Impacts of data quality on the setting of conservation planning targets using the species–area relationship
Correspondence: Kristian Metcalfe Durrell Institute of Conservation and Ecology, University of Kent, Canterbury, Kent CT2 7NR, UK.
Robert J. Smith, Durrell Institute of Conservation and Ecology, University of Kent, Canterbury, Kent CT2 7NR, UK.
The species–area relationship (SAR) is increasingly being used to set conservation targets for habitat types when designing protected area networks. This approach is transparent and scientifically defensible, but there has been little research on how it is affected by data quality and quantity.
We used a macrobenthic dataset containing 1314 sampling points and assigned each point to its associated habitat type. We then used the SAR-based approach and tested whether this was influenced by changes in (i) the number of sampling points used to generate estimates of total species richness for each habitat type; (ii) the nonparametric estimator used to calculate species richness; and (iii) the level of habitat classification employed. We then compared our results with targets from a similar national-level study that is currently being used to identify Marine Conservation Zones in the UK.
We found that targets were affected by all of the tested factors. Sample size had the greatest impact, with specific habitat targets increasing by up to 45% when sample size increased from 50 to 300. We also found that results based on the Bootstrap estimator of species richness, which is the most widely used for setting targets, were more influenced by sample size than the other tested estimators. Finally, we found that targets were higher when using broader habitat classification levels or a larger study region. However, this could also be a sample size effect because these larger habitat areas generally contained more sampling points.
Habitat targets based on the SAR can be strongly influenced by sample size, choice of richness estimator and the level of habitat classification. Whilst setting habitat targets using best-available data should play a key role in conservation planning, further research is needed to develop methods that better account for sampling effort.