Behavioural alterations as part of an agricultural edge effect

Abstract

Predation risk allocation hypothesis (Lima & Bednekoff, 1999) predicts that prey activity will decrease when predation risk is high only if there is a high temporal variation in the predation risk. Otherwise, it will remain uniform. Our study (Shapira, Sultan & Shanas, 2008) suggests a spatial component for this hypothesis. The temporal variation in the predation risk is governed by the moon phase, while the spatial variation is related to proximity to agricultural farms. The sandy habitat of the Arava valley, where many endemic species have evolved, is under increasing threat due to agricultural farming sprawl. Our study calls attention to land transformation effects, which may exceed diversity and community degradations by indirectly altering the behaviour of native species. The edge effect may therefore be wider than originally thought. We identified the red fox Vulpes vulpes as a major beneficiary, and an agent of this change. Blumstein (2008) is concerned that predators other than foxes, including past predators, may have caused the behavioural differences in gerbil foraging across the Israel– Jordan border. Gompper & Vanak (2008) also require more data on foxes predating gerbils. Gerbils are part of the red fox diet (Shapira et al., 2008), but as the fox is an omnivorous predator the relative contribution of gerbils to its diet may vary. In a 2-year study in the Arava valley (Shanas et al., 2006), we counted 30 incidents (out of 490 trappings) of foxes preying on gerbils caught in Sherman traps. The foxes actually learned to check the traps during the night and consume the ‘caught meal’ before our arrival in the morning. We agree that to prove that the red fox is the main predator causing fear in gerbils, more experiments are needed. However, the differential abundance of red fox across the border is supported by multiple observations, and now also by track count. Furthermore, the increased fox activity on the Israeli side of the border was apparent not only in transect tracks count, as Blumstein (2008) suggests, but also in the number of locations where fox activity was recorded. This latter measure was significantly higher close to farms. The proposed alternative methodology of capture–recapture for estimating population size has its own drawbacks and may prove to be misleading, especially when used on a relatively highly intelligent mammal such as the red fox. The multipredator theory, formulated by Blumstein (2006), suggests that anti-predator behaviour can be maintained by the presence of any predator, even if some of the natural predators are absent. In the case of the Arava valley, potential gerbil predators such as the sand cat Felis margarita may have become extinct near farms due to human activity. However, the giving-up-density (GUD) study (Shapira et al., 2008) seems to show that the gerbils’ anti-predator behaviour was maintained near farms most probably because of the replacement of cats, and perhaps sand foxes Vulpes ruppellii too, by red foxes as potential predators. Yet, even if the cause of the differential antipredator behaviour across the border is non-red fox past or present predators, the results still suggest that human activity may culminate in behavioural alterations in gerbils. The red fox has been listed as one of the worst invasive species (IUCN, 2005). While the direct effect on prey species is obvious, elucidating cascade effects is important, albeit challenging. For example, the introduction of the arctic fox Alopex lagopus into subarctic islands caused a cascade effect of transforming a plant community from grassland to tundra (Croll et al., 2005). Our study suggests that the effect of commensal predators should be assessed not only on the population level but also behaviourally. Quite possibly, factors other than farming (e.g. mismanagement of garbage) could have contributed to the increased abundance of foxes. But whatever the main cause, the consequences of humans settling and transforming the Arava sands into agricultural land are a greater abundance of foxes. Dickman (2008) suggested that human activity in the farms or roads may also somehow affect the gerbils’ behaviour. We favour the idea that farming has worked its effect on gerbils only