Cabbage stem flea beetle larval surveys (2015)

Abstract

The aim of the project was initially to assess the impact of the 2014/15 infestations of CSFB larvae, and the measures used to control them, on the winter oilseed rape crop in selected areas of England. Specific objectives were:

1. To determine whether high autumn levels of crop damage from adult CSFB are associated with large CSFB larval populations. 2. To provide preliminary information on the impact of CSFB larval number on yield. 3. To determine whether the presence of larvae is always indicated by presence of leaf-scarring. The work was subsequently extended and the larval surveys repeated in 2016.

In 2015, significant differences in larval populations were found between counties. Larval number was above the treatment threshold in all counties surveyed except Surrey. The largest population was 28 larvae per plant in Cambridgeshire. Larval populations dropped significantly between February and April. The majority of larvae were found in the petioles rather than the stems in both February (95%) and April (65%).

In 2016, significant differences in larval population were found between high and low risk sites and counties, with the largest populations in Cambridgeshire and the smallest in Lincolnshire and East Yorkshire. At half of the sites, larval populations were above the autumn treatment threshold of five larvae per plant. Between the 2015 and 2016 larval surveys, the larval population increased by 20.4% in Cambridgeshire and 131.1% in Bedfordshire (other counties were not assessed in both years).

As in 2015, the majority of larvae (96%) were found in the petioles rather than the stem in 2016. Regression analysis found that adult feeding damage at the cotyledon and three to four leaf stage, and the number of leaves per plant were significantly correlated with the number of larvae per plant in February/March. Stepwise regression analysis found that a model consisting of adult feeding damage at the cotyledon stage, the number of leaves in February/March and their interaction was most accurate at predicting larval number in February/March (explaining 64% of the variance).