Results of a food addition experiment in a north-central Chile small mammal assemblage: evidence for the role of "bottom-up" factors

Since 1989, effects of biotic interactions including predation and herbivor y have been examined in a replicated experimental study in a north-central Chilean semiarid thorn scrub community. Strong responses of small mammals a nd plants to El Nino Southern Oscillations (ENSO) have also been documented suggesting that "bottom-up" factors related to high rainfall are important . To simulate increased primary productivity effects on small mammals, ad l ib rabbit pellet additions were initiated in mid-1997 on unfenced grids nea r the experimental complex. Following the 1997 El Nino event with three tim es normal precipitation, numbers of small mammals during pre-addition month s and the first treatment year were similar on control and food addition gr ids. During the second year (1998-1999), a period of severe drought, food a dditions had significant positive effects on numbers of two predominantly h erbivorous "core" (resident) species, Octodon degus and Phyllotis darwini, and an omnivorous "quasi-core" (resident but highly fluctuating) species, A kodon olivaceus; however, all three species declined towards the end of the second treatment year, Two "opportunistic" (temporarily resident) species, Abrothrix longipilis (an insectivore) and Oligorvzomys longicaudatus (a gr anivore), showed no responses to food additions. An insectivorous marsupial , Thylamys elegans (also a "core species"), had significantly lower numbers on food addition grids. Changes in body weight distributions and proportio ns of reproductive individuals particularly in O. degus indicate in situ re sponses. Whereas no differences in residency, numbers of stations visited, and trappability were observed, energy compensation ratios greater than one suggest significant immigration in the second year. Thus, food additions e licited strong responses by herbivorous/omnivorous "core" and "quasi-core s pecies" whereas they had no effects on "opportunistic species". These resul ts reinforce the view that "bottom-up" factors influencing food availabilit y exert prevailing control on numerically important small mammal species by temporarily increasing carrying capacity, and that "top-down" factors (i.e ., biotic interactions) become important when small mammal numbers are at o r near their carrying capacity, Spatial dynamics may be important in explai ning declines of species populations exhibiting initially positive response s to food additions.