Plant species responses along a grazing disturbance gradient in Australiangrassland

Citation
Rj. Fensham et al., Plant species responses along a grazing disturbance gradient in Australiangrassland, J VEG SCI, 10(1), 1999, pp. 77-86
Citations number
45
Categorie Soggetti
Plant Sciences
Journal title
JOURNAL OF VEGETATION SCIENCE
ISSN journal
11009233 → ACNP
Volume
10
Issue
1
Year of publication
1999
Pages
77 - 86
Database
ISI
SICI code
1100-9233(199902)10:1<77:PSRAAG>2.0.ZU;2-M
Abstract
Floristic data from paired roadside-paddock analyses from grassland in cent ral Queensland, Australia, were ordinated. The mean direction of the vector s between these pairs was almost perfectly aligned with the indirect gradie nt represented by the first axis of Non-metric Multi-dimensional Scaling. I t confirms anecdotal evidence of a trend from infrequently grazed roadsides to constantly grazed paddocks. The increasing abundance of annual herbs an d grasses along this putative gradient is consistent with documented trends from elsewhere in the world. The response patterns of individual species a long the disturbance gradient is consistent with ecological theory predicti ng unimodal peaks in abundance along physical environmental gradients. The ancestral perennial dominants of the grasslands, Dichanthium sericeum and D . queenslandicum, exhibited a declining response to grazing disturbance. Ev en the generally unpalatable perennial grass Aristida leptopoda declined co nsiderably in the upper segments of the grazing disturbance gradient. A sui te of herbaceous trailing legumes had peaks in their abundance near the mid dle of the grazing disturbance gradient, trends that can be readily explain ed by the combination of their palatability and intolerance to competition from tall perennial grasses. Several species including the noxious exotic h erb Parthenium hysterophorus showed increasing abundance along the grazing disturbance gradient. The methodology may have application as a rapid metho d of assessing disturbance impacts elsewhere, and is most suited where a ma nagement differential between paired plots can be reliably generalized and where the physical environment is relatively monotonous.