Effect of Sample Placement on the Geostatistical Analysis of the Spatial Distribution of Lygus hesperus (Heteroptera: Miridae) in Lentils

Abstract

The effect of sample placement on the summary statistics and geostatistical analysis of the spatial distribution of Lygus hesperus Knight in lentils, Lens culinaris Medik., was evaluated for 10 stratified and three random sampling patterns for two adult and two nymphal distributions. Geostatistics uses spatial variation to determine the degree of association and dependence of spatially related data. All subsampling patterns accurately estimated the mean of these distributions. The hexagonal sampling pattern provided the best estimates of the spatial structure, whereas the random sampling patterns were least effective at estimating spatial structure. The smaller the number of samples, the lower the accuracy and reliability of the spatial structure. The minimum number of samples for these data would be between 35 and 100 samples, depending on the type of spatial structure and localized discontinuity. The higher the localized discontinuity and proportion of randomness, the greater the number of samples required to estimate the spatial structure.