We demonstrate the use of matrix models to assess the impact of biological control and other pest management strategies with a case study of Carduus nutans (nodding thistle) in New Zealand. Modeling can facilitate the choice of the best control agents, improve evaluation of biocontrol attempts, and assist in the design of integrated pest management plans.
Field data for ∼8000 mapped C. nutans plants at two sites in New Zealand were used to develop size‐structured matrix models. The matrix models confirm that both populations of C. nutans were increasing in number, as is expected of a noxious weed in its invasion phase. Elasticity analysis indicated that seed/seedling and small‐plant/seed transitions were more crucial to population growth than rosette survival rates. However, simulations of attack by the biocontrol agent Rhinocyllus conicus (nodding thistle receptacle weevil) showed that seed losses of ∼69% would be required to make the populations decrease in size, far more than the observed losses of 3040% in New Zealand, the United States, and Canada. This result accords with field observations of control failure in New Zealand, but not with records of successful control in North America.
Other management possibilities, including the use of grazing management to suppress germination and an integrated weed management approach, were also assessed, using both deterministic and stochastic matrix models. This work indicates that, with continued suppression of germination in conjunction with a reduction of the input to the seed bank, it may be possible to control C. nutans in New Zealand.
Katriona Shea Dave Kelly. (1998). ESTIMATING BIOCONTROL AGENT IMPACT WITH MATRIX MODELS: CARDUUS NUTANS IN NEW ZEALAND. Ecological Society of America, 8(3)