Posted by: oikosasa | February 4, 2013

Pasture – red kangaroo – dingo interactions

Read David Choquenot’s and David M. Forsyth’s new Early View paper “Exploitation ecosystems and trophic cascades in non-equilibrium systems: pasture – red kangaroo – dingo interactions in arid Australia” to learn more!

Here’s Dave’s background story to the study:

Fig1_Dave Forsyth at Kinchega National Park

This article had a long gestation. The seeds were sown in 2000, when two influential articles were published in The American Naturalist on the inter-related topics of trophic cascades (Schmitz et al. 2000 Am. Nat. 155, 141) and the Exploitation Ecosystems Hypothesis (EEH; Oksanen and Oksanen 2000 Am. Nat. 155, 703). After reading these articles we discussed the idea of adding the dingo (the top-order predator in mainland Australia) to Graeme Caughley’s two-link rainfall – pasture – red kangaroo model (Caughley & Gunn 1993 Oikos 67, 47), to test whether an empirically derived model could recreate EEH predictions and generate trophic cascades. In Caughley’s system, which was based on data collected in Kinchega National Park (western NSW; Image 1), prevailing productivity is tightly linked to rainfall through its effect on pasture growth and dieback. However, rainfall in this ecosystem is highly stochastic between seasons and years. Over the subsequent ten years we worked sporadically to test whether this system could produce dynamics consistent with the EEH and trophic cascades.

The model did reproduce the three zones predicted by the EEH, but a surprising outcome was the discovery of an additional zone at productivities above which the maximum densities of the dingo was achieved. The additional zone, in which kangaroo densities increased and pasture biomass declined due to the re-engagement of the kangaroo-pasture feedback loop, occurred because dingo densities are believed to be socially regulated (via dominant female infanticide): if dingo densities are instead constrained wholly by the availability of kangaroos then that zone disappears, kangaroos become less abundant and pasture biomass more abundant.


Increasing stochasticity in seasonal rainfall had sometimes counter-intuitive effects on model outcomes. High levels of stochasticty led to more frequent extinction of dingoes from the system, resulting in the re-engagement of the kangaroo-pasture feedback loops. Hence, increasing stochasticity led to increased attenuation in this system.

Roger Pech (Landcare Research, New Zealand) thoughtfully suggested that we use the normalized difference (rather than the absolute difference) of the log-response ratios to evaluate attenuation in this system. Roger’s suggestion will be appropriate to other studies assessing attenuation in trophic cascades.

Several journal reviewers also suggested that we assess the effects of potential diet switching by dingoes from kangaroos to reptiles, as has been observed in some areas of arid Australia. We found that prey switching by dingoes to reptiles weakened trophic cascades.

The role of the dingo as a trophic regulator has been the subject of much recent debate, with some scientists calling for culling to cease and reintroductions to be made in areas where it has been extirpated. The rationale for returning dingoes to previous densities in parts of their range focuses primarily on their potential to reduce the abundance of introduced red foxes and feral cats. However, our study suggests that additional benefits may occur through regulation of large kangaroo abundance, and the associated release of vegetation from grazing pressure. Depending on the degree to which the diversity of each trophic level is maintained by consumption-mediated co-existence, these changes may have flow on implications for amongst herbivores and vegetation biodiversity in these ecosystems.

Our study has generated testable predictions about interactions between top-order carnivores, their prey, and vegetation across productivity gradients. These predictions are obviously highly testable in Australia where dingo management is widespread. However, the generality of the predictions could also be tested in entirely different predator-driven ecosystems.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: