MODELLING THE FLOWER INFECTION STAGE OF STRAWBERRY GREY MOULD (BOTRYTIS CINEREA)
DC HARRIS and X XU
HRI-East Mailing, West Malling, Kent, ME19 6BJ, UK
Background and objectives
Grey mould, caused by Botrytis cinerea is the most important airborne disease of strawberry in the UK and in many other parts of the world, reducing both fruit yield and quality. Current control methods rely heavily on the use of fungicides and yet may be ineffective if conditions are very conducive to disease. There is a need to explore new approaches to disease control which are effective but rely minimally on fungicide use. The control of grey mould currently depends mainly on a prescriptive programme of protective sprays during flowering, since it is during this time that most of the infections leading to fruit rot occur. These sprays are applied regardless of whether conditions are favourable to infection. The influence of weather components and inoculum on flower infection in the field has been studied quantitatively during 1995, 1996 and 1997 in order to develop a mathematical model from which infection risk can be predicted. It is hoped that this will lead to disease management in which growers are able to use fungicides only when they are needed and to their best effect.
Materials and methods
Data on infection incidence, on spore numbers over the crop and for several weather components were obtained for successive 2-day intervals during flowering of the June-bearing cultivar Elsanta. Successive 'generations' of flowers were labelled at the 'white-bud' stage, harvested after 48 ;h, surface-sterilized and incubated for 1 ;week in sterile dishes under moist conditions to allow any infection by B. ;cinerea to develop. Daily spore counts were obtained using a Burkard volumetric trap. Air temperature, rainfall, relative humidity and surface wetness were all recorded automatically by instruments linked to a datalogger. Models for flower infection by B. ;cinerea were developed from the data.
Results and conclusions
A model based on two key weather variables, daytime vapour pressure deficit and night-time temperature gave very good correlation (0.8) between fitted and observed values. The model was not significantly improved by the inclusion of spore data, even though inoculum levels varied markedly in the three years. The apparent lack of importance of conidial numbers over the crop in determining infection incidence may be because the numbers were never below a level at which they became limiting or because the two key weather variables in the model are also important in determining spore numbers. Further work is in progress on validating the model in relation to June-bearing and over-bearing strawberry cultivars.