Blog written by Jennifer L. Chandler and John L. Orrock. Read the full paper here.
Because most plants die before becoming seedlings, the distribution and abundance of plants often depends upon the distribution and survival of plant seeds. Small mammals are ubiquitous granivores with the potential to determine the distribution and regeneration of plants and trees in forests. Despite their importance, patterns of rodent granivory can also be highly variable, making it difficult to predict how granivory will affect plant recruitment at large scales. While variation in productivity, seasonality, or latitude have been identified as important for predicting patterns of seed predation, often considerable variation in seed predation exists even after these factors are considered.
Since rodents are common prey of carnivores, knowledge of activity patterns of rodent predators may play a part in predicting hotspots and coldspots of seed consumption by rodents. Large carnivores can have effects that cascade down the food chain, altering ecosystem dynamics both when they are removed and reintroduced to a system. Top carnivores can affect abundance and behavior of mesocarnivores, which in turn affect abundance and behavior of their prey, usually herbivores, which can alter plant abundance and community composition. We hypothesized that distributions of apex predators can create large-scale variation in the distribution and abundance of mesopredators that consume small mammals, creating predictable areas of high and low seed survival.
The natural recolonization of northern Wisconsin by gray wolves (Canis lupus) presented a unique opportunity to test the hypothesis that interactions among carnivores affect seed consumption by rodents. By comparing areas recolonized by wolves to areas that had been essentially wolf-free since 1960, we could test whether apex predator presence indicates areas of low seed predation by rodents. Gray wolves competitively exclude coyotes (C. latrans), but better tolerate foxes (Vulpes vulpes, Urocyon cinereoargentus) because foxes have less diet overlap with wolves and are therefore, are less competition. Thus, areas with high wolf activity, such as wolf territories, are areas of relatively lower coyote activity, and higher fox activity. Foxes are expected to consume a greater proportion of small mammals, such as rodents that eat seeds, compared to coyotes. Consequently, we hypothesized that wolf territories may be areas of lower seed consumption due to the higher abundance of rodent predators.
Using multi-year field trials at sites inside and outside of 11 wolf territories in northern Wisconsin, USA, we evaluated whether rodent abundance and seed consumption were lower in wolf territories. At each site, we conducted live trapping sessions to survey rodent abundance. To measure seed consumption, we placed seed depots (plastic containers with known numbers of seeds of four tree species scattered on top of a layer of sand) at study locations for two-week periods, after which, seeds depots were collected and remaining seeds were counted. To confirm that differences between areas inside and outside of wolf territories were a result of differing interactions among carnivores, we also investigated several alternate explanations for the patterns in rodent abundance and seed survival that we observed. We measured a variety of habitat characteristics across our site, such as tree canopy cover, shrub cover, and presence of woody debris (all factors that can influence rodent abundance and activity), to rule out other potential causes of low rodent abundance and seed consumption that may be inherent in habitat where wolves preferentially establish territories.
Consistent with the hypothesized consequences of wolf occupancy, predation of seeds of three tree species was more than 25% lower inside wolf territories areas across two years. Rodent abundance was more than 40% lower in high-wolf areas during one of two study years: a result primarily driven by low southern red-backed vole (Myodes gapperi) abundance in wolf territories. The absence of significant habitat differences between high- and low-wolf areas that might affect rodent abundance or activity further supported these results. Together, our findings suggest that top-down effects of wolves on seed consumption by rodents and seed survival may occur inside wolf territories.
Accounting for the effects of interactions among carnivores on lower levels on the food chain may allow for more accurate predictions of large-scale patterns in seed survival and forest composition, as well as other important ecological processes. With the knowledge of the activity of relatively few individuals of an apex species (e.g., wolves), we were able to predict considerable variation in rodent abundance and seed survival. This finding has important practical applications in forest management; since the majority of U.S. forests rely upon natural regeneration of harvested forest stands (i.e., recruitment from seeds, as opposed to planting), understanding how top predators influence seed survival may allow forest managers to predict which stands are more likely to experience recruitment failure after harvest. Territory boundaries of apex predators may also predict patterns of ecological interactions that influence disease prevalence. For example, small mammals are important intermediate hosts of the bacteria that causes Lyme disease in humans. Areas between predator territories may be areas of high rodent abundance, and therefore, may indicate locations of increased Lyme disease risk to humans. Consideration of the top-down effects of carnivore interactions may shed new light on spatial patterns in many ecological processes with economic, human-health, and conservation consequences that may have otherwise been dismissed as anomalous.