Degree-day based phenological forecasting model of saddle gall midge (Haplodiplosis marginata) (Diptera: Cecidomyiidae) emergence

Outbreaks of saddle gall midge (Haplodiplosis marginata) affecting wheat and other cereals are difficult to anticipate and may not be identified until damage has occurred. Earlier work on this pest has shown that degree day models can be used to predict H. marginata emergence based on soil temperatures. Here, we. show how the availability of regular long-term trapping data can be used to update and improve upon this earlier model by predicting the progression of emergence. The emergence of adult H. marginata at three sites in the UK was monitored over two flight seasons using pheromone traps. The data confirmed the presence of multiple peaks in emergence over several weeks. Rainfall events followed by an accumulation of 512DD (+/- 9.11DD) above 0 degrees C could be used to predict peaks with greater accuracy than degree day accumulations alone. Cumulative percentage emergence as a function of degree day accumulations was best described by a probit model. The probit model predicted H. marginata emergence at other sites and years to within 4 days. Application of these models will enable growers to forecast peaks in emergence, make informed assessments of crop risk and time application of chemical controls appropriately and only where required. (C) 2017 Elsevier Ltd. All rights reserved.