Heracleum mantegazzianum in Secondary Successional Seres

A gradient analysis (correspondence analysis) of our vegetation relevés with H. mantegazzianum revealed a main sequence of vegetation types from managed grassland to ruderal grasslands to tall-herb communities and, finally, to woodlands. Subordinately, a parallel sequence from severely disturbed sites to tall-herb stands to woodlands was found (Fig. 8.4; for details see Thiele and Otte, 2006b). These two parallel sequences represented the main gradient

(first axis of the correspondence analysis) in the vegetation data. Towards the upper end of this gradient the age of abandoned or disturbed sites and the proportion of C-strategy (Grime et al., 1988; Grime, 2001) among resident plant species increased, whereas grassland management declined. Generally, these sequences of vegetation types with H. mantegazzianum can be interpreted as successional seres. Although some sites were actually in a stable state because of regular land use or maintenance, the majority of sites with H. mantegazzianum were in the process of secondary succession following abandonment or severe disturbances. These secondary successions mostly started from grassland swards, but some were from bare ground after severe disturbance (e.g. sand pits, mining or clearing of forests). Both seres will, ultimately, result in forests (Kahmen, 2004) unless land use is resumed or severe disturbances recur.

Along these successional seres, the cover of H. mantegazzianum showed a unimodal response (Fig. 8.5). In managed grasslands the cover of

Fig. 8.5. Response curves of H. mantegazzianum and selected resident species along the main gradient in 202 vegetation relevés from 20 study areas in Germany. Gradients were analysed by correspondence analysis (CA). Response curves were calculated from cover-abundance estimates on the modified Braun-Blanquet scale by Generalized Additive Models in canoco. The x-axis depicts the first CA axis representing a successional gradient from managed grasslands to tall-herb stands and woodlands. Along this gradient grassland species declined and tall herbs increased. The y-axis depicts predicted cover-abundances classes of the species. The nine classes of the modified Braun-Blanquet scale were coded numerically (1-9). The maximum predicted cover-abundance class of H. mantegazzianum (Her man) of '7' corresponds to 25-50% cover. Abbreviations of species names: Aeg pod = Aegopodium podagraria L., Hol lan = Holcus lanatus L., Ran rep = Ranunculus repens L., Urt dio = Urtica dioica L. Modified from Thiele and Otte (2006b).

Fig. 8.5. Response curves of H. mantegazzianum and selected resident species along the main gradient in 202 vegetation relevés from 20 study areas in Germany. Gradients were analysed by correspondence analysis (CA). Response curves were calculated from cover-abundance estimates on the modified Braun-Blanquet scale by Generalized Additive Models in canoco. The x-axis depicts the first CA axis representing a successional gradient from managed grasslands to tall-herb stands and woodlands. Along this gradient grassland species declined and tall herbs increased. The y-axis depicts predicted cover-abundances classes of the species. The nine classes of the modified Braun-Blanquet scale were coded numerically (1-9). The maximum predicted cover-abundance class of H. mantegazzianum (Her man) of '7' corresponds to 25-50% cover. Abbreviations of species names: Aeg pod = Aegopodium podagraria L., Hol lan = Holcus lanatus L., Ran rep = Ranunculus repens L., Urt dio = Urtica dioica L. Modified from Thiele and Otte (2006b).

H. mantegazzianum was constrained by land use and in successional seres starting on bare ground plant cover was generally low due to recent disturbance. The highest cover of H. mantegazzianum was found in young stages of succession or in sites where succession was blocked by permanent ongoing low-intensity maintenance, such as neglected road verges and grassland margins. With increasing successional age, cover of H. mantegazzianum declined again, whereas that of native tall herbs steadily increased. Finally, when woody components took over, cover of H. mantegazzianum was more and more constrained by increasing shade.

Hence, it appears that declining cover of H. mantegazzianum with increasing successional age is attributable to interspecific competition with other tall-herbs and woody species. However, it should be borne in mind that the data were single records from different sites in different successional stages and not multi-temporal observations of the same sites. Therefore, we do not have direct evidence for native species reducing H. mantegazzianum cover during succession at a particular site. The relationship between H. mantegazzianum cover and successional age could, on the one hand, be attributed to less successful invasion into old successional stages or, on the other hand, to successful invasion into young successional stages followed by declining H. mantegazzianum cover due to competition increasing with successional age.

Was this article helpful?

0 0

Post a comment