A stochastic simulation model for evaluating the concept of patch spraying

Abstract

The long‐term economic benefits of `patch' spraying are likely to be related to the initial spatial distribution of the target weeds, the demographic characteristics of the species and the weed control and crop husbandry practices to which they are subjected. This paper describes a stochastic simulation model developed to investigate the interaction between weed seed dispersal and patch spraying. Simulated weed plant and seed populations are generated and compared with data from field observations. Lloyd's Patchiness index is used to quantify the patchiness of the weed density distribution, and the parameter k of the negative binomial distribution is used as a measure of distribution shape. A method of assessing the spatial scale of weed aggregation is proposed, in which spatial weed density information is transformed into the frequency domain, using a discrete two‐dimensional Fourier transform. In this paper, we simulate `on/off' patch spraying (full or zero herbicide application rate). A quantitative analysis of the effects of sprayer resolution and weed seed dispersal range on the herbicide reduction and yield benefits from patch spraying is performed for three initial spatial seedbank distributions. The model is parameterized for the grass weed Alopecurus myosuroides Huds. Herbicide is applied in square areas (whose size is defined by the spatial resolution of the sprayer) in which mean weed density is greater than or equal to one plant m–2. For a system conforming to this specification we show that for the control of A. myosuroides, it is unlikely that patch spraying would be profitable in the long term if the control area is larger than 6 m × 6 m. In some circumstances higher resolution may be required.