Introduction: Fruit flies (Diptera: Tephritidae) pose a severe threat to cucurbit crops worldwide, with potential losses ranging from 30% to 100%. This study evaluated the effectiveness of agroecological and conventional protection practices, compared to untreated controls, in managing fruit fly infestations across different seasons in Morogoro, East-Central Tanzania.

Methodology: The study was conducted from March 2021 to September 2022. Field experiments employed a Randomized Complete Block Design with five replications tested three practices: agroecological crop protection, conventional pest management using synthetic pesticides, and untreated control. Each 45m x 45m plot contained watermelon, squash, and cucumber. Data on fruit blemishes, infestation rates, and yield (kg/ha) were analyzed using R software.

Results: Percentage of blemished fruits and infestation rates by Zeugodacus cucurbitae (Coquillet), Dacus ciliatus Loew, and Dacus vertebratus Bezzi were significantly affected by the interaction of management practices, seasons, and crop species, with the lowest infestation rates in conventional plots, followed by agroecological plots. Yield was significantly influenced by the interaction of season, crops, and management practice. Untreated control plots produced significantly less yield than those under agroecological and conventional management.

Conclusion: Agroecological practices effectively reduced fruit fly populations and blemished fruits, enhancing yield over consecutive seasons. These practices are comparable to conventional methods in mitigating fruit fly infestations in cucurbit crops.

Experimental setup and data collection

1. Experimental setup

Data were collected from 30 experimental plots established in two agroecological zones of the Morogoro region, namely mountainous and plateau. These zones were chosen because they are known to have a high prevalence of cucurbit crops, which were the focus of the current research, as well as the presence of cucurbit flies that have been reported in the area (Geurts et al. 2012; Mwatawala et al. 2006, 2009)

The experimental plots were set up in a factorial design using a randomized complete block design (RCBD) with five replications for each agroecological zone as study areas and the three farming practices as treatments. Experiments were conducted for four seasons: March to June 2021 (First season), July to September 2021 (Second season), March to June 2022 (Third season) and July to September 2022 (Fourth season).

Crop protection practices tested were;

ACP practices (formulated by GAMOUR): Attracting pollinating insects and trapping cucurbit infesting fruit flies using border crop (maize in our case); Mass trapping of flies and spot application of baits on border crop, in this case GF120 (Spinosad 0.24g/l) and Bio lure; Mulching using organic materials; Field sanitation by collecting damaged fruits and disposing them in an augmentarium and application of organic fertilizers (farmyard manure). GF120 was sprayed on the border crop at the intervals of seven days from when the cucurbits started to flower until seven days before harvest. We mixed GF120 and water at a dilution ratio of 1 L GF120:7 L water. 0.016 L of a mixture was sprayed on the upper side of maize leaves on a 0.4 m² spot and after every 10 m. Biolure was also placed on the border crop and changed weekly until seven days before harvesting.

Conventional practices: Included application of an insecticide Amadine^® 40EC (Dimethoate 400g/l, Shandong Binnong Technology Co. Ltd, China), at an application rate of 2.5 L/Ha; fungicide Daconil^® 720SC (Chlorothalonil 720g/l, Syngenta Crop Protection Ag, Switzerland), at an application rate 2.0 L/Ha; Industrial fertilizers (N: P: K 15:9:20 and CAN 15:4:26, basal and top dressing fertilizers, respectively). A tank mix of fungicide and insecticide was sprayed six times per season at the interval of 14 days from when 75% of the crops had emerged up to 14 days before harvest. In a 20 L sprayer tank, 0.125 L and 0.05 L of Chlorothalonil and Dimethoate, respectively, were mixed in a 20 L of water and sprayed on the cucurbits.) Untreated control plots: Plots without management practices except irrigation during the dry season and weeding were used as controls. Resulting in a total of 30 experimental plots grouped in 10 replication sites (i.e. five replication sites for each of the two agroecological zones, three plots per site, one for each treatment). Each replication site consisted of three plots, one for each treatment, and within each plot, three types of cucurbit crops were grown: watermelon, cucumber, and squash. The size of each plot was 45 meters by 45 meters, equal to 2025 square meters, and a distance of 100 meters separated the plots within replication sites.

2. Data collection

Number of blemished fruits by cucurbit flies

Data on the number of blemished fruits were collected randomly from 45 plants tagged for each cucurbit crop in each plot. The percentage of blemished fruits was recorded cumulatively from the fruit set through the harvesting stages. Fruits were sampled after they were set. The collected data were then transformed into a percentage of blemished fruits following methods described by Ghule et al. (2013) and Mwatawala et al. (2015).

Infestation rates

Fruits were randomly sampled from each plot per crop. The collected fruits were taken to the Sokoine University of Agriculture rearing facility, weighed, and placed in rearing containers. The fruits were reared using plastic containers with dimensions of 25 cm by 15 cm by 11 cm. Sampling and rearing methods were adopted from Mwatawala et al. (2009). Fruit flies that emerged from each reared fruit were identified and recorded per plot per crop. Fruit flies were identified using keys developed by Virgilio et al. (2014). Infestation rates were calculated as the number of emerged fruit flies per unit weight of fruits.

Yield

The yield of cucurbit fruits per crop species was determined by harvesting undamaged fruits from 45 tagged plants for every crop in each plot within an area of 67.5 square meters. The weight of the fruits (yield) was then calculated and converted into kilograms per hectare (Thuc and Minh 2022).