Higher taxa are effective surrogates for species in the selection of conservation reserves in estuaries

1. The lack of information about marine biodiversity is problematic for the selection of conservation reserves that aim to protect representative samples of biodiversity. A number of surrogate measures for biodiversity have been suggested as a potential solution to this problem. 2. The present study tested the effectiveness of using higher taxa of macroinvertebrates as a surrogate for species-level identification to depict spatial variation in species richness and assemblage variation and to select conservation reserves in one estuary in south-east Australia. 3. Spatial patterns of richness and assemblage variation for species were significantly correlated with patterns defined from genera, families, orders, classes, and phyla with a decline in the magnitude of correlation coefficients from finer to coarser resolutions. A network of reserves selected to include representatives of all phyla, classes, orders, families and genera coincidentally included 54%, 61.7%, 75%, 92.6%, 98.8% species in 8.3%, 13.9%, 17.7%, 44.4% and 58.3% of grid cells, respectively. However, only reserves selected for genera, families and orders performed significantly better than random selection. 4. Percentage of species represented by orders, families and genera in a realistic level of available grid cells for conservation (i.e. 13.9%) were very close ranging between 70 and 73.5%. A factor diminishing the performance of order as surrogate for species richness was related to the difficulty of identifying many macroinvertebrates to the order level. Therefore, it is concluded that genus- and family-level identification is an effective surrogate for species-level identification for conservation planning in estuaries. © 2008 John Wiley & Sons, Ltd