Assessing the conservation status of marine habitats: thoughts from a sandflat on the Isles of Scilly

• Biotope concept only useful at physical habitat level
• Setting fixed local limits on natural variability is almost always impractical.
• Temporal or spatial baselines cannot be established.
• AMBI and taxonomic distinctness are useful, not requiring such baselines.
• Biomass data offers more possibilities for assessment.

Statutory monitoring of the fauna of the ‘mudflats and sandflats not covered by seawater at low tide’ biotope complex on St Martin's Flats, a part of the Isles of Scilly Complex Special Area of Conservation, was undertaken in 2000, 2004 and 2009. The targets set by Natural England for “characteristic biotopes” were that “composite species, abundance and diversity should not deviate significantly from an established baseline, subject to natural change”. The three specified biotopes could not be distinguished, and instead three assemblages were subjectively defined based on sediment surface features. There were statistically significant natural changes in diversity and species composition between years, especially in the association initially characterised by the razor-clam Ensis, and possible reasons for this are discussed. It is suggested that setting fixed local limits on natural variability is almost always impractical. Two possible approaches to distinguishing between natural and anthropogenic changes are suggested; a change in ecological condition as indicated by AMBI scores, and a significant change in average taxonomic distinctness (Δ+) compared with expectation. The determination of species biomasses as well as abundances might also open more possibilities for assessment. The practice of setting objectives for a marine Special Area of Conservation (SAC) feature that include the range and number of biotopes cannot be supported, in view of the difficulty in ascribing assemblages to recognised biotopes. A more realistic definition of species assemblages might best be gained from examination of the species that consistently make a substantial contribution to the Bray–Curtis similarity among samples collected from specific sites.