Spatial and non-spatial analyses of insect counts in bulk-stored barley

Abstract

Stored grain insect species in bulk-stored barley were sampled during 15 consecutive weeks in two ways: direct sampling based on grain trier samples and indirect sampling based on probe trap captures. A total number of 22 insect taxa were found; this study focused on the six most abundant species and their natural enemies. Four aspects were addressed: (i) differences in insect counts when based on either probe trap captures or grain trier samples, (ii) usefulness of grain temperature and moisture content as explanatory variables for insect densities, (iii) density-dependent relationships between natural enemies and their hosts, and (iv) spatial and non-spatial analyses of insect counts. Both total captures and frequencies of insect taxa were consistently higher in captures with probe traps than insect numbers obtained from grain samples. Regression analysis with abiotic conditions and probe trap captures as explanatory variables provided good fits to insect counts in grain samples for four of the six insect species (R2-values>0.40). Using multi-regression analyses, we showed that: (i) the occurrence of natural enemies was only weakly associated with the abundance of hosts; (ii) grain moisture content and temperature appeared to be at least as important variables as host availability; and (iii) the predictive strengths of regression models were similar when based on either data from grain samples or probe traps. Spatial analyses (SADIE) of both sampling data suggested that all data sets followed a spatially random distribution; re-arrangement of the data provided insight into important aspects of SADIE analyses of small data sets. Non-spatial analysis (Lloyd's aggregation index) showed significant differences among species and that the level of non-spatial aggregation was quite sensitive to sampling method used.