Abstract
The prevalence and incidence of Bean pod mottle virus (BPMV) have been reported to be on the increase in the United States but little is known about the temporal and spatial dynamics of this virus within soybean (Glycine max) fields. A quadrat-based sampling method was developed to quantify the within-field spread of BPMV in soybean in 2006 and 2007. Twenty-five 30-cm-long quadrats were established within each row of soybean in field plots consisting of six rows, each 7.6 m long and spaced 0.76 m apart. Four treatments were used to influence the temporal and spatial dynamics of BPMV epidemics. Treatments were: (i) establishment of a point source of BPMV inoculum within soybean plots; (ii) lambda-cyhalothrin insecticide applied at the V1 and R2 growth stages; (iii) establishment of a BPMV inoculum point source, plus the application of foliar insecticide sprays at the V1 and R2 growth stages; and (iv) a nontreated, noninoculated control. All quadrats (census) were sampled beginning 25 days after planting; sampling continued every 8 to 11 days until plants were senescent. Sap from leaf samples was extracted and tested for BPMV by enzyme-linked immunosorbent assay. The incidence of BPMV per treatment was plotted against time to produce BPMV incidence curves for temporal analyses. In addition, positions of BPMV-positive quadrats were mapped for spatial analyses. BPMV was detected within soybean plots on the first sampling date in 2006 (30 May) and on the second sampling date in 2007 (21 June). The rate of BPMV temporal spread within treatments ranged from 0.11 to 0.13 logits/day in 2006 and from 0.05 to 0.07 logits/day in 2007. Doubling times for BPMV incidence among treatments ranged from 5.4 to 6.4 days in 2006 and from 10.0 to 14.1 days in 2007. Soybean plots that had the earliest dates of BPMV detection within quadrats (x) also had the highest BPMV incidence (y) at the end of the growing season (R2 = 66.5 and 70.4% for 2006 and 2007, respectively). Spatial analyses using ordinary runs, black-white join-counts, and spatial autocorrelation revealed highly aggregated spatial patterns of BPMV-infected quadrats over time. Bean leaf beetle population densities were linearly related to BPMV incidence (P < 0.0001) in both years, indicating that BPMV epidemics were greatly influenced by bean leaf beetle population density. To our knowledge, this is the first study to quantify the seasonal temporal and spatial dynamics of BPMV spread within soybean.
