Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Recruitment of Banksia spp. in an anthropogenically disturbed mediterranean climate type woodland in Western Australia

Crosti, Roberto (2011) Recruitment of Banksia spp. in an anthropogenically disturbed mediterranean climate type woodland in Western Australia. PhD thesis, Murdoch University.

PDF - Front Pages
Download (231kB)
PDF - Whole Thesis
Download (6MB)


Introduction Banksia woodland is the characteristic mediterranean type vegetation of the sand plains of the Swan Coastal Plain in the Perth region of Western Australia. Once extensive in the past the woodlands are nowadays severely fragmented and threatened by a variety of perturbations including weeds, fire and seed predation. Conservation and management of the woodlands depends very much upon the knowledge of past land use, type of original vegetation, frequency and magnitude of significant disturbances and surrounding land-use.

The results of a long-term study in the Kings Park bushland, comparing vegetation differences between the 1939 and 1999, showed that there have been substantial changes in the composition and structure of the woodland. In term of tree species, the most significant trends have been a decrease in Banksia spp.

Aim The research reported in this thesis aims for an understanding of the features and different stages of Banksia spp. recruitment in an environment such as the Kings Park bushland urban remnant. The final objective of this study is to assess if recruitment of Banksia spp. is affected by the perturbations mentioned above, ultimately influencing bushland conservation.

Materials &Methods The study was conducted both in urban bushland and in suburban areas in the Perth region. Several of the experiments and investigations on some specific stages of recruitment also included other species dominant in the bushland. The stages of recruitment investigated are: seed production and dispersal, seed germination and seedling establishment, seed predation.

Results In the Perth region B. attenuata and B. menziesii have a small degree of serotiny, the two species release seeds at different times of the year - B. attenuata from Summer to Autumn and B. menziesii from Spring to Summer. In post- fire conditions seed release is more synchronised than for unburned trees but the amount of seed released depends on the timing of the fire. Fires in early Summer lead to mass release of B. menziesii seeds and destruction of immature seeds of B. attenuata. Later fires destroy B. menziesii seed that has been shed on to the soil but synchronises release of B. attenuata seed.

Of the investigated locations Kings Park, and Ridge State Forest, have higher level of pre-dispersal seed predation compared to other locations. The post-dispersal surveys and experiment, both in inter-fire and post-fire environment show that in Kings Park in Winter, approximately 75% of the banksia seeds were consumed. Moreover, Banksia seed predation is not related to the density of seeds. Presence of leaf litter, weeds and plant material in the experimental unit reduced predation values. Most of the germination occurred on sandy soil, while presence of weeds and of a thick layer of leaf litter reduced germination numbers for the Banksia species. In post-fire conditions seedlings are also associated with “burn residuals” (accumulation of burned litter and ash). Most of the seeds that were sown in the soil and had germinated died.

Conclusion This project confirmed and quantified the decrease of banksia individuals (which until now was only assumed) over time and lack of their recruitment. Seed predation, both before and after seed release greatly affects banksia seed availability in Kings Park. Furthermore unfavourable sites for germination such as a thick layer of leaf litter and weeds are a great obstacle for establishment of large seeded species such as Banksia. Other dominant species, such as Allocasuarina fraseriana, Acacia saligna and Dryandra sessilis1 on the contrary do not seem to reduce their recruitment potential as predation (or removal) is not conspicuous and the various substrata do not reduce the ratio between the number of germinants and seedlings that survive. In order to boost banksia recruitment a number of management actions could be taken. In particular weeds and thick layers of leaf litter must be removed and a sandy soil substratum should be arranged especially in the areas where, even in the presence of banksia trees, the number of seedlings is low. Predation must be reduced, in particular through removal of pests.

Item Type: Thesis (PhD)
Murdoch Affiliation(s): School of Environmental Science
Supervisor(s): Ladd, Phil
Item Control Page Item Control Page


Downloads per month over past year