A simulation-based method to compare the pest suppression potential of predators: A case study with spiders

• Autor(a) principal: Benhadi-Marín, Jacinto
• Data de Publicação: 2018
• Outros Autores: Pereira, J.A., Barreales, David, Sousa, José Paulo, Santos, Sónia A.P.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10198/16919
• topic: Biological control; Ceratitis capitata; Functional response; Prey; R library; Spiders

Assessing and comparing the pest killing capacity of predators is a crucial but laborious task during the implementation of sustainable farming systems. Critical attributes of assessment include quantifying predator’s attack rate (a) and handling time (Th). The maximum attack rate (T/Th) (i.e. the maximum number of prey that can be attacked by a predator during the time interval (T) considered) could be a more precise and interpretable indicator of the potential suppression of pests exerted by a predator; however, its calculation only provides a point estimator usually derived from incomplete datasets (e.g. unbalanced or low replicated experimental designs) that could lead to draw wrong conclusions. We introduce simaR (simulation of maximum attack rates using R), an R library that generates 95% confidence intervals around estimates of the maximum attack rate that can be easily and intuitively used to compare across species. We validated the simulation method and used the empirical results of a controlled laboratory experiment to compare the maximum attack rates of spiders across a range of Medfly prey densities and illustrate how to use simaR with non-replicated partial data. Applying our method we found a significant effect of temperature on the maximum attack rate of two different guilds of spiders, the orb-weaver A. cucurbitina and the ambusher S. globosum that was not relevant regarding their attack rate and handling time. Our method compares different predator species and/or experimental conditions in a simple and reproducible procedure through an accurate, easy-to-use, fast and statistically robust analysis, based on simulation and bootstrapping, that can be used to assess the pest suppression potential of predators by simulating their functional responses from low-effort laboratory trials.