Facebook Twitter Flickr Google + Blogger

Australasian Plant Conservation

Originally published in Australasian Plant Conservation 19(1) June - August 2010, p 22-23

Fencing to exclude livestock grazing enhances biodiversity values of remnant York Gum–Jam woodlands in Western Australia

Rachel Standish1, Suzanne Prober2 and Georg Wiehl2
1 School of Plant Biology, University of Western Australia, Crawley.
2 Sustainable Agriculture Flagship and CSIRO Sustainable Ecosystems, Wembley, WA. Email: Suzanne.Prober@csiro.au

Figure 1. A fenced remnant of York Gum–Jam woodland showing significant York Gum recruitment. This site was fenced and burnt in c. 2003. No recruitment was evident in the near-by unfenced plot, which was burnt at the same time.
© 2009 CSIRO. Photo: Suzanne Prober

The wheat and sheep farming district of south-west Western Australia, known locally as the wheatbelt, retains an extraordinarily rich biological diversity despite extensive clearing of native vegetation and widespread land degradation. The unique features of the wheatbelt, including the ancient nutrient-impoverished soils and its relatively recent development for agriculture, make it difficult to predict limits to biodiversity persistence and recovery compared to agricultural landscapes elsewhere (Prober & Smith 2009; Standish & Hobbs 2009). Despite a long history of natural resource management programs that support biodiversity conservation in the region, the effectiveness of one of the oldest and most widespread interventions—fencing remnant vegetation to manage livestock grazing—has rarely been tested.

Our research project

We compared vegetation and soil condition in 29 unfenced and 29 fenced sites (paired), and 11 reference sites in York Gum (Eucalyptus loxophleba subsp. loxophleba)–Jam (Acacia acuminata) woodlands within the Avon catchment. These woodlands are rated high priority for biodiversity conservation. Fenced sites had been fenced for between two and 22 years, and reference sites (the least degraded woodlands we could find in the region) were selected as benchmarks.

We tested two specific hypotheses:

  1. Fencing facilitates recovery of degraded York Gum–Jam woodlands towards conditions of the reference woodlands; and
  2. Recovery of degraded woodlands towards reference condition after fencing is constrained by ecological or other limits.

What did we find?

Consistent with our first hypothesis, we found that fenced sites differed significantly from unfenced sites in species richness of most groups of native plants, native cover, exotic cover and tree recruitment, and the direction of these differences was generally towards reference conditions. Further, frequency of Jam increased with increasing time since fencing, exotic cover decreased with time since fencing, and fenced plots were more similar to reference plots in floristic composition than unfenced plots were. However, soil nutrient levels in topsoils did not differ between fenced and unfenced sites.

Our second hypothesis was also supported for most condition measures. On average, soil nutrients were elevated, exotic cover was higher, and native richness was lower in fenced compared with reference sites, and ordination analyses suggested soils and understorey of fenced sites reached reference condition in only a small subset of cases. This may be partly due to lack of sufficient time since fencing, but regression analysis suggested that recovery from exotic invasion is more limited at higher soil nutrient levels, and that recovery of native species richness is constrained by the persistence of exotics. On the other hand, recruitment of Jam in fenced sites was similar to that in reference sites. York Gum recruits were absent from 81 per cent of all sites (including reference sites), but there was a pulse of recruitment in some fenced plots that suggested fencing can enable episodic recruitment in conjunction with other disturbances (Fig. 1).

We conclude that fencing to exclude livestock grazing is often effective for enhancing the biodiversity conservation values of remnant York Gum–Jam woodlands. However, additional interventions are likely to be necessary to promote restoration in some cases, particularly where a lack of native propagules, elevated soil nutrients and concomitant invasion by exotic species, prevent the recovery of native species.

The full report (Prober et al. 2009) is available from the authors or from Wheatbelt NRM Inc. <http://www.wheatbeltnrm.org.au>.


This project was funded by Wheatbelt NRM Inc. with support from the Australian and Western Australian Governments. WWF-Australia, the Department of Environment and Conservation (WA) and the Department of Agriculture and Food (WA) contributed to the project. We thank the numerous participating land managers for their generous hospitality and for sharing their stories with us.


Prober, S.M. and Smith, F.P. (2009). Enhancing biodiversity persistence in intensively-used agricultural landscapes: a synthesis of 30 years of research in the Western Australian wheatbelt. Agriculture Ecosystems and Environment 132, 173-91.

Prober, S.M., Standish, R.J. and Wiehl, G. (2009). After the fence: condition of fenced, unfenced and reference York Gum – Jam woodlands in the Avon Catchment, Western Australia. CSIRO Sustainable Ecosystems and Sustainable Agriculture Flagship, Perth.

Standish, R.J. and Hobbs, R.J. (2009). Restoration of OCBILs in south-western Australia: Response to Hopper. Plant Soil DOI 10.1007/s11104-009-0182-z.