Event-driven competing risks

Abstract

The California red scale Aonidiella aurantii (Maskell) (Homoptera: Diaspididae) is a major pest of California citrus, with infestations causing growers significant financial losses. It has recently developed resistance to traditional insecticide sprays. An alternative suppression tactic is the release of a biological control agent, Aphytis melinus DeBach (Aphelinidae: Hymenoptera) that feeds on red scale. Although many aspects of the red scale Á/Aphytis interaction are now understood, it is difficult to differentiate the effects of temperature and population fluctuations in the field. To investigate such complex interactions, we propose a new stochastic modeling technique, based on event-driven competing risks, that incorporates details of life histories as well as the host Á/parasitoid interaction. Our continuoustime, individual-oriented modeling approach quantifies relationships among individuals and describes the resulting coupling between the interacting populations. The event-structured simulation drives time in contrast to the usual timedriven stochastic dynamic programming. Our system, developed in the public domain using the R statistical package, allows for different biological clocks, since both red scale and Aphytis development respond to temperature (degreedays) while searching female Aphytis follow a diurnal time schedule, contingent upon temperature-dependent egg maturation. #

Using discrete systems to exhaustively characterize the dynamics of an integrated ecosystem

“…The main originality of our work not only lies in the use of discrete models (e.g. Petri nets) in ecology, as some other studies started to explore the promising avenue of discrete models too (Baldan et al, ; Ewing et al, ; Gaucherel et al, , ). Rather, it concerns the discrete and qualitative conception of the integrated ecosystem forming, we think, the backbone of any ecosystem.…”

“…Petri nets are similar to Boolean networks (Kauffman, 1969;Thomas & Kaufman, 2001), as they are well-adapted to formalize network topological changes. To our knowledge, such discrete and qualitative models have rarely been used in ecology (but see Ewing, Yandell, Barbieri, Luck, & Forster, 2002;Cordier, Largouët, & Zhao, 2014;Baldan, Bocci, Brigolin, Cocco, & Simeoni, 2015;Gaucherel, Théro, Puiseux, & Bonhomme, 2017). Among discrete models, Petri nets are powerful tools that enable the rigorous formalization of exhaustive changes in an ecosystem's structure.…”

Quantifying Insect Predation: A Comparison of Three Methods for Estimating Daily per Capita Consumption of Two Aphidophagous Predators

“…While predatory Heteroptera are primarily noctural in their activity (Pfannenstiel andYeargan 2002, Van Laerhoven et al 2003), many other insect predators and parasitoids are primarily diurnal (Nakamura 1997, Ewing et al 2002, Pfannenstiel and Yeargan 2002, Hougardy and Mills 2007. Although we know of no studies on the diel activity patterns of the larvae of C. nigricornis and H. axyridis, we follow Dixon (2000) in assuming that aphidophagous predator larvae are diurnal.…”

A Method for Analyzing Survivability in the Context of a One-on-One Engagement

“…However, we also realize that a competing risk framework does not necessarily lead to the nice analytical solutions that Williams and Ancker obtained in their studies. Consequently, we ask ourselves what benefits do we accrue by employing competing risks in our study of stochastic duels, and we respond by highlighting the work of Ewing et al (2002) who modeled the complex interactions of a pest with a California citrus crop. In their model, Ewing et al assumed a hazard function that was reasonably well behaved over time (2002, p. 41); as a result, they described the occurrence of events in an ecosystem via this hazard function.…”