Abstract
Quantifying how anthropogenic land use affects wildlife communities is critically important for balancing socioeconomic activities with wildlife conservation and management. Wildlife often perceive human activity as a risk and respond by increasing anti-predator behaviors (e.g., vigilance or fleeing) and/or partitioning themselves in space or time. Alternatively, some species may be drawn to human activities and benefit from anthropogenic subsidies such as food, water, or habitat features. Regardless, shifts in animal behavior and spatiotemporal activity patterns can alter key ecological processes that help maintain wildlife communities. For example, human activities can impact predator-prey interactions by altering spatiotemporal overlap of predators and prey or by altering predator foraging efficiency. Predation can shape the structure and function of wildlife communities and regulate prey populations via top-down processes. Thus, quantifying how human activities influence predator-prey interactions enhances our ability to make well-informed decisions regarding wildlife management. However, quantifying the consumptive and non-consumptive effects of predators on prey and factors that influence patterns of predation is challenging because most prey have to contend with multiple predators with different functional traits. Each species of predator may respond to human activity differently and, hence, predator-specific approaches are necessary to fully understand predator-prey dynamics and how humans affect those dynamics. Therefore, identifying the explicit predator species within ecological communities and determining rates of predator-specific mortality can help elucidate the functional role specific predators play within ecological communities and how human land use influences predator-specific patterns of predation. For the greater sage-grouse (Centrocercus urophasianus; hereafter sage-grouse), nest predation is the primary cause of nest failure and can influence population dynamics. A diverse suite of predators are known to predate sage-grouse nests and thus, generalizing about how specific habitat, landscape characteristics, and human land use influence nest fate has been challenging. Greater sage-grouse populations have declined across their range in western North America and declines have been attributed to habitat loss, habitat degradation, and land use. Many land use activities thought to be responsible for sage-grouse population declines are assumed to increase nest predation, yet few studies have evaluated the effects of land use activities on explicit sage-grouse nest predators (e.g., the effects of land use on abundance and nest foraging behavior of specific predators). Domestic cattle (Bos taurus) grazing is often assumed to increase nest predation by reducing grass height (and hence nest concealment) and thereby facilitates nest detection by predators. Grass height is lower on actively grazed areas (hereafter pastures), but sage-grouse nest success appears to be similar on pastures grazed at varying intensities. The structural effects of livestock grazing (i.e., reduced nest concealment) could be offset by a numerical response of one or more sage-grouse nest predators to the presence of cattle (i.e., cattle may cause a localized numerical response in some nest predators). A reduction in the number of nest predators at the pasture scale could explain similar patterns of nest success on pastures grazed at varying intensities. That is, the effects of the two mechanisms (i.e., reduced nest concealment and a numerical response) could counteract each other if both were valid.
Chapter 1 evaluates one prediction of the numerical response hypothesis: a decreased probability of at least one sage-grouse nest predator predating sage-grouse nests in pastures with cattle relative to pastures without cattle present during the nesting season. Cameras placed at nest sites are often used to identify nest predators. However, deploying cameras can increase nest abandonment and affect nest fate by providing predators with olfactory and visual cues of nest locations. We leveraged the power of molecular methods to identify sage-grouse nest predators via predator DNA collected from eggshells from predated sage-grouse nests. We then used this information to evaluate the influence of habitat, nest-site characteristics, and grazing on predator-specific nest mortality. We also deployed artificial nests with trail cameras to determine the accuracy of our molecular method by comparing the predator species captured via trail camera with our molecular results. Our proof-of-concept study showed that non-invasive, molecular methods provide an accurate method for identifying nest predators. Our molecular analyses detected the species captured via trail camera at 95% of our artificial nests. We collected samples from 114 predated sage-grouse nests and detected predator DNA from 76 (67%) of those nests. We did not find evidence that domestic cattle grazing influenced predator-specific nest mortality. We found a negative relationship between shrub canopy cover and the probability of coyote (Canis latrans) predation and a negative relationship between ambient temperature and the probability of both coyote and corvid (Corax spp. and Pica hudsonia) predation. Our study highlights that management efforts to increase sage-grouse nest success may vary spatially depending on the predominant predator within the system and that non-invasive genetic methods to determine predator-specific nest mortality can mitigate problems and biases associated with nest cameras.
Chapter 2 evaluates a second prediction of the numerical response hypothesis: one or more nest predator species avoid cattle by partitioning themselves in space and thus, are unavailable to consume sage-grouse nests on pastures with cattle. Additionally, the effects of grazing (i.e., reduced concealment at nests) could be offset if one or more nest predators partition themselves temporally to avoid cattle. Altering diel activity patterns could result in increased nest predator activity during portions of the day when they are less efficient at locating sage-grouse nests. We deployed motion-sensor cameras across six pastures to evaluate whether coyotes (a primary sage-grouse nest predator) altered spatiotemporal activity patterns in response to cattle. Contrary to the numerical response hypothesis, the probability of detecting coyotes was positively associated with cattle detections at camera sites. Moreover, coyotes did not shift their diel activity patterns in response to cattle being in the pastures. Thus, in our system, similar patterns of sage-grouse nest success in grazed and non-grazed pastures cannot be explained by the numerical response hypothesis or by a temporal avoidance of cattle by coyotes.
Our study did not find evidence to support the numerical response hypothesis. In our system, the presence of cattle did not influence predator-specific nest mortality. Further, the presence of cattle did not result in coyotes (the predominant nest predator in our system) altering their spatiotemporal activity patterns. In fact, we found that the probability of detecting coyotes increased with cattle detections suggesting that certain habitat characteristics (e.g., a relationship between increased forage quality for cattle and increased prey availability for coyotes) may offset any perceived risk that coyotes may have of cattle. Regardless, similar patterns of sage-grouse nest success on pastures with varying grazing intensity cannot be explained by the numerical response hypothesis. Current grazing practices (e.g., the number of cattle within a pasture and duration of grazing) may provide sage-grouse with areas of nesting refugia (i.e., areas within a pasture with adequate nest concealment) which could explain similar patterns of nest success on pastures grazed at varying intensities. However, continued research evaluating explicit mechanisms that link domestic cattle grazing to sage-grouse demographics is needed.