Journal of Applied Ecology 2001 38, 689 – 698 © 2001 British Ecological Society Blackwell Science, Ltd The influence of management regime and altitude on the population structure of Succisa pratensis: implications for vegetation monitoring CHRISTOPH BÜHLER and BERNHARD SCHMID Institut für Umweltwissenschaften, University of Zürich, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland Summary 1. Environmental change and land use may alter the vegetation typical of wetland hab- itats. However, commonly used techniques for vegetation monitoring that are based on species composition are often not suitable to detect gradual changes in the structure of vegetation at an early stage. In this study we tested if this methodical deficiency could be overcome by observation of the stage structure of a perennial target species. 2. We studied the density and frequency distribution of four phenological stages of the perennial plant Succisa pratensis in 24 calcareous fens in eastern Switzerland. These fens differed in management type (mowing, cattle-grazing) and altitude (low, medium, high). Among grazed fens, the intensity of management was quantified by direct observa- tion. Species composition and canopy structure of the plant community surrounding S. pratensis were also measured. 3. High altitude had a positive effect on the density of adult plants of S. pratensis only in mown fens, whereas in grazed fens adult density was highest at medium altitude. 4. Local densities of seedlings and vegetative adults within a fen were higher in mown than in grazed fens, and lower in fens of the lowest altitude level than of the two higher ones. However, there were no differences in the relative proportions of all four pheno- logical stages among main factors except that significantly fewer seedlings occurred in fens of the lowest altitude level. 5. It is the intensity rather than the mere type of management regime that is crucial for shaping populations of S. pratensis: for grazed fens, there is a negative relationship between intensity of grazing and density of adult plants of S. pratensis. Moreover, the number, the relative proportion of seedlings and the seed-set are all negatively cor- related with grazing intensity. 6. The species composition of the plant community, but not its physical structure, was significantly related to the relative proportion of seedlings of S. pratensis, which supports the indicator qualities of the target species. 7. Monitoring the population structure of one or several target species provides important indicator information about the stability of a whole plant community. For the target-species approach we propose to use characteristic but frequently and steadily occurring species instead of rare or endangered ones. Key-words: calcareous fens, indicator species, management strategies, phenological stage classes, wetland conservation. Journal of Applied Ecology (2001) 38, 689 – 698 Introduction Most of the methods designed to monitor vegetation change in species-rich grassland rely on the periodical inspection of the species composition within a certain number of permanent plots at a site. Although numer- ous efforts have been made to optimize this method (number, size, shape, distribution of plots, frequency of census; Greig-Smith 1983; Goldsmith 1991; Weber et al. 1995; Kammer 1998), this ‘community’ approach gen- erally suffers from the problem that gradual changes in species composition may be recognized too late, i.e. Correspondence: Dr Bernhard Schmid, Institut für Umwelt- wissenschaften, Universität Zürich, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland (fax 0041 16355711; e-mail bschmid@uwinst.unizh.ch).