Stress triangle: do introduced predators exert indirect costs on native predators and prey?

Abstract

Non-consumptive effects of predators on each other and on prey populations often exceed the effects of direct predation. These effects can arise from fear responses elevating glucocorticoid (GC) hormone levels (predator stress hypothesis) or from increased vigilance that reduces foraging efficiency and body condition (predator sensitive foraging hypothesis); both responses can lead to immunosuppression and increased parasite loads. Non-consumptive effects of invasive predators have been little studied, even though their direct impacts on local species are usually greater than those of their native counterparts. To address this issue, we explored the non-consumptive effects of the invasive red fox Vulpes vulpes on two native species in eastern Australia: a reptilian predator, the lace monitor Varanus varius and a marsupial, the ringtail possum Pseudocheirus peregrinus. In particular, we tested predictions derived from the above two hypotheses by comparing the basal glucocorticoid levels, foraging behaviour, body condition and haemoparasite loads of both native species in areas with and without fox suppression. Lace monitors showed no GC response or differences in haemoparasite loads but were more likely to trade safety for higher food rewards, and had higher body condition, in areas of fox suppression than in areas where foxes remained abundant. In contrast, ringtails showed no physiological or behavioural differences between fox-suppressed and control areas. Predator sensitive foraging is a non-consumptive cost for lace monitors in the presence of the fox and most likely represents a response to competition. The ringtail's lack of response to the fox potentially represents complete naiveté or strong and rapid selection to the invasive predator. We suggest evolutionary responses are often overlooked in interactions between native and introduced species, but must be incorporated if we are to understand the suite of forces that shape community assembly and function in the wake of biological invasions.

Figure 1. Stress triangle: Non-consumptive costs of introduced predators on native predators and prey, and the potential pathways of these costs.

The ‘predator stress hypothesis’ predicts that fear will drive a chronic glucocorticoid (GC) response that leads to immunosuppression and consequent increases in parasite load. The ‘predator sensitive foraging hypothesis’ predicts that animals will reduce their foraging efficiency and hence lose body condition. This may lead directly to immunosuppression and an increase in parasite load, or indirectly by stimulating a GC response. These pathways can be set in train by an introduced predator via competition or the risk of predation on native predators and prey, respectively, forming the base of the triangle. Arrows show direction of pathway, with symbols representing an increase (+) or decrease (−) in an individual’s response.

Figure 2. Map of the Cape Conran Coastal Park and Murrungowar state forest study region, situated in Far East Gippsland, Victoria (37°48′S 148°52′E).

Dots indicate the locations of the 76 transect lines used for trap and sand pads to sample the lace monitor (Varanus varius). Colour of dots indicates treatment area; black represents fox (Vulpes vulpes) suppression areas (baited with 1080 poison) with low densities of V. vulpes, white represents control areas (baited with non-toxic baits) with high densities of V. vulpes.

Figure 3. Percentage of common ringtail possums (Pseudocheirus peregrinus) that adopted low, medium or high vigilance behaviours in response to capture in areas of fox suppression (with low densities of foxes) and control (high densities of foxes).

Capture responses were categorized into 3 vigilance levels; low (successful capture, no escape attempt), medium (delayed escape behaviour), and high (immediate escape behaviour).

Figure 4. Foraging choice (%) of lace monitors (Varanus varius) in areas of fox suppression (with low densities of foxes) and control (high densities of foxes).

Choice of foraging behaviour was between high risk-high reward sites (represented by traps) vs. low risk-low reward sites (represented by sand pads).