Language selection


Develop spatial modelling and forecasting tools for the management of carrot rust fly and carrot weevil in carrot producing regions of Eastern Canada

Project Code: PRR14-020

Project Lead

Suzanne Blatt - Agriculture and Agri-Food Canada


Validate degree day models developed in Quebec for use in Nova Scotia, Prince Edward Island, Ontario, and British Columbia to determine the spatial pattern and timing of migration into the carrot crop by carrot rust fly and carrot weevil and evaluate the potential effects of soil texture, soil fungi and parasitism on pest populations

Results Summary


Carrot rust fly (CRF) and carrot weevil (CW) are two major pests of carrot causing serious economic damage in Eastern Canada. Larvae of both pests create tunnels in the carrot roots, making them unmarketable. Furthermore, secondary infections by bacteria and fungi may occur in these tunnels and cause important post-harvest damage prior to and during storage of carrots. Insecticides are the primary method of control. Proper timing of insecticide sprays based on population thresholds would allow growers to judiciously and efficiently use insecticides in a focused integrated pest management (IPM) program. A previously funded project by Pesticide Risk Reduction PRR07-090 Development of bioclimatic models to forecast the dynamics of two insect pests: carrot weevil and carrot fly updated degree day (DD) models developed in Quebec in the mid-1980s using extensive new datasets. These updated models can effectively predict emergence of the first generation of both pests in Quebec, but not of subsequent generations. However, it was unknown whether they could accurately predict populations of these pests in other provinces.

This 3 year project investigated the performance of the Quebec bioclimatic models for CRF and CW in Nova Scotia (NS), Prince Edward Island (PEI), Ontario (ON), and British Columbia (BC), by determining the spatial distribution of pest populations in the field and assessing correlations between carrot damage and adult trap captures. The project also investigated the presence of naturally occurring parasitoids and entomopathogenic fungi in pest populations as well as carrot damage caused by click beetles. This work was identified as a priority for action under Agriculture and Agri-Food Canada's (AAFC) Pesticide Risk Reduction Strategy for Root Insect Pests of Carrot, Parsnip and Onion.


Degree day models and spatial distribution

The CRF was studied in all four provinces. There were six field sites in NS, twelve in PEI, ten in ON and five in BC. Yellow sticky traps were placed in NS, PEI and BC fields and monitored two or three times each week. In ON, re-useable orange carton traps and yellow acrylic and carton traps were used.

The CW was studied in NS and ON as neither PEI nor BC report having this pest in their region. Over the course of the study, 28 field sites in NS and 14 in ON were monitored for CW emergence. Modified Boivin traps were used in NS, while regular plate traps were used in ON. Traps were placed in the fields in early spring and checked two to three times each week for CW presence. When possible, weather station units were placed in the fields of NS. In ON, weather data was taken from the closest weather station operated by Environment Canada.

In 2014, the spatial study deployed traps along two transects within the fields, at least 3 meters (m) in from the edge with 50 m between transects and 50 m between the traps within transects. A total of 10 traps in each field were monitored two or three times weekly. In 2015 and 2016, the spacing was altered to better capture any impact closer to the edges of the fields. Carrots in the fields were assessed for damage by CRF and CW and these observations were correlated with adult trap captures in the field.

Other studies

In ON, studies comparing trap colour (yellow vs orange) and different materials of yellow traps were conducted to evaluate their efficacy at capturing CRF.

During field sampling in NS in 2014, it was noted that adult click beetles (wireworms) were also attracted to the modified Boivin traps in the spring. Laboratory studies were initiated to evaluate whether traps infested with adult click beetles would deter CWs from entering the trap.

Carrot damage by CW larvae is often confused with damage from click beetle larvae. A laboratory study using potted carrots was conducted to compare the damage inflicted by both species.

There is a long held belief that CW has minimal capacity to fly and migrates into the field in the spring by crawling on the soil surface. To test this hypothesis, a variety of soil textures was used to monitor CW movement in the laboratory using an arena bioassay.

Entomopathogenic fungi have been reported to affect CRF in Europe and parasitoids have been associated to CW populations in NS. CRF and CW species were analysed for infection by fungi and parasitoids, respectively.


Fields in BC had very low CRF incidence and results were inconsistent to draw conclusions. In all other fields and years, it was found that the current DD models developed in Quebec did not accurately predict emergence of either species. This could be due to population adaptation to the specific region which may require more or less heat units for development. Further work, through a recent project PRR17-020 Applying updated bioclimatic information to validate models and refine Integrated Pest Management program recommendations for carrot rust fly and carrot weevil management continues to collect data towards the development of DD models specific for ON, NS and PEI.

Contrary to the widely held belief that both CRF and CW are found mainly along the perimeters or edges of a field, this study demonstrated that both species can be readily found within the field. CW, previously thought to migrate into the fields from the edges, was found well into the field early in the season. This provides strong, new evidence that CW can overwinter within the carrot field and potentially survive crop rotation. It is also possible that the flight capacity of CW has been greatly underestimated. In light of these findings, careful consideration should be placed on whether or not a perimeter or border spray for control should be discouraged.

Carrot damage did not correlate well with trap captures for either species. In both cases, trap captures were lower than carrot damage observed in the field. For CW, this is potentially due to the inability to monitor adults once carrots reach the fourth true-leaf stage and forms a canopy. At this stage, oviposition on the carrot plant begins and the CW lifecycle progresses under the carrot canopy likely for the remainder of the season as the fresh carrots in the field are more attractive and numerous to the weevils than the carrot root baits in the traps. A more effective bait to monitor CW throughout the season may serve to improve these correlations. For CRF, it is likely that the sticky traps used in this study were insufficient at capturing the complete CRF adult population. A new bait, GF-120TM Fruit Fly Bait investigated through another project PRR15-030 Investigation of GF-120 to suppress carrot rust fly populations in Eastern Canada and British Columbia for control of CRF in carrot could be further explored as a lure for sticky traps.

It was found that CRF was more attracted to the yellow traps and trap material was not as important as trap colour.

Laboratory studies showed that there was no negative impact by click beetles in traps on CW captures. These studies also demonstrated that while carrot damage from both species can be similar, there were subtle differences. Sharing this knowledge with growers would allow them to make better decisions about pesticide applications. If growers could tell apart the two types of larvae, pesticides would only be applied when CRF was identified. There are currently no pesticides available against click beetles in carrots. Therefore, with better knowledge, unnecessary pesticide applications would be potentially avoided.

Laboratory studies found no evidence of entomopathogenic fungi or parasitoid presence in the pests analysed as part of this project.

This study demonstrated that the DD models developed 30 years ago in Quebec did not predict emergence of either test species in the provinces studied. The use of more effective trapping baits and the collection of more emergence data within these provinces will allow the development of more accurate models for CRF and CW, which in turn will guide growers in the judicious and effective use of insecticides based on population thresholds.

Report a problem on this page
Please select all that apply:
Date modified: