Temporal Patterns in the Abundance of a Critically Endangered Marsupial Relates to Disturbance by Roads and Agriculture

Abstract

The aim of this study was to investigate how landscape disturbance associated with roads, agriculture and forestry influenced temporal patterns in woylie (Bettongia penicillata) abundance before, during and after periods of rapid population change. Data were collected from an area of approximately 140,000 ha of forest within the Upper Warren region in south-western Australia. Woylie abundance was measured using cage trapping at 22 grid and five transect locations with varying degrees of landscape disturbance between 1994 and 2012. We found evidence that the distribution and abundance of woylies over time appears to be related to the degree of fragmentation by roads and proximity to agriculture. Sites furthest from agriculture supported a greater abundance of woylies and had slower rates of population decline. Sites with fewer roads had a greater abundance of woylies generally and a greater rate of increase in abundance after the implementation of invasive predator control. The results of this study suggest that landscape disturbance is less important at peak population densities, but during times of environmental and population change, sites less dissected by roads and agriculture better support woylie populations. This may be due to the role these factors play in increasing the vulnerability of woylies to introduced predators, population fragmentation, weed species invasion, mortality from road collisions or a reduction in available habitat. Strategies that reduce the impact of disturbance on woylie populations could include the rationalisation of forest tracks and consolidation of contiguous habitat through the acquisition of private property. Reducing the impact of disturbance in the Upper Warren region could improve the resilience of this critically important woylie population during future environmental change.

Fig 1. Location of 22 grids and five transects where woylie monitoring was conducted in the Upper Warren, south-western Australia, between 1994 and 2012.

Grey shading represents state owned forest managed by the Western Australian Department of Parks and Wildlife. White shading represents privately owned land used for livestock farming and forestry.

Fig 2. Mean capture rate at 22 grid locations between 1994 and 2009.

Grey solid line represents the mean capture rate of the 18 sites that had similar abundance curve profiles and the grey shading indicates standard error. The remaining lines represent sites that were unusual either because they had different shaped abundance curves (K2-2, K3-1 and K3-2) or peaked and declined earlier than all other sites (C1). Bars associated with the mean of the similar 18 sites in 2009 indicate standard error.

Fig 3. Mean capture rate at five transects in the Upper Warren between 2001 and 2012.

Fig 4. Mean capture rate of grid sites over time grouped by level of disturbance.

a) Road density. Low: <1500 m, n = 6. Intermediate: 1600–2400 m, n = 6. High: >2400 m, n = 6. b) Proximity to agriculture. Low: >900 m, n = 4. Intermediate: 400–810 m, n = 6. High: <300 m, n = 8. c) Time since timber harvest. Low: 44 yrs in 1994, n = 2; 45 yrs in 1995, n = 3; 46 yrs in 1996 onwards, n = 2. Intermediate: 21–24 yrs in 1994, n = 10; 25 yrs in 1995, n = 6; 26 yrs in 1996 onward, n = 5. High: harvested in 1995 or 1996, n = 9 in 1995 and n = 11 in 1996 onwards. Transparent grey vertical line indicates time of commencement of quarterly aerial fox baiting. In Fig 3C, site classification varied between 1994 and 1996 so 1994 and 1995 are represented by data points only. In 1994, there were no sites classified as highly disturbed by timber harvesting. Bars indicate standard error.

Fig 5. Mean capture rate of trap points at Balban transect over time grouped by level of disturbance.

a) Proximity to agriculture. Low: >2000 m, n = 11. Intermediate: 1000–2000 m, n = 18. High: <1000 m, n = 22. b) Road density. Low: <1500 m, n = 16. Intermediate: 1500–2000 m, n = 16. High: >2000 m, n = 17. Bars indicate standard error.

Fig 6. Mean capture rate of trap points at Warrup transect over time grouped by level of disturbance.

a) Time since timber harvesting. Low: >42 years, n = 29. Intermediate: 22 years, n = 16. b) Proximity to agriculture. Low: >2000 m, n = 11. Intermediate: 1000–2000 m, n = 27. High: <1000 m, n = 12. c) Road density. Low: <1500 m, n = 12. Intermediate: 1500–2000 m, n = 10. High: >2500 m, n = 13. Note: there was no high disturbance category for time since timber harvesting for the Warrup transect. Bars indicate standard error.

Fig 7. Mean capture rate of the Boyicup, Camelar and Moopinup transect trap points over time grouped by level of disturbance.

a) Proximity to agriculture. Low: >2000 m, n = 47. Intermediate: 1000–2000 m, n = 47. High: <1000 m, n = 49. b) Road density. Low: <1200 m, n = 40. Intermediate: 1200–1800 m, n = 50. High: >1800 m, n = 53. Bars indicate standard error.

Fig 8. Analysis of Principal Coordinates (PCO) based on the time since timber harvesting, proximity to agriculture and road density of each of 22 grids between 1996 and 2009.

Blue vector overlays represent Pearson’s correlation coefficients of mean capture rate during a particular year against the PCO axes. Black vector overlays represent Pearson’s correlation coefficients of these variables against the PCO axes. Vector length indicates strength of correlation. The analysis was based on Euclidian distances calculated from square-root transformed values.

Fig 9. Analysis of Principal Coordinates (PCO) based on the time since timber harvesting, proximity to agriculture and road density of 242 trap points along five transects in the Upper Warren (Boyicup, Moopinup, Camelar, Warrup and Balban).

Vector overlays labelled with year represent Pearson’s correlation coefficients of mean capture rate during a particular year against the PCO axes. Vector overlays labelled with landscape variables represent Pearson’s correlation coefficients of these variables against the PCO axes. Vector length indicates strength of correlation. The analysis was based on Euclidian distances calculated from square-root transformed values.