DYNAMICS OF FRUIT FLIES POPULATIONS IN TWO MANGO PRODUCTION ZONES (KORHOGO AND FERKÉ) IN NORTHERN CÔTE D’IVOIRE

Fruit flies are a major constraint to the production and export of mangoes by causing the rot of mangoes. The objective of this work is to follow the fluctuation of population of flies in the northern part of Côte d'Ivoire between May 2015 and October 2016. A network of trapping flies using food baits (the Torula and Ceratrap) and sexual attractants (Methyl Eugenol and Trimedlure) were set up in Korhogo and Ferké mango orchards with the forest Diabaté Kamonon and the forest Petit Clerc as reference area. During trapping, 11 species of fruit flies were identified at all study sites. Among these species, Bactrocera dorsalis and Ceratitis cosyra, were the most abundant in the area. In general, diversity indices were low. The Ferké’s Forest has the highest Shannon index (H = 1.20). This index was higher than the index found in Ferke's orchards (0.9398). At Korhogo the Shannon index in orchards (H = 0.84) is higher than in the forest (H = 0.41). It has been observed that the genus Bactrocera swarmed on mango orchards when precipitation was plentiful. While the genus Cerartitis only appeared when precipitation was low. In addition, knowledge of the ecology and fluctuation of fruit flies, will allow to develop a site monitoring device through a trapping system with sexual and food attractants.


Introduction
The mango, third export fruit of Côte d'Ivoire, generates more than seven billion FCFA per year in Côte d'Ivoire [1]. This crop is threatened by the presence of Tephritidae, of which more than 4,000 species of this family are known throughout the world. They are among the most devastating crop pests in the world [2]. These different species have a wide distribution because they cover the tropical and subtropical zones as well as the temperate zones [3]. Their habitats range from humid forests to dry savannas [4]. Pregnant females sting plant tissues, deposit their eggs on fruit and vegetable species such as citrus fruits, apples, mangos, watermelons, zucchini, tomatoes [5]. Eggs are also observed on Asteraceae family Adventives such as grasses, woody plants and shrubs [6]. Classified as quarantine pests, fruit flies are a limiting factor for fresh fruit and vegetable exports from some countries. In Australia, fruit flies loss from an annual fruit production of $ 850 million is estimated at $ 100 million [6]. In southwestern Burkina Faso, the damage was higher, estimated at 64% thanks to the presence of Bactrocera dorsalis on mangoes [7]. In Ivory Coast, Bactrocera dorsalis is responsible for more than 69% of the damage on mango [8]. In addition to Bactrocera dorsalis, various other species are beginning to appear on mango orchards. The inventory of flies species and the monitoring of their fluctuation is necessary to optimize the fight against these pests.

Study Area
The study was conducted at the National Center for Agronomic Research (CNRA) fruit research station Lataha (Korhogo) (9 ° 34 'N and 5 ° 34' W) and at the CNRA of Ferké. (5 ° 55 -3 ° 30 N and 10 ° 30 -8 ° 35 W): The climate in these two zones is Sudanese with two seasons: a dry season, from November to April, and a rainy season, May to October. Annual precipitation over the last three decades varies between 800 and 1500 mm with an average temperature ranging between 22 ° C and 28 ° C. The vegetation consists mainly of savannas with trees and grass. The soils are either ferralitic types derived from granites or shales, or of tropical ferruginous type. The average annual rainfall is 1400 mm in a wet year and 1000 mm in the dry year. The natural vegetation consists of savannah trees. The soils are of ferralitic type. Populations of fruit flies were harvested from the orchards of these two stations with natural savanna as the reference area. The plots sampled are as follows: At the two research stations: mango orchards that are full of varieties. The cultivated varieties are "Kent", "Keitt", "Gallery" and "Amelie" with a density of 100 feet per hectare. The reference medium consists of two savannas: In Korhogo, the natural forest of "DIABATE Kamonon" (DK), located to Lataha, has been sampled. (09 ° 26.122 ' W and 005 ° 08.323 'N) [9]. In Ferké the natural forest of "Petit Clerc" has been sampled (10 ° 26.122 W and 005 ° 08.323' N [10].

Characteristic of the Traps
Three types of traps were used for flies harvesting. These are "TephriTrap" type traps, "Mc-PhailTrap" type traps and local traps. The operating principle of these different traps is as follows: Traps of the "Tephri Trap" type are yellow boxes with perforations. The yellow color is recognized as the most attractive for flies. In these traps, closed with a lid, are introduced into the nucelle located under the lid of the trap, parapheromones called "sexual attractants" (Methyl-Eugenol (ME) or Trimedlure (TRI)). These insecticide-associated substances have a similar odor to female flies and specifically attract adult male flies. Sexual attractants, provided in capsule form, have been placed.
"Mc-PhailTrap" type traps are also yellow in color. Three granules of a high-protein liquid food attractant (the "Torula Yeast") were milled and dissolved in 300 ml of water. To this solution was added 3 g of Borax. Borax has a conservative power and prevents the decomposition of flies captured in the trap [11]. The resulting mixture, combined with an insecticide, was put in each trap. These traps attract male and female flies that come to feed.
Local traps are made locally with plastic bottles. 500 ml of the Ceratrap solution (protein-rich food bait), combined with an insecticide, were taken and placed in each trap. These perforated traps attract both male and female flies.
The insecticide used to kill flies in all these traps was DDVP or Dichlorvos (2,2-dichlorvosvinyl dimethyl phosphate)

Trapping Device for Flies
In each mango orchard, just as in the forest zone, the different traps were arranged according to a randomization on three lines. On each line four traps were placed (2 were content sexual attractants and 2 others were food attractants). Each line represented a repetition. Each trap was then hooked to a branch of the host plant at a height of about 1.5 m above the ground [12] away from direct sunlight. The traps were arranged in a square figure or parallelepiped [11]. The branch supporting the trap was covered with solid fat as a barrier to prevent the predatory activity of the ants on adults trapped flies in traps. A total of 12 traps per site were installed. The optimum distance between two lines was about 40 m to avoid interferences between attractants [13]. The distance between treatments on the same line was 30 m (every 3 trees for orchards).
Each week, flies captured by the traps were collected and a systematic rotation of the traps was performed. Captured flies were counted by trap and then kept in vials containing 70% alcohol. Precipitation was recorded at each site during the entire trapping period.

Identification of Flies
The flies captured in the different traps were identified to the species using identification keys [14,15,16] under a binocular microscope.

Data Analysis
Statistical analysis of the data Ecological indices have been chosen to describe Tephritidae populations. For each medium, the species richness (S), which is the total number of species sampled, was calculated. Then, thanks to the EstimateS software (version 7.0) [17], the Shannon index (H ') and the equitability index (E) were calculated for each area. The abundance of flies (the number of individuals captured per site) was determined and related to rainfall.

Diversity of Tephritidae Caught in Traps
During trapping, 11 species of fruit flies grouped into genera were identified at all study sites (Table I) (Table II).

In the Korhogo Area
In the mango orchard, B. dorsalis was the most species captured during the sampling period. The number of other flies, of which C. cosyra was the largest, varied in the same order as that of B. dorsalis, but with very low numbers (Fig. 1).

In the Ferké Area
In mango orchard of this area, the capture of B. dorsalis was important from May to July of the two years. The peaks of fly catches were located in June with 19667 and 6 436 individuals harvested respectively in 2015 and 2016. During these months we also observed a precipitation in the area. The other flies, including C. cosyra, appeared at the same time, but the numbers are very small. From September to April, total numbers of these species remained relatively low and constant ( Figure 1).
In abundance of other flies, including C. cosyra, evolved in the same direction as that of B. dorsalis. The peaks of these flies were in January 2016 and in August of the two years (Fi. 1).

Discussions
A total of 11 species of fruit flies grouped into 4 genera were captured by the traps installed. This high diversity was linked to the fact that the study was conducted in two mango production areas, in orchards and in the forest. Most of these species have already been reported in Côte d'Ivoire by the work of [19] but also in other parts of Africa [20,21]. Of these species of fruit flies harvested, Bactrocera dorsalis appears to be the most important species. This result has also been confirmed by several authors [22,23,13]. Native to Asia, Bactrocera dorsalis has spread around the world due to transcontinental trade in fresh vegetables [20]. It would have adapted to the humid and warm climatic conditions so that it developed properly, thus diverting the initial predominance of C. cosyra in most parts of West Africa and particularly in Côte d'Ivoire [24]. The abundance of Bactrocera dorsalis observed from May to October in relation to Ceratitis cosyra coincided with the fructification period of the host plants, particularly the mangoes that flies like. The heavy precipitation recorded during these months, accompanied by increase of the relative humidity of the environment, would also explain the abundance of this species. Our work has shown that in all four sites, Bactrocera genus swarms in mango orchards when abundant rainfall while Cerartitis genus appears only when precipitation is low. This observation has been made by several authors [13,25,19]. The low proportion of Ceratitis cosyra compared to Bactrocera dorsalis may be explained by inter-lartial food competition and the aggressiveness of adult of B. dorsalis to other species. This situation would cause the displacement of many species to other host plants in the natural environment [26]. The important proportion of fruit flies present in traps in natural forest could be explained by the fact that at the end of the mango season the flies species take refuge in its area.

Conclusions and Recommendations
Fruit flies influence the production of mangoes in the northern region of Côte d'Ivoire (Korhogo and Ferké). During trapping, 11 species of fruit flies were identified at all study sites. Of these species, Bactrocera dorsalis and Ceratitis cosyra were the most abundant in the area. In general, diversity indices were low. The Ferké's Forest recorded the highest Shannon index (H = 1.20). This index was higher than the index found in Ferke's orchards (0.9398). In Korhogo, the Shannon index in the orchards (H = 0.8429) was higher than in the forest (H = 0.411). It has been observed that the genus Bactrocera swarms on mango orchards when precipitation was plentiful. While the genus Cerartitis only appears when precipitation was low. Knowledge of the ecology and fluctuation of fruit flies, will develop a site monitoring device through a trapping system with sexual and food attractants. This monitoring device will provide alerts to trigger the fight against fruit flies.