Modelling a two-dimensional spatial distribution of mycotoxin concentration in bulk commodities to design effective and efficient sample selection strategies

Mycotoxins in agricultural commodities are a hazard to human and animal health. Their heterogeneous spatial distribution in bulk storage or transport makes it particularly difficult to design effective and efficient sampling plans. There has been considerable emphasis on identifying the different sources of uncertainty associated with mycotoxin concentration estimations, but much less on identifying the effect of the spatial location of the sampling points. This study used a two-dimensional statistical modelling approach to produce detailed information on appropriate sampling strategies for surveillance of mycotoxins in raw food commodities. The emphasis was on deoxynivalenol (DON) and ochratoxin A (OTA) in large lots of grain in storage or bulk transport. The aim was to simulate a range of plausible distributions of mycotoxins in grain from a set of parameters characterising the distributions. For this purpose, a model was developed to generate data sets which were repeatedly sampled to investigate the effect that sampling strategy and the number of incremental samples has on determining the statistical properties of mycotoxin concentration. Results showed that, for most sample sizes, a regular grid proved to be more consistent and accurate in the estimation of the mean concentration of DON, which suggests that regular sampling strategies should be preferred to random sampling, where possible. For both strategies, the accuracy of the estimation of the mean concentration increased significantly up to sample sizes of 40–60 (depending on the simulation). The effect of sample size was small when it exceeded 60 points, which suggests that the maximum sample size required is of this order. Similar conclusions about the sample size apply to OTA, although the difference between regular and random sampling was small and probably negligible for most sample sizes.