Insect exclusion fencing is not an effective option for the management of swede midge in organic cole crop production

Braden G. Evans & Dr. Rebecca H. Hallett, School of Environmental Science, University of Guelph

The swede midge, Contarinia nasturtii, is a small, inconspicuous brown fly from the family Cecidomyiidae, the ‘gall forming’ midges (Figure 1).  It is an invasive insect from Eurasia which has become established in North America, expanding its range across Canada and the United States since it was first recorded here in Ontario in 2001.  This pest insect attacks economically important cole crops in the Brassica family, including broccoli, cauliflower, cabbage, and canola, among many others. The gregarious larvae live and feed among the compressed leaves surrounding the developing vegetable head, leading to direct damage to the marketable portion of the host plant.

Management programs in conventional systems rely on insecticides, pheromone-based action thresholds, crop rotation and management of cruciferous weeds to control local swede midge populations.  While this approach produces acceptable yields for conventional growers, organic producers do not have access to the most effective synthetic insecticides for swede midge pest management.  Damage levels due to swede midge are variable from year to year, but severe infestations in recent years have resulted in losses of up to 100% in organic fields, causing some producers to abandon broccoli production temporarily.

Our research project, funded by the OMAF & MRA New Directions program, is aimed at developing organically-acceptable swede midge management practices for cole crop producers.  As a potential physical control tactic, we have evaluated insect fencing barriers over the past three field seasons.  Insect exclusion fencing can be an effective pest management tool by blocking the dispersal of insects into cultivated areas.  These physical barriers have shown some effectiveness in cultivated brassicas, such as reducing cabbage root maggot (Delia radicum) damage in rutabaga fields, and appeared to hold some potential against a weak-flying insect, such as the swede midge.

Field trials consisted of enclosing 5 meter rows of broccoli with 5-foot tall, 1-mm black ‘no-see-um’ nylon mesh panels supported by aluminum posts anchored to the ground with rebar.  The enclosed fenced plots harbored approximately 45 broccoli plants (Figure 2).

Treatments for comparison consisted of 5 replicates of (A) fenced plots, (B) fenced plots with a soil nematode application and (C) unfenced control plots.  Damage levels to the developing broccoli heads were recorded weekly and assigned a rating from 0-3, based on the severity of damage.

At the end of the season, despite a lower average damage level in fenced plots, each of the fenced and unfenced treatments exhibited damage that exceeded the threshold for marketability, which is defined as a rating of 1 or less.  The fencing appeared to delay the onset of elevated damage symptoms, resulting in lower damage ratings overall, but did not effectively block swede midge dispersal (Figure 3).

Swede midge may have entered the fenced plots through occasional tears in the material or windblown gaps arising between adjacent panels. A sticky trap survey conducted within fenced plots in 2013 found swede midge on cards at 6 and 8-feet above the soil, suggesting that adults were able to fly over the upper boundary of the fence.

An unexpected consequence of the fencing treatments, observed in 2011, was a sudden, localized increase of Imported Cabbage Worm (ICW) larvae within fenced plots.  High numbers of ICW pupae were found within the folds of fabric in the corners of the fenced plots (Figure 4).

Subsequently, very high numbers of ICW larvae in these plots caused nearly complete defoliation of the fenced broccoli.  The unfenced, control plots had only minor feeding damage by ICW.  Our comparisons of insect diversity in the fenced and unfenced plots suggest that, in addition to the favorable habitat for pupation offered by the fencing, the fencing may have led to reduced numbers of beneficial parasitic wasps within plots, and thus favoured ICW populations.

Our results indicate that insect exclusion fencing is not a sufficiently effective physical barrier to swede midge dispersal.   Although fencing has shown some promise against other species, such as the cabbage root maggot, our results indicate that fencing is not a suitable approach for management of the swede midge.