Meers: Improving crop risk assessment tools for bertha armyworms
Date: March 2016
Term: 3 years
Status: Completed
Researcher(s): Scott Meers, Alberta Agriculture and Forestry, Brooks AB; Owen Olfert, Agriculture and Agri-Food Canada, Saskatoon SK
SaskCanola Investment: n/a
Total Project Cost: n/a
Funding Partners: Alberta Crop Industry Development Fund
Project Summary
Although outbreaks are sporadic, the bertha armyworm continues to be an economically significant pest of canola in western Canada. Researchers initiated a three-year study in 2012 to revalidate the relationship between the trap catches and larval populations, as well as to more clearly define the risk categories and fine-tune the monitoring system to improve forecasting efficiency. Overall, the research has reconfirmed the validity of the current bertha armyworm monitoring and forecasting system and that current warning thresholds based on moth catches appear to be working well.
Although outbreaks are sporadic, the bertha armyworm continues to be an economically significant pest of canola in western Canada. The bertha armyworm monitoring and forecasting system, based on the original pheromone trapping research done in 1987, has seen major improvements over the years. These advancements include changes in trap design, an improved pheromone, which is 2x more attractive, and dramatic growth in the canola industry and hybrid canola acreage over the last 20 years.
Researchers in Alberta and Saskatchewan initiated a three-year study in 2012 to revalidate the relationship between the trap catches and larval populations, as well as to more clearly define the risk categories and fine-tune the monitoring system to improve forecasting efficiency. The objectives of this research were to determine the optimum density of traps to maximize accuracy and minimize overlap, and to determine the relationship between adult counts in pheromone traps and in-field larval numbers. The project also tested the impact of trap height and field location on bertha armyworm moth catch, and set out to develop and test a system that would reduce bumblebee catch in bertha armyworm traps without affecting the moth catch for forecasting.
Researchers selected a trap density study area based on increasing pheromone trap catches in 2011, which included 35 townships of cropland covering parts of three counties east of Edmonton. This area also turned out to be located in a region of a bertha armyworm outbreak in 2012. Two unitraps were established in each field 2 m from the field edge and 50 m apart. Bertha armyworm moths were counted weekly from mid-June throughout July, with larval sampling in selected fields with pheromone traps plus nine surrounding fields. In addition, larval counts were also carried out in Saskatchewan in 2012 and 2013 in fields by AAFC Saskatoon staff. Trap heights of 150 cm (high) and 75 cm (low) were compared to determine if trapping efficiency was impacted. Moth catches were evaluated in relationship to canola acreage in the previous year from the surrounding area, using both the amount of canola in l, 2 and 3 mile radii around the trap field and canola within the township.
The results of the study showed that the traps tended to catch more moths in areas of intense canola production the previous year. The number of moths was correlated with the number of canola fields close by, and held true for canola within 1, 2 and 3 miles of the trap. Pheromone traps in townships with significant canola production the previous year resulted in a predictable increase in moths caught, compared to lower areas with less canola the year previous. This suggests that traps should be placed in areas with high concentrations of canola the previous year. The proximity of last year's canola was also more important than any differences that may be created by prevailing wind direction. Results showed no significant difference in the number of moths caught in traps at a height of 150 cm compared to 75 cm, so the recommendation on trap height placement will not change. As well, the location of the trap along the field margin produced no significant difference.
The results of the damage counts due to larval feeding showed that damage was not always the highest in the fields with pheromone traps, which indicates that traps are more indicative of the risk to an area rather than to specific fields. This also suggested that areas with large canola acreage one year, followed by less canola production the next year, may have a slightly higher risk of damaging larval populations the subsequent year than expected. Some progress was made in the development and testing of a trap design to reduce bumblebee catch, however additional design adjustments and testing are needed.
Overall, the research has reconfirmed the validity of the current bertha armyworm monitoring and forecasting system and that current warning thresholds based on moth catches appear to be working well. More traps will always give a better read on bertha armyworm populations but due to limitations the current rate of approximately 5 per county or a total of 300 in the province appears to be doing a good job as long as they are relatively well distributed. Although populations occasionally crash due to disease outbreaks, the current pheromone trapping system and trap density is very good at identifying areas that are at risk of bertha armyworms at damaging levels.
Scientific Publications
Olfert, O., Meers, S., Gavloski, J., Hartley, S., Otani, J. (2013). Bertha armyworm in western Canada. 2012 Crop Variety Highlights and Insect Pest Forecasts. Saskatoon Research Centre Technical Bulletin No. 2013-01. Pp. 8-9.
Olfert, O., Meers, S., Gavloski, J., Hartley, S., Otani, J. (2014). Bertha armyworm in western Canada. 2013 Crop Variety Highlights and Insect Pest Forecasts. Saskatoon Research Centre Technical Bulletin No. 2014-01. Pp. 8-9.
Full Report PDF: Improving crop risk assessment tools for bertha armyworms