Identifying factors that may be responsible for regulating the size of animal populations is a cornerstone in understanding population ecology. The main factors that are thought to influence population size are either resources (bottom-up), or predation (top-down), or interspecific competition (parallel). However, there are highly variable and often contradictory results regarding their relative strengths and influence. These varied results are often interpreted as indicating "shifting control" among the three main factors, or a complex, nonlinear relationship among environmental variables, resource availability, predation, and competition. We argue here that there is a "missing link" in our understanding of predator-prey dynamics. We explore whether the landscape-of-fear model can help us clarify the inconsistencies and increase our understanding of the roles, extent, and possible interactions of top-down, bottom-up, and parallel factors on prey population abundance. We propose two main predictions derived from the landscape-of-fear model: (1) for a single species, we suggest that as the makeup of the landscape of fear changes from relatively safe to relatively risky, bottom-up impacts switch from strong to weak as top-down impacts go from weak to strong; (2) for two or more species, interspecific competitive interactions produce various combinations of bottom-up, top-down, and parallel impacts depending on the dominant competing species and whether the landscapes of fear are shared or distinctive among competing species. We contend that these predictions could successfully explain many of the complex and contradictory results of current research. We test some of these predictions based on long-term data for small mammals from the Chihuahuan Desert in the United States, and Mexico. We conclude that the landscape-of-fear model does provide reasonable explanations for many of the reported studies and should be tested further to better understand the effects of bottom-up, top-down, and parallel factors on population dynamics.

No abstract

No abstract

Drylands worldwide are experiencing shrub encroachment into grasslands with potential consequences for biodiversity and ecosystem services. Climate change could increase the rate of shrub encroachment, amplify precipitation variability, and thus alter bottom‐up processes for animal communities. Desert rodents are important biodiversity elements of arid grasslands and shrublands that exert strong effects on soil, vegetation, and other animal species. We used long‐term data from the Jornada Basin Long Term Ecological Research site in the Chihuahuan Desert of southern New Mexico to ask whether bottom‐up control of desert rodents changes across shrub encroachment gradients. Our design included spatial blocks with replicated ecological states representing transitions from black grama (Bouteloua eriopoda) to honey mesquite (Prosopis glandulosa). Grassland‐to‐shrubland transitions did not produce degraded ecosystems, on average, with reduced net primary production or decreased rodent biomass. However, more rodent biomass was supported on unencroached grasslands following droughts whose frequency and severity may increase in southwestern United States. Hence, the observed evenness in rodent biomass across ecological states should be sensitive to climate change. The best predictors of rodent biomass also differed markedly for two trophic groups. This outcome was explained by considering core‐transient dynamics. Granivores were mostly core species that regularly occurred on sites and responded to lagged net primary production at local scales, whereas folivores included transient species (especially Sigmodon hispidus) that responded to lagged precipitation at broader scales via spillover dynamics. Bottom‐up processes for desert rodents across shrub invasion gradients were understood by integrating lagged responses to productivity pulses with core‐transient structuring of communities.