Risk analysis of Sclerotinia sclerotiorum for biological control of Cirsium arvense in pasture: Ascospore dispersal

Natural levels of Sclerotinia sclerotiorum ascsospores in the Canterbury re gion were determined over 3 years by trapping depositing ascospores in dish es containing a selective agar. Mean levels in 'horticulture', 'biocontrol- pasture', 'mixed cropping' and 'pasture' strata were 115, 56, 10 and 3 asco spores m(-2) day(-1), respectively. Ascospore deposition downwind of small experimental biocontrol sites was measured on 2 days in 1994 and 9 days in 1997 in late spring. Exponential depletion models scaled up to represent a 1 ha biocontrol site, revealed that dispersing ascospores declined to natur al levels at downwind distances of 2.5-7.9 m. These results imply that biol ogical weed control in pasture using S. sclerotiorum creates no greater ris k of crop disease than does horticulture, and that under the conditions of our experiments, an isolation distance of 8 m would have sufficed. However, such a safety zone may be inadequate under certain meteorological conditio ns not encountered in the experiments when ascospores may disperse in large r numbers over longer distances. To complete the information required to bu ild a mechanistic model of spore dispersal (beyond the scope of this paper) which would cope with a variety of meteorological conditions, two studies were conducted on the dynamics of apothecium formation and ascospore releas e. In a two-year study, apothecium formation was confined to the spring (Se ptember-November), and population size peaked in mid October. In a 5-day st udy, ascospore release occurred during the daytime, reaching a maximum late morning on frost-free days and a lower maximum mid afternoon on days with morning frost.