Non-linear time-series analysis of the interaction between the citrus whitefly and the whitefly-specialist ladybird

A comprehensive understanding of the top-down effects of natural enemies on agricultural pests is essential for achieving effective biological control in integrated pest management. However, it is typically difficult to identify causal effects between the interacting species from time-series data, which have often been monitored for pest forecasting purposes in agricultural ecosystems, as it is likely to involve non-linear (state-dependent) population dynamics. In this study, we applied a recently developed framework of non-linear time-series analysis (empirical dynamic modelling) to determine top-down and bottom-up effects between the citrus whitefly Dialeurodes citri and the whitefly-specialist ladybird Serangium japonicum. We used weekly monitoring data for the two species collected over 4 years in pesticide-free citrus groves located in Shizuoka Prefecture, central Japan. Although we were able to identify time-delayed positive effects of D. citri abundance on S. japonicum abundance, we failed to detect any significant causal effects of S. japonicum abundance on D. citri abundance. Moreover, weather variables (temperature and rainfall) were found to have only a negligible effect on the population dynamics of the two species. Our findings indicate that bottom-up rather than top-down effects predominate in the weekly dynamics of this predator–prey system. On the basis of these observations, we discuss whether S. japonicum would be an effective agent for the biological control of D. citri in open-field systems.