Biological control of invasive plant species: A stochastic analysis

Abstract

Biological control agents are regarded as a relatively safe method to control weeds. However, their impact on weeds can be relatively low and unpredictable. The aims of this article were to: (i) assess whether or not a weevil (Apion onopordi) and a mycoherbicide (Sclerotinia sclerotiorum) are desirable as biological agents for the control of Californian thistle (Cirsium arvense) in New Zealand despite their uncertain effectiveness; (ii) identify the combination of control options that is optimal to control the thistle; (iii) analyze the economic consequences of excluding chemicals from the weed control strategy; and (iv) assess the feasibility of the eradication of this weed. Two optimization models were developed and compared: one deterministic model and one stochastic model. The results showed that taking into account the stochastic effectiveness of biological agents can change the optimal integrated strategy, particularly if the biological control agent is relatively expensive. However, for a cheaper biological agent, the stochastic efficacy is less likely to change the optimal control strategy. On the basis of the modeling results, the authors argue that, in the context of the agri‐environmental setting of this article's case study, chemicals can be replaced by more environmentally friendly control options at a relatively low cost. The authors also show that the eradication of the thistle is unlikely, at least given the efficacy of the existing control methods.