Abstract

Since early 1989, we have conducted a larger-scale ecological manipulation in a semiarid thorn scrub community in north-central Chile. We have excluded vertebrate predators (raptors and mammalian carnivores), and larger small mammal herbivores/competitors (i.e., degus, Octodon degus) from replicated 0.56-ha plots, and monitored small mammal population and plant responses over more than ten years. Repeated measures ANOVAs on minimum number known alive (MNKA) estimates of small mammals for a six-year period (1990-1996) spanning an El Nino event in 1991-1992 showed strong responses of some species to predator exclusions (e.g., O. degus; Darwin's leaf-eared mouse, Phyllotis darwini; the chinchilla-rat, Abrocoma bennetti). However, responses varied in time with significant effects during pre-El Nino (1990-1992) and El Nino (1992-1994) periods (i.e., O. degus), or pre-El Nino and post-El Nino (1994-1996) periods (P. darwini, A. bennetti). Other species showed no responses to predator exclusions (e.g., olivaceous field mouse, Akodon olivaceus; long-haired field mouse, Abrothrix longipilis; long-tailed rice rat, Oligoryzomys longicaudatus). Some effects of competitor (degu) exclusions were detected (e.g., A. bennetti during the El Nino and post-El Nino periods; O. longicaudatus during the El Nino). 'Top-down' factors (i.e., biotic interactions) appear to have greater effects on 'core species' (i.e., P. darwini, O. degus) which persist in the thorn scrub. Other species (e.g., A. longipilis, O. longicaudatus) are transitory residents or 'opportunistic' with lesser effects of biotic interactions, and their populations may be controlled by source-sink dynamics. All species had strong responses to the 1991-1992 El Nino indicating primary control by 'bottom-up' factors.