Coupled Oscillators in an Agroecosystem: Integrating Direct and Indirect Effects.

AbstractAgricultural pests are increasingly appreciated as subjects of ecology. One particular case, a pest in coffee production, is analyzed here using the conceptual framework of complex systems, increasingly acknowledged as having an obvious home in the field of ecology, notorious for its complex structures. The particular case analyzed here arguably falls under the control of the complexity of the ecological system rather than of a simple magic bullet of population regulation. The system, which has been under study in southern Mexico for the past quarter century, is analyzed through the lens of neutral oscillations of the classical nondissipative Lotka-Volterra system. Based on three consumer/resource pairs (populations of [1] an ant, [2] a scale insect, [3] a beetle predator of the scale insect, [4] a fungal pathogen of the scale insect, and [5] a fly parasitoid of the ant), this five-dimensional system is well known qualitatively. Coupling all agents through both direct effects and trait-mediated indirect effects, the behavior of the neutral oscillation form of the system reveals a complex set of behaviors, including harmonized invariant sets, chaos, and/or quasiperiodicity. Such behaviors are well-known subjects in the science of complex systems and, it is argued, are ultimately sufficient to effect a degree of regulation on the pest, independent of explicit density-dependent feedback. Control of the system is thus seen as arguably actuated through its complexity, independent of any classic dissipative force.

Spatiotemporal Foraging Dynamics of Solenopsis invicta (Hymenoptera: Formicidae) and its Potential Effects on the Spatial Structure of Interspecific Competition.

The ant communities on coffee farms in the West/Central Mountains of Puerto Rico are composed of mainly invasive species, although many have a long history of occupation and are effectively naturalized. The ecological forces that maintain such communities are thus of interest, and are evidently related to the spatial patterns in which they inevitably occur. Furthermore, the spatial patterns in which members of the native ant community forage almost certainly include limitations related to the structure of the networks of subterranean foraging tunnels that extend from the nest mounds of Solenopsis invicta. Here we explore some details of that structure. We ask, what is the pattern of foraging exit holes and the gaps between them, and how does that pattern change from farm to farm and from time to time? We encounter typical underground foraging trails punctuated by foraging exits, which, we propose, create a structure above ground of relatively small foraging exits in a matrix of effective foraging gaps. This pattern varies from nest to nest and farm to farm. Other ant species clearly occupy those gaps and seem to gain some of their resilience in the system from this peculiarity of S. invicta's foraging area structure.

Can Conflicting Selection from Pollinators and Nectar-Robbing Antagonists Drive Adaptive Pollen Limitation? A Conceptual Model and Empirical Test.

AbstractPollen limitation is widespread, despite predictions that it should not be. We propose a novel mechanism generating pollen limitation: conflicting selection by pollinators and antagonists on pollinator attraction traits. We introduce a heuristic model demonstrating antagonist-induced adaptive pollen limitation and present a field study illustrating its occurrence in a wild population. For antagonist-induced adaptive pollen limitation to occur, four criteria must be met: (1) correlated attraction of pollinators and antagonists; (2) greater response by antagonists than pollinators to altered investment in attraction traits; (3) reduced investment in pollinator attraction, leading to pollen limitation; and (4) higher fitness for plants with reduced investment in pollinator attraction. We surveyed nectar robbery and reproductive output for 109 Odontonema cuspidatum (Acanthaceae) plants in a pollen-limited population over 2 years and used experimental floral arrays to evaluate how flower number affects pollination and nectar robbery. Both pollinators and nectar robbers preferred larger floral displays and nectar robbery reduced reproductive output, suggesting conflicting selection. Survey and experimental data agreed closely on the optimum flower number under antagonist-induced pollen limitation; this number was substantially overrepresented in the population. While criteria for antagonist-induced adaptive pollen limitation are restrictive, the necessary conditions may often be realized. Considering interactions beyond the plant-pollinator dyad illuminates previously overlooked mechanisms generating pollen limitation.

A Keystone Ant Species Provides Robust Biological Control of the Coffee Berry Borer Under Varying Pest Densities.

Species' functional traits are an important part of the ecological complexity that determines the provisioning of ecosystem services. In biological pest control, predator response to pest density variation is a dynamic trait that impacts the provision of this service in agroecosystems. When pest populations fluctuate, farmers relying on biocontrol services need to know how natural enemies respond to these changes. Here we test the effect of variation in coffee berry borer (CBB) density on the biocontrol efficiency of a keystone ant species (Azteca sericeasur) in a coffee agroecosystem. We performed exclosure experiments to measure the infestation rate of CBB released on coffee branches in the presence and absence of ants at four different CBB density levels. We measured infestation rate as the number of CBB bored into fruits after 24 hours, quantified biocontrol efficiency (BCE) as the proportion of infesting CBB removed by ants, and estimated functional response from ant attack rates, measured as the difference in CBB infestation between branches. Infestation rates of CBB on branches with ants were significantly lower (71%-82%) than on those without ants across all density levels. Additionally, biocontrol efficiency was generally high and did not significantly vary across pest density treatments. Furthermore, ant attack rates increased linearly with increasing CBB density, suggesting a Type I functional response. These results demonstrate that ants can provide robust biological control of CBB, despite variation in pest density, and that the response of predators to pest density variation is an important factor in the provision of biocontrol services. Considering how natural enemies respond to changes in pest densities will allow for more accurate biocontrol predictions and better-informed management of this ecosystem service in agroecosystems.

Clusters of ant colonies and robust criticality in a tropical agroecosystem.

Although sometimes difficult to measure at large scales, spatial pattern is important in natural biological spaces as a determinant of key ecological properties such as species diversity, stability, resiliency and others. Here we demonstrate, at a large spatial scale, that a common species of tropical arboreal ant forms clusters of nests through a combination of local satellite colony formation and density-dependent control by natural enemies, mainly a parasitic fly. Cluster sizes fall off as a power law consistent with a so-called robust critical state. This endogenous cluster formation at a critical state is a unique example of an insect population forming a non-random pattern at a large spatial scale. Furthermore, because the species is a keystone of a larger network that contributes to the ecosystem function of pest control, this is an example of how spatial dynamics at a large scale can affect ecosystem service at a local level.