captures the richness and relative abundance of species (Magurran 2021). Species diversity metrics can be applied at different scales. Alpha diversity is the species diversity within a focal site and applies at smaller (community) scales (Magurran et al. 2021, Chiarucci et al. 2011). Beta diversity captures differences in species composition among sites, and is usually applied at larger scales e.g. to capture differences in species composition among habitat types within an area (Chiarucci et al. 2011). Gamma diversity captures overall species diversity at landscape scales (Anderson et al. 2018).

The species groups monitored provide information about different aspects of an ecosystem. Using 2-3 species groups dramatically increases the overall coverage of collective species responses (Larrieu et al. 2018). A synthesis of many biodiversity studies found that birds and plants were the best collective surrogate for unsurveyed species (Westgate et al. 2017). The species group chosen will provide information on different aspects of the overall ecological community. Plants are the most commonly assessed group and generally have higher correlation with other species groups (Burrascano et al. 2018). Butterflies respond to habitat conditions and the presence of their food plants in the area, beetles to habitat conditions and small-scale habitat heterogeneity (Burrascano et al. 2018, Ferris and Humphrey 1999). Birds have been linked to vegetation structure in forest systems and habitat diversity at larger scales (Ferris and Humphrey 1999).

Species diversity monitoring at larger scales can return higher species richness as more environmental variation and greater numbers of rare species are captured across larger areas (Magurran 2021). However, thorough monitoring at large scales can be challenging and there is a chance that rare species won’t be detected.

Methodology summary

Step 1 – decide on species groups to monitor
Using >2 species groups captures overall biodiversity more accurately. Species groups representing different parts of the ecosystem should be selected and the species listed below have monitoring protocols available to assess them:

  • Plants – have the highest number of correlations with other species groups
  • Carabid beetles – predatory so will reflect changes in their prey, respond to habitat changes differently to butterflies/moths
  • Spiders – generalist predators providing additional insights to carabids
  • Butterflies – respond to changes in vegetation abundance and quality, easy to recognise
  • Moths – good national data for comparison, reflect environmental change
  • Birds – reflect habitat structure and larger-scale habitat diversity, easy to recognise
  • Bats – sensitive to habitat change, roost site availability and invertebrate food sources
  • Frogs – ubiquitous predatory amphibian, feeding on invertebrates
  • Wild herbivores (Rabbits and Deer) – drivers of pant composition and structure, dropping counts to estimate relative abundance
  • Spittle bugs – xylem feeding invertebrates that are easy to spot and ecologically well-understood in the UK
  • Crane flies – soil-dwelling and important food source for other groups

Step 2 – collect data

  • Plants – National Plant Monitoring Scheme – 5 x 5 m plots (or 10 x 10 m plots in woodland) – 3 plots per habitat type within 1 km square
  • Carabid beetles – UK Environmental Change Network (ECN) ground predators protocol – 3 transects (10 traps per transect 10 m spaced), representing different vegetation types within central 9 ha representing site
  • Spiders – UK ECN spiders protocol  – 3 transects (10 traps per 100 m transect), representing different vegetation types within central 9 ha representing site
  • Butterflies – UK Butterfly Monitoring Scheme  – 1-2 km transect representing habitats within 1 km square (subsections of transects)
  • Moths – UK ECN moths protocol  – 1 light trap in centre of 1 ha square
  • Birds – UK Breeding Bird Survey  – 2 transects per 1km square per site, one transect per half, habitat features recorded
  • Bats – UK ECN bats protocol  – 2 transects per 1km square per site, one transect per half, habitat features recorded
  • Frogs – UK ECN frogs protocol  – shallow ponds/ditches within site
  • Wild herbivores (Rabbits and Deer) – UK ECN rabbits and deer protocol – 1-2 km transect representing habitats within site (subsections of transects)
  • Spittle bugs – UK ECN spittle bugs protocol  – 20 quadrats 0.25 m2 randomly in the central 1 ha of site
  • Crane flies – UK ECN crane flies protocol  – 50 x 40 m central area divided into 20 subplots, 10 cm diameter and 10 cm depth cores taken randomly from within subplots

Sampling layout
For sites <0.1 km2 (<10 ha)

  • Plants – 3 5 x 5 m plots per habitat type (or 10 x 10 m plots if in woodland)
  • Carabid beetles and spiders – 1-3 100 m transects (10 traps per transect) per habitat depending on size of site and habitat conformation
  • Moths – trap in centre of site (if >1 ha), traps attract moths from 30 m radius, so if there are large enough areas of distinct habitat consider multiple traps located in these
  • Spittle bugs – 20 0.25 m2 quadrats
  • Crane flies – 20 subplots in central 50×40 m area
  • Butterflies, bats, birds and wild herbivores – a modified transect approach might be possible – linking presence to smaller scale habitat heterogeneity will be harder

For sites 0.1-1 km2 (10-100 ha)

  • Plants – 3 5 x 5 m plots per habitat type (or 10 x 10 m plots if in woodland)
  • Carabid beetles and spiders – 3 transects per habitat type (traps spaced every 10 m, aim for 10 traps = 100 m transects)
  • Moths – traps attract moths from 30 m radius, so if there are large enough areas of distinct habitat consider multiple traps located in these
  • Spittle bugs – 20 0.25 m2 quadrats
  • Crane flies – 20 subplots in central 50×40 m area
  • Bats and birds – in sites 1 km2 or close to 1 km2 2 c.1 km transects, in smaller sites a modified transect approach might be possible
  • Butterflies and Wild Herbivores – in sites 1 km2 or close to 1 km2 transects 1-2 km transect, in smaller sites a modified transect approach might be possible

For sites >1 km2 (>100 ha)

  • Plants – for each 1 km square 3 5 x 5 m plots per habitat type (or 10 x 10 m plots if in woodland)
  • Carabid beetles and spiders – per 1 km square 3 transects per habitat type (traps spaced every 10 m, aim for 10 traps = 100 m transects)
  • Moths – traps attract moths from 30 m radius, so if there are large enough areas of distinct habitat consider multiple traps co-located with plant plots
  • Spittle bugs – 20 0.25 m2 quadrats per 1 km square
  • Crane flies – 20 subplots in central 50 x 40 m area per 1 km square
  • Bats and birds – for each 1 km square 2 transects (where large areas of one habitat are present sampling might be possible focussed on habitat type)
  • Butterflies and Wild Herbivores – for each 1 km square 1-2 km transect (where large areas of one habitat are present sampling might be possible focussed on habitat type might be possible)

Step 3 – calculate diversity indices
Point diversity

  • Calculated at the level of plot or trap
  • Simpson’s diversity index (D) is recommended (Magurran 2004)
  • Simpson’s diversity index can be calculated using the diversity function in the vegan package in R
  • Simpson’s diversity index is often presented as the complement (1−D)
  • As the complement of Simpson’s increases so does diversity
  • Can assess variation in point diversity by habitat type if multiple habitats present at site

Alpha diversity

  • In sites <10 ha, calculated at the level of habitat for plants, carabid beetles, spiders, moths, crane flies if possible, calculated at the site level for bats, birds and butterflies
  • In sites 10-100 ha, calculated at the level of habitat for plants, carabid beetles, spiders, moths, crane flies, calculated at the site level for bats, birds and butterflies
  • In sites >100 ha, calculated per habitat for all species if possible
  • Simpson’s diversity index (D) is recommended (Magurran 2004)
  • Simpson’s diversity index can be calculated using the diversity function in the vegan package in R
  • Simpson’s diversity index is often presented as the complement (1−D)
  • As the complement of Simpson’s increases so does diversity

Beta diversity

  • Between habitat diversity – likely only to be relevant for sites >10 ha, with multiple large areas of habitat
  • A simple metric of beta-diversity is Whittaker’s beta diversity, which can be calculated using the betadiver function in the vegan package in R
  • Whittaker’s beta diversity = (total site species richness)/(mean habitat species richness)
  • See also Similarity for measures of differences in the assemblages of species between habitats

Gamma diversity

  • Total diversity at the site or landscape scale, calculated across all habitat types
  • Often calculated as species richness

Other metrics that can be derived

  • Species richness
  • Species evenness metrics

User guide for vegan available.
Information on the vegan package and examples of visualisation.

See Mammal biomass, Invertebrate biomass, Vegetation biomass for more information on considerations relating to these groups.

Metric threshold or direction of change

Generally increases in species diversity could be seen as positive, however the species composition underlying change in these metrics should be considered and increases in target species (e.g. species characteristic of target habitats) are important.

Where data has been collected following standardised UK-wide methods (e.g. UK Breeding Bird Survey), species-level trends over time can be compared to national trends.

Increases due to the presence of invasive or non-native species may not be desirable.

Technological innovations

  • Environmental DNA (eDNA) can facilitate accurate and efficient species surveys that could replace traditional surveys. Sensitivity varies depending on the focal taxon. (Fediajevaite et al. 2021, Deiner et al. 2017).
  • For bulk invertebrate samples metabarcoding can be used to identify all species present. The detection of all species in a sample can be improved by sorting taxa by size (Elbrecht et al. 2017).
  Close

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

Scale

  • Community
  • Landscape

Cost

  • Medium

Tier

  • Tier 1

Technical expertise

  • High

Standardised methodology

  • Partial