Résumé
Fruit flies (Diptera: Tephritidae) are considered major pests of mango (Mangifera indica) worldwide. Their management, with an increasing focus on replacing synthetic pesticides with alternative eco-friendly methods, is a challenging issue. A new fruit fly species, the oriental fruit fly, Bactrocera dorsalis (Hendel), recently invaded Reunion Island, further increasing fruit fly incidence on mango crops, which were so far mainly infested by the peach fruit fly Bactrocera zonata (Saunders). Field studies have shown that fruit fly damages can be reduced by adjusting the harvest date, as fruit fly infestation increases with fruit maturity. At the same time, the harvest date is an important factor in determining the shelf life of stored fruit and the quality of fully ripened fruits ready for consumption. For example, harvesting mangoes too early can reduce their quality. A modelling approach was developed to optimize harvest timing with the dual objectives of improving fruit quality and reducing fruit fly damage. To do this, a fruit fly-mango modelling framework was developed by coupling a model that predicts the probability of a fruit being infested by fruit flies as a function of fruit maturity, with ecophysiological and empirical models that predict the growth and maturity of fruit on the tree and the quality (Brix) of stored fruit when ripe. Using the model, we virtually investigated the effect of harvest date and frequency, which alter fruit maturity at harvest, on fruit fresh mass, fruit quality and fruit infestation by fruit fly. The simulation results show that by harvesting the fruit at an earlier stage of maturity, such as the green mature stage, it is possible to greatly reduce the risk of mango infestation while ensuring that fruit fresh mass and quality are only minimally affected.