COS 120-6 - Biodiversity loss in kwongan heathland and the use of simulation modeling to assess management strategies

Thursday, August 11, 2011: 3:20 PM
18D, Austin Convention Center
Nancy Shackelford1, Richard J. Hobbs2, Michael Renton3, Kristine Brooks4 and Michael Perring3, (1)Mitacs, Victoria, BC, Canada, (2)School of Plant Biology, The University of Western Australia, Perth, Australia, (3)School of Plant Biology, University of Western Australia, Crawley, Australia, (4)Narrogin, Department of Environment and Conservation, Narrogin, Australia
Background/Question/Methods

Biodiversity loss is a recognized threat worldwide.  Human population expansion and economic development have lead to habitat loss, resource depletion, land use change, and dramatic disturbance regime shifts.  In this presentation, we focus on kwongan, a sandplain heath system in one of the world’s biodiversity hotspots, the southwest of Western Australia.  Kwongan heathland contains immense floristic biodiversity and endemism.  In the past few decades, there has been heightening concern from land managers that this biodiversity is slowly being lost, even in the most pristine areas.  To quantify this hypothesized loss, we updated published floristic surveys taken in 1983 on a government-owned nature reserve.  We repeated those surveys in 2010 and assessed individual species loss and overall decreases within functional groups.  We used the biodiversity data to estimate a relationship between density of an invading tree, Allocasuarina huegeliana, and reduction in diversity levels.  This allowed us to construct a mechanistic population model of A. huegeliana and estimate its impacts on biodiversity in the heath.  We used the model to predict how different management options based on various burning regimes and tree removal would affect density over time.  This enabled us to identify management options likely to maintain target threshold levels of biodiversity.

Results/Conclusions

We found that in the last 27 years, dramatic loss of species richness occurred almost uniformly over functional types, implying a significant species decline that is not an artifact of the sampling time of year or the poor climatic conditions of 2010.  There is a significant correlation between the rates of loss and the densities of the native invasive tree.   However, from the simulation, we found no threshold of state change from heath to A. huegeliana woodland; increasing fire frequency controlled the tree regardless of initial starting population size and the regularity of fire did not affect rates of density increase.  This implies that A. huegeliana population dynamics should not vary dramatically with an irregular fire occurrence.  When planning management, primary consideration can thus be given to heath health and an intermediate fire regime established.  Additionally, we found that control could be maintained with the implementation of fire alone, without the added necessity of managed removal.  An important variable in all simulated management was the mortality rate of A. huegeliana during a fire.  It is pivotal that fires for A. huegeliana management achieve high mortality levels or the population increases occur at exponentially faster rates.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.