Root maggots (Delia spp., Diptera: Anthomyiidae) in prairie canola (Brassica napus L. and B. rapa L.): Spatial and temporal surveys of root damage and prediction of damage levels

The levels of infestation and damage to canola taproots caused by crucifer-feeding root maggots (Delia spp.)(Diptera: Anthomyiidae) were determined through surveys of commercial crops across the canola growing regions of western Canada. Canola root damage caused by Delia spp. maggots was measured at the end of the season by determining the percentage of plants infested (PPI) per field and estimating average damage levels (DL, on a scale of 0 to 5) to taproots. Over the course of the survey, 2890 canola fields were sampled. Infestation by root maggots occurred wherever canola is grown in western Canada. In the study, 96% of the fields surveyed in Manitoba and Saskatchewan and 99.8% of the fields in Alberta showed evidence of root maggot feeding, with PPI much higher than in provincial surveys of a decade previously. Geographical ecoregion significantly affected both PPI and DL. The greatest amount of damage over the largest area was found in western and northwestern Alberta, although localized areas with severely damaged roots occurred along the northern edge of the entire Parkland ecoregion. Over the entire survey area, Brassica rapa L. roots had greater PPI and DL than B. napus L. roots, although species differences were not significant in Saskatchewan or Manitoba. Data from the survey were correlated with weather variables to develop a model predicting the levels of damage inflicted by root maggots to canola roots. Of the weather variables investigated, temperature, especially the average temperature in the preceding July and August, had the greatest in fluence on PPI and DL. The warmer the temperatures of the preceding July and August, the lower the PPI and DL in the current year. Likewise, previous year’s July and August precipitation influenced PPI and DL negatively. Other weather parameters had little consistent influence on PPI and DI. The predictive model for PPI and DL, developed from the 1995–1998 survey data, incorporated canola species, ecoregion, previous July-August and September-October temperatures and precipitation. The validity of the model was assessed using a set of survey data collected in Alberta from 1981 to 1983. Regression equations of predicted results on actual results were significant. However, r 2 values for the regressions were low, in part because of the few ecoregions surveyed and the lack of canola species data recorded in the earlier survey, and because of the large increase in infestation levels since the earlier data wer