Validation of degree-day models for predicting the emergence of two fruit flies (Diptera: Tephritidae) in northeast Egypt

We estimated thermal developmental thresholds (T-0) and degree-day (DD) constants for the immature stages of two tephritid pests, Bactrocera zonata (Saunders) and Ceratitis capitata (Weidenmann). Males of both species were trapped in an Egyptian guava orchard during the fruiting seasons of 2016 and 2017 and trap catches were compared with peak flights predicted by the DD model based on local weather data. Ceratitis capitata had faster development than B. zonata at 20 and 25 degrees C, but their overall developmental rate was similar at 30 and 35 degrees C. The thermal threshold of development (T-0) of B. zonata was higher than that of C. capitata, indicating greater sensitivity to cold. Although 35 degrees C yielded the fastest development of both species, survival was higher at 30 degrees C, with B. zonata experiencing a slight advantage, suggesting better tropical adaptation. Immature development of B. zonata and C. capitata was estimated to require 338 and 373 d, respectively, and 616 and 424 DD for a complete generation. Trap catches over both seasons showed good correspondence to peaks of fly activity predicted by the DD models; deviations from expectation ranged from 0 to 7 d for both fly species. Both species had four overlapping generations per season, with B. zonata abundance peaking in the first generation in both years, but only in 2016 for C. capitata. The models predict about eight and 12 generations per year in northeast Egypt for B. zonata and C. capitata, respectively. These models should be useful for timing pest control measures to coincide with periods of peak fly activity in fruit orchards.

Automated summary

The study estimated the thermal developmental thresholds and degree-day constants for immature stages of two tephritid pests, showing that Bactrocera zonata is more sensitive to cold while Ceratitis capitata develops faster. Trap catches corresponded well with peak fly activity predicted by DD models.