MEPS 162:71-78 (1998)  -  doi:10.3354/meps162071

Relationships between spatial patterns of benthic assemblages in a mangrove forest using different levels of taxonomic resolution

M. G. Chapman*

Centre for Research on Ecological Impacts of Coastal Cities, Marine Ecology Laboratories A11, University of Sydney, New South Wales 2006, Australia

The use of multivariate techniques to quantify spatial and temporal patterns in assemblages requires decisions about the resolution to which taxa are identified. Similar spatial patterns in assemblages are often found whether coarse or fine levels of taxonomic resolution are used. Whether coarser resolution retains patterns of relative difference in assemblages between habitats at a hierarchy of spatial scales has not been widely examined. Where organisms are patchy at many spatial scales, numerous replicates are needed to represent assemblages adequately. Identifying taxa into broad groups is generally quicker and cheaper than identification to species, so coarser resolution may release limited resources for the intensive sampling necessary for populations of highly dispersed and patchy organisms. In this study, intertidal epifaunal assemblages in urban mangrove forests were examined at scales of metres, tens of metres and hundreds of metres using 2 different levels of taxonomic resolution: 93 taxa and 9 broad groups. Bray-Curtis measures of dissimilarity were also compared between the 2 levels of taxonomic resolution to test the hypothesis that quantitative differences in the assemblages at a hierarchy of spatial scales were retained with different levels of taxonomic resolution. Assemblages varied at each spatial scale. Similar patterns were obtained at each level of taxonomic resolution because the Bray-Curtis measures of dissimilarity were strongly correlated between the 2 levels. More variable sites, as identified using one level of taxonomic resolution, were also more variable using a different level of taxonomic resolution. Also, sites that were similar to each other when 93 taxa were used remained similar when the taxa were grouped into 9 groups. Transforming the data to the fourth root retained patterns of differences between sites, but altered patterns of variability within sites. The relative value of putting limited resources into increasing the size of the sample as opposed to increased taxonomic resolution in such habitats is discussed.


Spatial pattern · Variance · Multivariate analyses · Benthos · Mangrove


Full text in pdf format