Spatial scales of predator-prey interactions in model and experimental ladybug-aphid systems

Ecological processes operate at a variety of spatial and temporal scales. Therefore, our understanding about a system may depend on the observation scale. Predator-prey interactions include top-down effects (TD; negative effects of predator on prey) and bottom-up effects (BU; positive effects of prey on predator). Focusing on a ladybug-aphid system, we hypothesized that TD and BU are spatial scale-dependent. Top-down effects are predicted stronger at smaller scales. Bottom-up effects are predicted (i) stronger at larger scales if population dynamics is dominated by demographic processes; (ii) alternatively, stronger at smaller scales if dispersal is dominating. We tested these predictions using field experiment and computer simulation. We arranged 81 potted plants (Iva fructescens) in a triangular array and counted daily their aphid herbivores (Uroleucon ambrosiae) and ladybugs (Cycloneda sanguinea). TD and BU indices were calculated from the time-series data for each plant and for larger scales after summing over each 3-plant, 9-plant, and 27-plant sets. We found that TD effects were significant at all scales, whereas BU effects were only significant at smaller scales. The trend for BU effects confirmed prediction (ii), suggesting that dispersal is the main cause of the observed BU effects. This idea was also supported by simulating the model that excluded predator demographic terms and showed similar results. In conclusion, ladybug-aphid interactions are dependent on spatial scale. Understanding the mechanisms creating scale-dependency provides an avenue for linking behavioral processes with community consequences.