Project
title / Adapting pest management strategies for novel growing methods on a perennial crop.
/ DEFRA
project code / HH1935SHO

Department for Environment, Food and Rural Affairs CSG 15

Research and Development

Final Project Report

(Not to be used for LINK projects)

Two hard copies of this form should be returned to:
Research Policy and International Division, Final Reports Unit
DEFRA, Area 301
Cromwell House, Dean Stanley Street, London, SW1P 3JH.
An electronic version should be e-mailed to
Project title / Adapting pest management strategies for novel growing methods on a perennial crop.
DEFRA project code / HH1935SHO
Contractor organisation and location / Horticulture Research International
East Malling
Kent ME19 6BJ
Total DEFRA project costs / £ 369,067
Project start date / 01/04/99 / Project end date / 31/10/02
Executive summary (maximum 2 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.

CSG 15 (Rev. 6/02) 3

Project
title / Adapting pest management strategies for novel growing methods on a perennial crop.
/ DEFRA
project code / HH1935SHO

Hops are susceptible to attack by two arthropod species that are capable of destroying the crop entirely (damson-hop aphid and two-spotted spider mite), and to several other mostly less serious pests. After three centuries of control by pesticide sprays, damson-hop aphid is now at the threshold of being managed entirely biologically by a combination of partially aphid resistant dwarf plants and the activity of natural enemies. Currently available hop cultivars are highly susceptible to aphids, so some intervention with a pesticide is necessary on those cvs as part of an integrated pest management programme under development. Previous DEFRA-funded research showed that spider mites too could be controlled without pesticides by supplementing naturally occurring enemies with a well timed inoculative release of the alien predatory mite Phytoseiulus persimilis. However, the rates of predator release used in those studies are economically non-viable in commerce. More effective exploitation of P. persimilis was the main objective of the part of this research project undertaken at HRI-East Malling.

Since organophosphorous pesticides were withdrawn from use on hops after resistance by aphids and spider mites rendered them ineffective (reinforced more recently by public disquiet), several species of insects that were important pests before the 1950’s have become troublesome. Among these, caterpillars of currant pug moth and root-feeding weevils are potentially the most damaging. Hop growers may only use broad-spectrum pyrethroid insecticide sprays against these resurgent pests; pesticides that are toxic to natural enemies of aphids and spider mites yet are ineffective against these pests owing to their earlier developed resistance. Within this project, the main aims of work undertaken by ADAS-Rosemaund were to assess the seriousness of caterpillars and weevils as pests of dwarf hops and to develop a control strategy against current pug moth based on its sex pheromone.

Field experiments were made on two dwarf hop cultivars during 1999, 2000 and 2001 to compare alternative strategies for managing spider-mite pests by exploiting inoculative release of P. persimilis as a supplement to naturally-occurring predators. Pest-in-first tactics, whereby plants were seeded with commercially-sourced spider mites and allowed to establish before predators were released, had no detectable influence on subsequent pest and predator population development and were an unnecessary cost. Indeed, the prior release of the pest was counterproductive in 2000, as significantly more P. persimilis and other phytoseiids were recorded on the non-release plots than on those where the pest had been either released in one patch, or were spread uniformly on all plants in the row. Release of P. persimilis proved to be unnecessary in 2001, as naturally-occurring predators, particularly phytoseiid mites, were sufficiently numerous to regulate spider mite numbers below the economic damage threshold of 10 spider mite active stages per leaf throughout the crop cycle. In 1999 and 2000, predators maintained pest numbers below the economic damage threshold on both hop cvs where P. persimilis was released at the equivalent of 10 predators per plant. Pest numbers exceeded the damage threshold in predator non-release plots. However, P. persimilis competed with naturally-occurring phytoseiids in 2000, as significantly more phytoseiids were recorded from the non-release than the predator release plots. The pattern of P. persimilis release was found to be unimportant. Spider mite control was equally effective whether the predator dispersed naturally from one central release point within a row of plants, or whether it was spread uniformly between plants. Single-point release should prove the cheapest option.

Results from releases of P. persimilis in the centre row only of three rows of dwarf hops at a rate equivalent to 3 predators per plant were more variable than in the experiments where the predators were released at ca 10 per plant. In 1999, the predators controlled spider mites on cv Herald, whereas on First Gold, although they had a significant effect when compared with predator non-release plots, they did not prevent numbers exceeding the economic damage threshold. In 2000, the reverse pattern was observed; spider mite populations were maintained below the economic damage threshold on First Gold but not on Herald. In both years, P. persimilis soon spread aerially from the release row to adjacent rows and by both aerial and ambulatory dispersal within the release rows. Spider mite populations developed at similar rates on predator-release and non-release rows within predator release plots, and within the individual rows of each plot, irrespective of the pattern of introduction of the predator. The results suggest that it is not necessary to release predators on every row of dwarf hops; natural aerial dispersal by the predator is exploitable commercially. It is argued that the numbers of P. persimilis released during the course of this work and the control they exerted was economically competitive with that provided by an acaricidal spray programme.

Intensive studies were made throughout the day of the spread of marked P. persimilis through the hop hedge as they dispersed from vials fixed near the base at the mid-point of a 32.5 * 2.5 m high row of hops. As expected, the majority of marked predators were recovered from zones nearby the release point. However, the strong natural dispersal powers of the predator were demonstrated in one of the experiments when 5 marked predators were recovered in a zone 15 m from the release point just 10 hours after the vials were vented. A computer model was written that simulates predator movement within the hedge throughout the day. The model parameters were verified using data obtained from two further days of intensive sampling. In a future project, the predator dispersion model could be further developed as a tool to predict optimal patterns of predator release for larger row lengths. Furthermore, if that model was combined with a future population dynamics model, it would be possible to provide hop growers with predictive tools for deciding optimal release rates for P. persimilis in a biocontrol programme.

Currant pug moth (CPM) was the commonest caterpillar pest on commercial dwarf hops in the west-midlands. CPM mainly overwinters in dead vegetation left after harvest. Carabid and staphylinid beetles were the commonest ground-dwelling predators, and their seasonal abundance coincided with that of CPM. Parasitoids eliminated laboratory cultures of CPM, so prevented studies on its sex pheromone. Mirid bugs and caterpillars damaged foliage and cones at commercial farms. In pot experiments, vine weevils reduced hop root mass by around half. In a similar field study, 90% fewer vine weevil larvae were recovered than were released. The molluscicides methiocarb and metaldheyde had no significant effect on vine weevil larval numbers.

Opportunities were exploited to publicise the findings and to transfer the technologies developed to interested parties. Results were presented at 18 organised meetings, mostly held in the west-midland and south-east hop growing regions, and at 4 scientific conferences. In addition, informal presentations and demonstrations were made to hop growers visitors at HRI-East Malling, the majority being conducted on the experimental plots. Ten of the presentations were given at meetings where the audience consisted almost entirely of hop specialists, growers and their industry advisors.

CSG 15 (Rev. 6/02) 3

Project
title / Adapting pest management strategies for novel growing methods on a perennial crop.
/ DEFRA
project code / HH1935SHO
Scientific report (maximum 20 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.

CSG 15 (Rev. 6/02) 3

Project
title / Adapting pest management strategies for novel growing methods on a perennial crop.
/ DEFRA
project code / HH1935SHO

The scientific objectives of the spider-mite biocontrol components undertaken at HRI-East Malling, were:

1.  To determine the rates of dispersal of Phytoseiulus persimilis within and between rows of dwarf hops.

2.  To quantify the optimum patterns of release of phytoseiids on dwarf hops consistent with achieving reliable and adequate regulation of the numbers of the two-spotted spider mite.

3.  To model mathematically the initial dispersal of predatory mites in relation to the numbers and positioning of release points.

4.  To determine the species composition and quantify the damage caused by lepidopteran caterpillars to dwarf hops.

5.  To identify the sex pheromone of the currant pug moth and to use it in the field for identifying a treatment threshold.

6.  To quantify damage caused by root-feeding weevils on dwarf hops and to evaluate the impact of naturally-occurring predators and pathogens.

Objective 1 was addressed by intensive studies of the movement of ca 104 marked predators from single release points in individual hedges of dwarf hops during 12 h, and by examining the season-long spread and control of pests within and between hedges following an inoculative release of the predator on the middle row only of three-row plots of hops. The season-long field experiments, outlined above, provided data for objective 2 also. In addition, a factorial experiment, repeated in each of the three years, examined the long-term effects of establishing prey populations on the hop plants several weeks before release of the predator. The data from two days of the intensive studies were used to calculate parameters for variables incorporated in the simulation model. The model was verified with data from two further 12 h intensive studies. Objectives 4-6 were contracted to ADAS-Rosemaund.

Objectives 1-3: Improving biocontrol of Tetranychus urticae on dwarf hops with Phytoseiulus persimilis.

Introduction

The dense hedge structure of low trellis (≤ 3 m) and dwarf hop systems provide a favourable habitat for biocontrol of two-spotted spider mite (Tetranychus urticae Koch) using phytoseiid mites, as predators can disperse freely within the hop hedge (Campbell & Lilley, 1999a,b; Lilley & Campbell, 1999; Moosher, 1999; Vostřel, 2001). Moosher (1999) reported a 90% reduction in T. urticae populations by introduced Typhlodromus pyri Scheuten on low-trellis hops in Germany. Similar reductions were reported by Vostřel (2001) in the Czech Republic using T. pyri and Neoseiulus californicus (McGregor). In UK, inoculative releases of Phytoseiulus persimilis Athias-Henriot helped predators maintain pest populations below the suggested economic damage threshold of 10 leaf-1 (Strong & Croft, 1995) on dwarf hops (Campbell & Lilley, 1999a; Lilley & Campbell, 1999; Barber et al., 2003a; Jones et al., 2003). In the UK experiments, P. persimilis also dispersed aerially (Jung & Croft, 2001) to colonise and control spider mites on nearby rows of hops.

In previous DEFRA-funded experiments with P. persimilis on dwarf hops (Campbell & Lilley, 1999a,b; Lilley & Campbell, 1999; Barber et al., 2003a), a pest-in-first strategy was used, as recommended for control of T. urticae with this predator on glasshouse crops (Stenseth, 1985). On hops, the principle aims of introducing the pest were 1) to ensure that sufficient prey mites were available to aid establishment of the predators, and 2) to improve experimental efficiency by reducing natural plot-to-plot variability in mite numbers. However, pest supplementation adds expense and may be unnecessary for hop growers. In all previous studies on dwarf hops, predators and prey were distributed uniformly on every plant within the hop hedge. Purposeful exploitation of P. persimilis' ambulatory and aerial dispersal behaviours (Jung & Croft, 2001) offers potential for further cost savings, thereby improving biocontrol's competitiveness with acaricides used currently (see discussion).

Commercial exploitation of ambulatory dispersal of introduced predators depends upon the speed with which the predators spread from their points of release and locate individual metapopulations of prey (Strong et al., 1999). That knowledge is crucial for making informed decisions on the distances to allow between release points. The issue was addressed here by monitoring the dispersion during 10 h of marked P. persimilis from single points of release. Key factors influencing initial dispersal patterns were then modelled mathematically with the aim of identifying whether improvements could be made to release strategies.

Materials and Methods

General

Field experiments were carried out in 1999, 2000 and 2001 on mature hops planted in 1996. The plantation consisted of dwarf hop cvs 'Herald' and 'First Gold' in 112 rows, each of which contained 66 plants, spaced 0.5 m apart. Rows were oriented N-S and were separated by 2.5 m of bare soil between rows and 5-10 m of bare soil between plots. The climbing stems were supported by polypropylene netting (14 cm square mesh) suspended vertically between horizontal wires at 0.15 m and 2.5 m. Damson-hop aphid (Phorodon humuli) was controlled by a soil-drench of imidacloprid and powdery mildew (Sphaerotheca humuli) by sprays of triforine and bupirimate. Downy mildew (Pseudoperonospora humuli) was controlled by sprays of copper oxychloride plus metalaxyl before phytoseiids were released, and then by chlorothalonil. Fungicides were applied at manufacturers’ recommended rates and were selected to minimise harm to the phytoseiids. Fertilisers were applied as for normal farm practice.

Phytoseiulus persimilis and T. urticae in vermiculite were bought from BCP Ltd., Wye. The 2.5 x 7.5 cm plastic tubes of P. persimilis nominally contained either 1000 or 2000 individuals depending on the experiment. The tubes were rotated to thoroughly mix the predators among the vermiculite. The mixture was then sprinkled on basal leaves where most T. urticae are found early in the season (Lilley, et al., 1999). The operator walking pace and sprinkling rate for treatments in which mites were released on every plant were established in preliminary trials. Predators in five randomly selected tubes of nominally 1000 individuals were counted under a binocular microscope.