Bactrocera dorsalis (Diptera : Tephritidae) : dynamique des populations, parasitisme par Fopius arisanus (Hymenoptera : Braconidae) et distribution prédictive du parasitoïde en Afrique

Abstract

Bactrocera dorsalis (Hendel) is a major constraint to the development of the fruit and vegetable sector in Cameroon. Its principal parasitoid, Fopius arisanus (Sonan) has been introduced – from Hawaii, USA - for the biological control of this pest in Africa. To date, no scientific studies have been carried out on the ecology of the B. dorsalis-F. arisanus system in Cameroon. The present study, therefore, aims to fill this knowledge gap by providing basic data necessary for the development of an integrated pest management program of B. dorsalis in Cameroon. The study focused on five specific objecties: i) study the seasonal fluctuations of B. dorsalis populations in relation to host plant phenology; ii) determine the host plants and trophic preferences of B. dorsalis; iii) evaluate the influence of the host fruit species of B. dorsalis on parasitism and biological parameters of F. arisanus; iv) evaluate the influence of temperature on life history of F. arisanus; v) model the Afro-tropical distribution of F. arisanus under a climatic change scenario. A weekly monitoring of a trapping network as well as the phenology of the plants present in the orchards were carried out for four years (from 2013 to 2016) in the orchards located in Foumbot (zone III) and Nkolbisson (zone V). Fruits of cultivated and wild plants were sampled over the span of three years (2013-2015), in six and four locations of agroecological zones III and V of Cameroon. Laboratory choice and no-choice experiments were conducted to evaluate the parasitism of B. dorsalis eggs by F. arisanus in fruits of 11 host plant species. This parasitism was also evaluated at the level of the IITA mixed orchard at Nkolbisson. Biological parameters of the parasitoid and its host B. dorsalis were measured using adults emerging in no-choice experiments. Phenological modelling of F. arisanus was performed using complete life table data collected at constant temperatures ranging from 15 to 35°C. The Insect Life Cycle Modelling platform was utilized for model selection, simulations of the demographic parameters and mapping the Afro-tropical distribution of F. arisanus. Nine species of fruit flies were identified in the two targeted agro-ecological zones: Bactrocera dorsalis, Ceratitis cosyra, C. anonae, C. capitata, C. bremii, C. punctata, Dacus bivittatus, D. punctatifrons and Perilampsis sp. The dominance of B. dorsalis was high in both orchards (over 98% abundance). Total food-bait and parapheromone trap captures of B. dorsalis varied significantly among years in the Foumbot orchard, while in the Nkolbisson orchard no differences among the years were observed. Regardless of fruit fly species sex, these captures were more abundant during the wet season (June) in zone III and during the small rainy season (May) in zone V. The population peaks coincided with the presence of mangoes and guavas in the orchards, although it was identified on the fruits of 16 out of 21 host plant species sampled. In addition to B. dorsalis, three other species of Tephritidae emerged from the fruits. The infestation per fruit unit weight was higher on M. indica and Irvingia gabonensis which are the preferred hosts. The absolute parasitism of B. dorsalis eggs by F. arisanus was high in Psidium guajava, Irvingia gabonensis and Irvingia wombolu fruits in choice experiments, in P. guajava fruits in no-choice experiments, and in Erybotrya japonica fruits in the field. Of all the fruits tested in the laboratory, body and hind tibia lengths of both parasitoid sexes emerging from M. indica were longer than on the others. The longest pre-imaginal development time for both sexes was obtained on E. japonica and the shortest on Carica papaya. The parasitoid development time from egg to adult estimated by the probit model was 64.82 and 18.97 days at 15 and 30 °C; the minimum temperature threshold estimated by the linear regression was 7.62 and 7.73 °C for the egg-larva and pupa stages, respectively, with thermal constants of 181.382 and 222.922 degree-days; the maximum temperature threshold estimated by the Logan 1 submodel was 33.75°C for the egg-larva stage and 32.15°C for the pupae; the immature mortality was higher at temperatures 15 and 30oC. The maximum fertility obtained from the seconddegree polynomial function was 205 eggs / female at 22°C. The life table parameters simulated between 15 and 35oC were significantly affected by temperature; accordingly, the distribution predicted showed that the Central and Eastern Africa offer the best thermal conditions for the survival and establishment of the parasitoid under the current temperature conditions, with a predicted expansion of its limits to southern Africa in 2050. In Cameroon, the western highlands, humid forest areas and high guinea savannah zones are suitable for F. arisanus introduction. The presence of Bactrocera species, the quality and availability of host fruits and the appropriate climatic conditions are necessary for the survival and permanent establishment of F. arisanus in the new areas of introduction.