Developing tools for monitoring and forecasting of onion fly Delia antiqua in Norway

Abstract: The onion fly Delia antiqua (Diptera: Anthomyiidae) is a major pest in onion growing areas in Norway. Currently used methods for monitoring and forecasting attack seem to work un-satisfactorily, causing reduced effects of control measures. In this study, we intended to develop a selective monitoring trap by combining colour and odour sensory cues. White sticky traps and blue sticky- and water traps, with and without odour, were tested in a field experiment in Vestfold, Norway in 2016, along with fly registrations in commercial onion fields in Vestfold, Rogaland and Nord-Trøndelag in 2015-2017. Onion fly pupae were also collected from these fields in autumn 2016 to study the phenology of D. antiqua in laboratory emergence tests. The resulting data were used together with climate data to test SWAT for simulation of onion fly population dynamics in Norway.Results showed that the blue sticky traps with odour were the most selective and effective. However, as the onion fly occurrence was generally very low in 2016, the results need verification from additional field experiments. High amounts of non-target catches on the sticky traps (white > blue) showed that catches on sticky traps were very time and labour-intensive compared to the water traps. In addition, several fly individuals with damaged identifying features were found on sticky traps, making a reliable morphological identification difficult. Molecular methods might be a possibility tool to achieve a more efficient identification. Emergence of one population of D. antiqua pupae at two temperatures indicated a theoretical base temperature of 3.9 °C for post-diapause development. From these data, the day-degree requirement for 50% emergence would equal about 255 day-degrees (d°, Celsius). Studies of more populations at more than two temperatures are needed to obtain reliable results. Nevertheless, modelling with our Norwegian data in SWAT showed a very close relationship between forecasts and the actual first fly and first peak occurrence of D. antiqua in the field.With data from more Norwegian D. antiqua populations and emergence data from a wider range of temperatures, we might be able to improve the model adaption in SWAT further. Additionally, after refining the monitoring system, a decision support tool for Norwegian onion producers may be established.