Environmentally associated spatial changes of a macrozoobenthic community in the Saemangeum tidal flat, Korea

Estuarine tidal flats are both ecologically and economically important, hence developing methods to reliably measure ecosystem health is essential. Because benthic fauna play a central role in the food web of tidal flats, in this study we set out to quantitatively describe the intertidal zonation of macro-invertebrates and their associations with specific environmental parameters along three transects in the Saemangeum tidal flat, Korea. The abundance and biomass of intertidal fauna with respect to five environmental parameters (i.e., shore level, mud content, coarse sand content, water content, and organic content) were measured, to identify environmental factors that influence macrofaunal distribution in intertidal soft bottom habitats. A total of 75 species were identified, with dominant species showing distinct zones of distribution along all transects. The number of species recorded in each transect was found to be dependent on sediment characteristics and salinity. Cluster analysis classified the entire study area into three faunal assemblages (i.e., location groups), which were delineated by characteristic species, including (A) 'Periserrula-Macrophthalmus' (B) 'Umbonium-Meretrix'. and (C) 'Prionospio-Potamocorbula'. Four environmental variables (i.e., shore level, water content, mud content, and organic content) appeared to determine factors that distinguished the three faunal assemblages, based on the discriminant analysis. The faunal assemblage types of the sampled locations were accurately predicted from environmental variables in two discriminant functions, with a prediction accuracy of 98%. It should be noted that the zonation of benthos in the lower section (C) of Sandong had been affected by the construction of a nearby dike, while this parameter had remained essentially unchanged at the other two location groups (A-B). Overall, the zonation of benthos from the Saemangeum tidal flat was explained adequately by the measured environmental variables, implying that faunal assemblages are closely associated with certain combinations of abiotic factors. The identification of such reliable associations may facilitate the development of statistical models to predict faunal distributions based on environmental variables at both local and regional scales. The entire study area was embanked in 2006 (one year after this study), and an integrated plan was set into force to develop claimed land into industrial, residential and agricultural districts, which also included a partial restoration program of the tidal flats located near to the study area. (C) 2011 Elsevier B.V. All rights reserved.