Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Action: Delay herbicide use Natural Pest Control

Key messages

Read our guidance on Key messages before continuing

Natural enemies: Two randomised, replicated, controlled trials from Australia and Denmark found more natural enemies when herbicide treatments were delayed. One of the studies found some but not all natural enemy groups benefited and fewer groups benefitted early in the season.
Weeds: One randomised, replicated, controlled study found more weeds when herbicide treatments were delayed.
Insect pests and damage: One of two randomised, replicated, controlled studies from Canada and Denmark found more insect pests, but only for some pest groups, and one study found fewer pests in one of two experiments and for one of two crop varieties. One study found lower crop damage in some but not all varieties and study years.
Yield: One randomised, replicated, controlled study found lower yields and one study found no effect.

Crops studied were beet and oilseed.

Supporting evidence from individual studies


A randomised, replicated, controlled study in 2001 on Mors, Denmark (Strandberg & Bruus Pederson 2002) found more ladybird (Coccinellidae) and sawfly larvae (Symphyta) in mid-July in plots receiving late applications of herbicide (averaging 1.80 and larvae/0.9 m², respectively) at recommended (0.30 and 0.2 larvae) or early (0.05 and 0.05 larvae) spraying dates. More rove beetles (Staphylinidae), adult ground beetles (Carabidae) and money spiders (Linyphiidae) also occurred when plots were sprayed late (42 rove beetles, 6 ground beetles and 33 money spiders/0.9 m²) rather than at recommended (19, 3 and 28, respectively) or early dates (18, 3 and 22). Rove beetles and money spiders showed similar patterns in mid-June. Groups that did not show an effect were not presented. Some pest groups such as planthoppers (Delphacidae) and leaf beetles (Chrysomelidae) were also more numerous in plots sprayed late. Plots treated later had more weeds (0.2-52.3 weeds/m²) than plots treated at recommended or early dates (0.6-18.8 and 0.1-5.7 weeds, respectively). Beet Beta vulgaris yields were similar between treatments (870-970 t root/ha). Glyphosate was sprayed in three timing treatments: early application (25 ­May and again 27 June), recommended application (14 June and 5 July) and late application (27 June and 16 July).


Four randomised, replicated experiments in 1999-2001 in Alberta, Canada (Dosdall et al. 2003) found root maggot Delia spp. damage was generally lower in canola Brassica napus sprayed with herbicide at late rather than early crop growth stages. Two experiments found lower root damage in plots sprayed at six-leaf (damage rating of 2.5) rather than four-leaf (2.3 rating) and two-leaf (2.2 rating) stages in one of two canola cultivars, when averaged across sites, years and other treatments. Another experiment found this effect for two out of three cultivars and a fourth experiment found plots sprayed at the six-leaf stage had lowest damage in 1999 and 2001 (ratings of 2.7-3.2, 3.1-3.4 and 3.0-3.4 in six-, four- and two-leaf stages, respectively) but no effect in 2000. One experiment found less root maggot eggs in plots treated at the six-leaf (0.8 eggs/plant) rather than four- (1.4 eggs) and two-leaf (1.6 eggs) stages for one of two cultivars, but another experiment found only slight differences. Canola seed yields varied but were slightly lower in the six-leaf (392-3,265 kg/ha) than the two-leaf (672-3,458 kg/ha) stage treatments in three experiments. Glufosinate was sprayed in three timing treatments replicated four times. Root damage was scored 1 to 5.


A randomised, replicated, controlled study in 2000 in Darwin, Australia (Paynter 2003) found more green twig-mining moths Neurostrota gunniella (natural enemies of the weed mimosa Mimosa pigra) survived when herbicide treatment to kill mimosa plants was delayed. Survival of green twig-mining moths was similar in herbicide-treated plots vs. untreated controls when herbicides were applied 23 days (51-61 vs. 48% of larvae survived to become adults) after moth egg laying, but survival was lower in the herbicide-treated plots when treated at 9 days (0-8% vs. 49%) or 16 days (5-30% vs. 80%) days since eggs were laid. Green twig-mining moth survival was similar for the three types of herbicide. Adult green twig-mining moths were released on 29 August to lay eggs on mimosa plants. Fluroxypyr, tebuthiuron and metsulfuron methyl herbicides were then applied at 9, 16 and 23 days after egg-laying. Treatments were tested in 5 x 5 m plots, each containing 20 mimosa plants grown in 30 cm-diameter pots. The number of larvae found before herbicides were applied was estimated from the number of leaflets mined by larvae. After herbicide treatment all emerging adult moths were collected using nets placed over the plants.


A randomised, replicated, controlled study in 2001 on Mors, Denmark (Strandberg et al. 2005, the same study as Strandberg & Bruus Pederson 2002) found more arthropods (including insects and spiders) in July in plots receiving late herbicide applications (averaging approximately 525 arthropods/m²) compared to plots receiving herbicide at recommended (290 arthropods) or early (230 arthropods) spraying dates. The diversity of arthropod groups and species in July was also higher in plots treated late (20 groups/0.9 m²) rather than at recommended (16 groups) or early (12 groups) spraying dates. This study did not distinguish between pest and natural enemy arthropods and aphids (Aphidoidea), thrips (Thysanoptera), mites (Acari) and springtails (Collembola) were not included. More weeds occurred in mid-May to mid-August in plots sprayed late (2-89 g weed dry weight/m²) than in plots sprayed at recommended (2-10 g dry weight) or early (2 g dry weight) dates. The experiment took place in 20 x 20 m plots of beet Beta vulgaris. Glyphosate herbicide (roundup ready) was sprayed in three timing treatments: early (25 May and again 27 June), recommended (14 June and 5 July) and late (27 June and 16 July) applications. Treatments were replicated four times. Arthropods were sampled with a Dietrick vacuum sampler.

Referenced papers

Please cite as:

Wright, H.L., Ashpole, J.E., Dicks, L.V., Hutchison, J., McCormack, C.G. & Sutherland, W.J. (2019) Some Aspects of Enhancing Natural Pest Control. Pages 603-625 in: W.J. Sutherland, L.V. Dicks, N. Ockendon, S.O. Petrovan & R.K. Smith (eds) What Works in Conservation 2019. Open Book Publishers, Cambridge, UK