A plea for using qualitative aspects in the interpretation of ecological field data as revealed by the proof of carabid beetle assemblages of a pristine salt marsh

Abstract. The evaluation of ecological field data can be done by an increasing number of quantitative methods. The application of these methods often is often blind against two kinds of problems: (i) the data often do not meet the requirements of a method, e.g., as an ultra-metric structure of the data in the case of hierarchical cluster analysis. In such cases, the result will be misleading because the presentation of results is ultra-metric independent on the structure of the data. (ii) Most of the animals are able to move actively or may drift passively by wind, etc. Therefore, species occurring by accident like vagrants have to be eliminated from the assemblage of animals at a particular site before a quantitative method is applied. In addition, the result of a quantitative analysis has to be checked for its ecological plausibility. This is a qualitative step, which can only be done by taking into account the known data on biology and ecology of the species. Some pitfalls of an exclusive application of quantitative methods will be demonstrated in this paper using a data set of salt marsh Carabidae.