Population dynamics of Aspergillus flavus in the air of an intensively cultivated region of south-west Arizona

Airborne propagules of Aspergillus flavus were quantified to investigate population dynamics of A. flavus in a region of south-west Arizona prone to epidemics of aflatoxin contamination of cottonseed. Air was sampled continuously from May 1997 to March 1999 at two sites using Burkard cyclone samplers. One sampler was initially at the centre of 65 ha of cotton treated with an atoxigenic strain of A. flavus to manage aflatoxin contamination of cottonseed. The second sampler was 0·6 km from the treated field. Total fungal colony-forming units (CFU) sampled ranged from 17 to 667 and from 9 to 1277 m at the untreated and treated sites, respectively. Counts of A. flavus ranged from 0 to 406 m and from 0 to 416 per m at the untreated and treated sites, respectively. Aspergillus flavus comprised 1–46 and 1–51% of the total cultured fungi at the treated and untreated sites, respectively. Peaks in total fungal and A. flavus CFU coincided with boll maturation and cotton harvest (days 251–321). Autoregression analysis suggested that there was no difference in total fungal CFU between treated and untreated sites, but the analysis showed that the quantity of A. flavus decreased at the treated site. This is probably caused by changes in cropping making the conditions less conducive to growth and reproduction of A. flavus in the surrounding fields. The incidence of the S strain of A. flavus was highest between May and August. The L strain accounted for up to 100% of the A. flavus sampled in the other months, and autoregression analysis showed that the L strain accounted for a greater overall proportion of the A. flavus population at the treated site compared with the untreated site. Autoregression analysis also showed the vegetative compatibility group of the applied strain was a greater proportion of L-strain A. flavus at the treated site (5–75%) than at the untreated site (0–65%), although this decreased with time. The quantity of A. flavus sampled at both treated and untreated sites was correlated with air and soil temperature. Large quantities of A. flavus occurred in the soil (up to 34 474 CFU g) of cotton fields and on cotton plant parts and debris (up to 272 461 CFU g) adjacent to the cyclone samplers. Aspergillus flavus is a major constituent of the airborne mycoflora associated with cotton fields in south-west Arizona when temperature is conducive to fungal growth. Although application of atoxigenic A. flavus altered the proportion of A. flavus strains and vegetative compatibility groups in the aerial mycoflora, the total quantity of A. flavus remained similar to that in untreated fields. Dispersal of A. flavus between fields suggests that atoxigenic fungi will be most effective in area-wide management programmes.