Research Article Received: 5 January 2014
Revised: 25 June 2014
Accepted article published: 14 July 2014
Published online in Wiley Online Library: 13 August 2014
(wileyonlinelibrary.com) DOI 10.1002/ps.3857
Prospects for the control of apple leaf midge Dasineura mali (Diptera: Cecidomyiidae) by mass trapping with pheromone lures Peter L Lo,a* James TS Walkera and D Max Sucklingb Abstract BACKGROUND: Apple leaf midge, Dasineura mali (Kieffer), poses quarantine issues for some apple export markets because larvae occasionally pupate in the stem end and calyx of fruit. Pheromone-baited oil-filled containers were used in 1 ha orchard plots to trap adult male D. mali in order to test the potential for mass trapping to reduce populations. RESULTS: Mass-trapped plots had 97% fewer adult males in pheromone traps and 48% fewer larvae per shoot in the second D. mali generation compared with untreated areas. Oil traps caught on average 900 000 D. mali per plot over 11 weeks during the second and third generations. Catches averaged 9200 per trap at plot corners. By comparison, catches were 51% lower 10–25 m away along borders, 80% lower at the midpoint of borders and 95% lower >7 m from plot edges. Fruit infestation was low (four out of 8000 apples). CONCLUSION: The attractiveness of the pheromone, monophagous habit and low mobility of D. mali enhance the prospects for successful mass trapping. Countering this are high populations, multivoltinism and aspects of mating behaviour. Mass trapping would probably have been more effective had it been in place season long and conducted over successive years. It needs refinements and more study before becoming a feasible control option for D. mali. © 2014 Society of Chemical Industry Keywords: Dasineura mali; apple leaf midge; Cecidomyiidae; sex pheromone; mass trapping
1
INTRODUCTION
Pest Manag Sci 2015; 71: 907–913
Feeding by larvae on the upper leaf surface causes the edges of leaves to roll tightly, forming characteristic distorted red-tinged leaves. The larvae develop for 2–3 weeks within these rolls and emerge when they are about to pupate. The great majority of mature larvae drop to the ground where they spin a tough silken cocoon and emerge as adults about 2 weeks later. Males emerge slightly ahead of females, mate as soon as the latter appear and oviposition begins the same day.4 In the laboratory, individual males mated up to 5 times, with a median of 2.7 times.4 Previous New Zealand studies reported that D. mali has three to five, and usually four, generations in Manawatu,10 Nelson11 – 13 and Central Otago.14 All of these studies determined the phenology by examining shoots for D. mali eggs. Peaks of oviposition by D. mali in the North Island typically occurred about early October, mid-December, early February and late March. D. mali larvae occasionally pupate in the calyces and stem ends of apples, which poses a much greater problem than the feeding damage.7 This infestation of fruit constitutes a quarantine issue for
∗
Correspondence to: Peter Lo, Plant and Food Research, Private Bag 1401, Havelock North 4157, New Zealand. E-mail: [email protected]
a The New Zealand Institute for Plant and Food Research Limited, Havelock North, New Zealand b The New Zealand Institute for Plant and Food Research Limited, Christchurch, New Zealand
www.soci.org
© 2014 Society of Chemical Industry
907
The cecidomyiid family of Diptera encompasses several thousand species, many of which are important agricultural pests. The female-produced sex pheromones of 16 species, including apple leaf midge, Dasineura mali (Kieffer), have been identified and synthesised during the past 15 years.1 The various cecidomyiid sex pheromones are highly attractive to conspecific males,1 and this potency has opened up new possibilities for population monitoring and pheromone-based control systems. There is the potential for managing pest species with synthetic pheromone through mating disruption,2 attract-and-kill3 or mass trapping.4 D. mali is a monophagous pest of apples (Malus × domestica) in New Zealand, northern Europe, North America and Argentina.5 It has been present in New Zealand since at least 1950.6 Feeding by larvae can seriously affect the growth of young trees by stunting new shoots, but it is generally inconsequential for mature trees, although larval damage to flowers causes deformed fruit. The extent of midge damage varies widely, but individual trees can suffer damage to up to 100% of shoots and 41% of leaves.7 The average loss of leaf area from D. mali damage can reach 33%, with an equivalent reduction in photosynthetic output.8 The effect on fruit yield has not been quantified in apple, but in pear a 10% increase in yield occurred when D. pyri (pear leaf midge) was controlled.9 D. mali overwinters in the soil as mature non-diapausing larvae.10 In spring and summer, females lay their eggs on the unfurling leaves of actively growing shoots, and they hatch after 3–6 days.5
www.soci.org export markets where D. mali is absent, such as Australia, California, China, Japan and Taiwan. Broad-spectrum organophosphate insecticides formerly applied against other pests also kept D. mali under control. The removal of organophosphate insecticides since 2000 under pipfruit integrated fruit production in New Zealand,15 together with increased scrutiny of D. mali from regulatory authorities, has led to the status of this pest being elevated. To reduce fruit infestation by D. mali cocoons, the pest is managed in orchards with insecticides, and post-harvest by apple washers in packhouses. Biological control by the parasitic wasp Platygaster demades10,13,16 and predators17,18 also contributes to reducing populations. Nevertheless, additional on-orchard control methods that contribute to a multistrategy risk minimisation approach are needed to augment existing methods for sustainable management of D. mali. The female-produced sex pheromone of D. mali was identified as (Z)-13-acetoxy-8-heptadecen-2-one. Its synthesis has enabled populations to be monitored,19 as well as mating disruption and attract-and-kill to be trialled.3 Mass trapping of D. mali was first tested in a young orchard of one-year-old trees.4 While that study achieved a high degree of suppression (96%) of adult male catches, a lack of shoot infestation in both treated and untreated plots indicated that there was little colonisation of the trial area by females. The present paper reports on a field trial to test the potential for control using mass trapping in a mature orchard. The phenology of D. mali in Hawke’s Bay, New Zealand, was also documented to provide data to support the interpretation of the trial results. This is the first published New Zealand study to monitor D. mali populations using the sex pheromone.
2
MATERIALS AND METHODS
2.1 Pheromone lures The racemic pheromone compound was synthesised in an eight-step procedure that produced an overall yield of 64% from 6-bromohexanoic acid. The product was 92% pure by gas chromatography and contained 4.5% of the corresponding E-isomer. Lures were prepared on red rubber septa (Thomas Scientific, Swedesboro, NJ) at indicated loadings. 2.2 Phenology in the year of the study Populations of D. mali were monitored in five apple orchards near Hastings in Hawke’s Bay, New Zealand (latitude 39.6 S, longitude 176.8 E), with two delta traps per orchard. Each trap had an 18 × 19 cm sticky base and a 3 μg pheromone lure, and they were
P L Lo, J T S Walker, D M Suckling positioned ∼0.6 m above the ground. Traps were put out on 21 August 2006 and checked approximately weekly until 26 March 2007. 2.3 Mass trapping trial design The trial site was a 36 ha mixed-cultivar commercial apple orchard in Hawke’s Bay (Fig. 1). Eight 1 ha plots (approximately 100 m by 100 m) were established, which encompassed 23–29 tree rows, depending on the row spacing between cultivar blocks. Four plots were selected for mass trapping, and 750 traps were put out in each of these plots on 13–14 November 2006 at the start of the second emergence period for D. mali. The traps were as evenly spaced as the tree rows allowed, with an average distance of 3.6 m apart. A row of one apple cultivar (either ‘Royal Gala’, ‘Braeburn’ or Pink Lady®) was chosen in the centre of the treated plots to assess D. mali shoot damage and infestation. The remaining four plots were untreated and were at least 80 m from a treated plot. They were matched with a treated plot for cultivar, rootstock, tree size and age. Because sticky traps would saturate rapidly and require frequent replacement, a trap based on a tephritid fruit fly trap known as a ‘Lynfield trap’20 was used instead. Each trap comprised a cylindrical 680 mL plastic container suspended from the lid by a wire. This wire was hung from irrigation pipes so that the top of the trap was approximately 60 cm above ground. Each container had four evenly spaced 16 mm diameter holes, 20 mm below the rim. A rubber septum loaded with 2 mg of D. mali pheromone was suspended in the middle of each container, which had about 150 mL of vegetable oil to drown midges that entered the trap. The orchard received a standard integrated fruit production spray programme.21 None of the insecticides that were applied for other pests was active against D. mali adults or larvae, except for mineral oil and carbaryl. Oil was applied only in late winter, well before any midge emerged. Carbaryl was applied to some varieties for fruit thinning at a lower rate than the insecticide rate. 2.4 Monitoring and assessments 2.4.1 Pretreatment Two sticky traps were placed 50 m apart in the centre of each plot on 8 November 2006. Each trap contained a sticky base with a 2 μg D. mali pheromone lure, and males were counted after 2 and 5 days. D. mali shoot damage was assessed on 20 November 2006 by examining 250 leaves per plot on spur clusters on one-year-old wood and terminal shoots. These clusters and shoots were
908
Figure 1. Diagram of the trial site, showing the location of plots and transects.
wileyonlinelibrary.com/journal/ps
© 2014 Society of Chemical Industry
Pest Manag Sci 2015; 71: 907–913
Control of Dasineura mali (apple leaf midge) by mass trapping with pheromone lures selected without bias as much as feasible, while walking slowly along a central row in each plot. This date was after the emergence of the overwintering generation and measured damage caused by the first generation of larvae, which were not affected by the deployment of mass trapping. 2.4.2 Post-treatment After the deployment of the oil traps, D. mali infestations were assessed in five ways. Firstly, two sticky traps per plot with 2 μg pheromone lures were checked for adult males fortnightly between 24 November 2006 and 5 February 2007. Secondly, on 22 December 2006, 21 oil traps per plot were collected at 10 m intervals along two transects that bisected the centre of plots from the midpoints of each side. These traps were replaced with new ones. This trapping period from mid-November to late December encompassed the first midge generation. Based on the pattern of catches in the first assessment of oil traps, a more comprehensive set of traps (32 per plot) was sampled on 29 January 2007. The interval between mid-November and late January covered the first two midge generations. This set comprised each corner trap and those 10 m and 25 m away on either side of the two corners that had the highest catches in the December assessment. In addition, the first four traps at the midpoint of each side moving into the plot were also collected (these were adjacent to traps sampled in the December transects), plus four randomly selected central traps >20 m from the plot edges. In the laboratory, the oil trap contents were poured onto trays, and the numbers of D. mali were counted or estimated in clusters of 100. Thirdly, the number of leaves with D. mali damage on actively growing shoots were counted on a transect running along the central row of each plot. Assessments were conducted after the second and fourth peaks of male catches on 7–8 December 2006 and 7 March 2007 respectively. In December, starting at the first tree in the plot, 20 shoots were examined at 10 m intervals along the row (total 220 shoots). In March, when trees had largely stopped growing, ten shoots were examined at each of five locations: within 10 m of the plot edges, 20–30 m from the edges and within 5 m of the plot centre. Fourthly, the numbers of larvae inside rolled leaves were counted in December. Starting at the first tree in the plot and thereafter at 25 m intervals along the same row used for shoot assessments, ten infested shoots were sampled at each location (total 50 per plot). The oldest leaf with an unbroken roll on each shoot was soaked in
www.soci.org
a 10% alcohol solution with a drop of detergent to encourage the larvae out of the rolls, which were also broken open. The final assessment was an examination at harvest of 1000 apples per plot: 100 apples from each of ten picking bins. 2.5 Data analysis Pretreatment counts of sticky trap catches and shoot damage between treatments were compared by one-tailed t-tests. The proportions of damaged leaves were angular transformed before analysis to stabilise the variances. Trap catches were compared using percentage reduction compared with the control [100 − 100*(catch in treatment/catch in control). Shoot damage, larval counts and the estimated number of larvae per shoot were compared between trapped and untreated plots by fitting a Poisson generalised linear mixed model using a log link function, with treatment as a fixed effect, and site and site/treatment as random effects. No consistent trends in damage along transects were observed, so the location data for each plot were combined. The significant P-value was
Revised: 25 June 2014
Accepted article published: 14 July 2014
Published online in Wiley Online Library: 13 August 2014
(wileyonlinelibrary.com) DOI 10.1002/ps.3857
Prospects for the control of apple leaf midge Dasineura mali (Diptera: Cecidomyiidae) by mass trapping with pheromone lures Peter L Lo,a* James TS Walkera and D Max Sucklingb Abstract BACKGROUND: Apple leaf midge, Dasineura mali (Kieffer), poses quarantine issues for some apple export markets because larvae occasionally pupate in the stem end and calyx of fruit. Pheromone-baited oil-filled containers were used in 1 ha orchard plots to trap adult male D. mali in order to test the potential for mass trapping to reduce populations. RESULTS: Mass-trapped plots had 97% fewer adult males in pheromone traps and 48% fewer larvae per shoot in the second D. mali generation compared with untreated areas. Oil traps caught on average 900 000 D. mali per plot over 11 weeks during the second and third generations. Catches averaged 9200 per trap at plot corners. By comparison, catches were 51% lower 10–25 m away along borders, 80% lower at the midpoint of borders and 95% lower >7 m from plot edges. Fruit infestation was low (four out of 8000 apples). CONCLUSION: The attractiveness of the pheromone, monophagous habit and low mobility of D. mali enhance the prospects for successful mass trapping. Countering this are high populations, multivoltinism and aspects of mating behaviour. Mass trapping would probably have been more effective had it been in place season long and conducted over successive years. It needs refinements and more study before becoming a feasible control option for D. mali. © 2014 Society of Chemical Industry Keywords: Dasineura mali; apple leaf midge; Cecidomyiidae; sex pheromone; mass trapping
1
INTRODUCTION
Pest Manag Sci 2015; 71: 907–913
Feeding by larvae on the upper leaf surface causes the edges of leaves to roll tightly, forming characteristic distorted red-tinged leaves. The larvae develop for 2–3 weeks within these rolls and emerge when they are about to pupate. The great majority of mature larvae drop to the ground where they spin a tough silken cocoon and emerge as adults about 2 weeks later. Males emerge slightly ahead of females, mate as soon as the latter appear and oviposition begins the same day.4 In the laboratory, individual males mated up to 5 times, with a median of 2.7 times.4 Previous New Zealand studies reported that D. mali has three to five, and usually four, generations in Manawatu,10 Nelson11 – 13 and Central Otago.14 All of these studies determined the phenology by examining shoots for D. mali eggs. Peaks of oviposition by D. mali in the North Island typically occurred about early October, mid-December, early February and late March. D. mali larvae occasionally pupate in the calyces and stem ends of apples, which poses a much greater problem than the feeding damage.7 This infestation of fruit constitutes a quarantine issue for
∗
Correspondence to: Peter Lo, Plant and Food Research, Private Bag 1401, Havelock North 4157, New Zealand. E-mail: [email protected]
a The New Zealand Institute for Plant and Food Research Limited, Havelock North, New Zealand b The New Zealand Institute for Plant and Food Research Limited, Christchurch, New Zealand
www.soci.org
© 2014 Society of Chemical Industry
907
The cecidomyiid family of Diptera encompasses several thousand species, many of which are important agricultural pests. The female-produced sex pheromones of 16 species, including apple leaf midge, Dasineura mali (Kieffer), have been identified and synthesised during the past 15 years.1 The various cecidomyiid sex pheromones are highly attractive to conspecific males,1 and this potency has opened up new possibilities for population monitoring and pheromone-based control systems. There is the potential for managing pest species with synthetic pheromone through mating disruption,2 attract-and-kill3 or mass trapping.4 D. mali is a monophagous pest of apples (Malus × domestica) in New Zealand, northern Europe, North America and Argentina.5 It has been present in New Zealand since at least 1950.6 Feeding by larvae can seriously affect the growth of young trees by stunting new shoots, but it is generally inconsequential for mature trees, although larval damage to flowers causes deformed fruit. The extent of midge damage varies widely, but individual trees can suffer damage to up to 100% of shoots and 41% of leaves.7 The average loss of leaf area from D. mali damage can reach 33%, with an equivalent reduction in photosynthetic output.8 The effect on fruit yield has not been quantified in apple, but in pear a 10% increase in yield occurred when D. pyri (pear leaf midge) was controlled.9 D. mali overwinters in the soil as mature non-diapausing larvae.10 In spring and summer, females lay their eggs on the unfurling leaves of actively growing shoots, and they hatch after 3–6 days.5
www.soci.org export markets where D. mali is absent, such as Australia, California, China, Japan and Taiwan. Broad-spectrum organophosphate insecticides formerly applied against other pests also kept D. mali under control. The removal of organophosphate insecticides since 2000 under pipfruit integrated fruit production in New Zealand,15 together with increased scrutiny of D. mali from regulatory authorities, has led to the status of this pest being elevated. To reduce fruit infestation by D. mali cocoons, the pest is managed in orchards with insecticides, and post-harvest by apple washers in packhouses. Biological control by the parasitic wasp Platygaster demades10,13,16 and predators17,18 also contributes to reducing populations. Nevertheless, additional on-orchard control methods that contribute to a multistrategy risk minimisation approach are needed to augment existing methods for sustainable management of D. mali. The female-produced sex pheromone of D. mali was identified as (Z)-13-acetoxy-8-heptadecen-2-one. Its synthesis has enabled populations to be monitored,19 as well as mating disruption and attract-and-kill to be trialled.3 Mass trapping of D. mali was first tested in a young orchard of one-year-old trees.4 While that study achieved a high degree of suppression (96%) of adult male catches, a lack of shoot infestation in both treated and untreated plots indicated that there was little colonisation of the trial area by females. The present paper reports on a field trial to test the potential for control using mass trapping in a mature orchard. The phenology of D. mali in Hawke’s Bay, New Zealand, was also documented to provide data to support the interpretation of the trial results. This is the first published New Zealand study to monitor D. mali populations using the sex pheromone.
2
MATERIALS AND METHODS
2.1 Pheromone lures The racemic pheromone compound was synthesised in an eight-step procedure that produced an overall yield of 64% from 6-bromohexanoic acid. The product was 92% pure by gas chromatography and contained 4.5% of the corresponding E-isomer. Lures were prepared on red rubber septa (Thomas Scientific, Swedesboro, NJ) at indicated loadings. 2.2 Phenology in the year of the study Populations of D. mali were monitored in five apple orchards near Hastings in Hawke’s Bay, New Zealand (latitude 39.6 S, longitude 176.8 E), with two delta traps per orchard. Each trap had an 18 × 19 cm sticky base and a 3 μg pheromone lure, and they were
P L Lo, J T S Walker, D M Suckling positioned ∼0.6 m above the ground. Traps were put out on 21 August 2006 and checked approximately weekly until 26 March 2007. 2.3 Mass trapping trial design The trial site was a 36 ha mixed-cultivar commercial apple orchard in Hawke’s Bay (Fig. 1). Eight 1 ha plots (approximately 100 m by 100 m) were established, which encompassed 23–29 tree rows, depending on the row spacing between cultivar blocks. Four plots were selected for mass trapping, and 750 traps were put out in each of these plots on 13–14 November 2006 at the start of the second emergence period for D. mali. The traps were as evenly spaced as the tree rows allowed, with an average distance of 3.6 m apart. A row of one apple cultivar (either ‘Royal Gala’, ‘Braeburn’ or Pink Lady®) was chosen in the centre of the treated plots to assess D. mali shoot damage and infestation. The remaining four plots were untreated and were at least 80 m from a treated plot. They were matched with a treated plot for cultivar, rootstock, tree size and age. Because sticky traps would saturate rapidly and require frequent replacement, a trap based on a tephritid fruit fly trap known as a ‘Lynfield trap’20 was used instead. Each trap comprised a cylindrical 680 mL plastic container suspended from the lid by a wire. This wire was hung from irrigation pipes so that the top of the trap was approximately 60 cm above ground. Each container had four evenly spaced 16 mm diameter holes, 20 mm below the rim. A rubber septum loaded with 2 mg of D. mali pheromone was suspended in the middle of each container, which had about 150 mL of vegetable oil to drown midges that entered the trap. The orchard received a standard integrated fruit production spray programme.21 None of the insecticides that were applied for other pests was active against D. mali adults or larvae, except for mineral oil and carbaryl. Oil was applied only in late winter, well before any midge emerged. Carbaryl was applied to some varieties for fruit thinning at a lower rate than the insecticide rate. 2.4 Monitoring and assessments 2.4.1 Pretreatment Two sticky traps were placed 50 m apart in the centre of each plot on 8 November 2006. Each trap contained a sticky base with a 2 μg D. mali pheromone lure, and males were counted after 2 and 5 days. D. mali shoot damage was assessed on 20 November 2006 by examining 250 leaves per plot on spur clusters on one-year-old wood and terminal shoots. These clusters and shoots were
908
Figure 1. Diagram of the trial site, showing the location of plots and transects.
wileyonlinelibrary.com/journal/ps
© 2014 Society of Chemical Industry
Pest Manag Sci 2015; 71: 907–913
Control of Dasineura mali (apple leaf midge) by mass trapping with pheromone lures selected without bias as much as feasible, while walking slowly along a central row in each plot. This date was after the emergence of the overwintering generation and measured damage caused by the first generation of larvae, which were not affected by the deployment of mass trapping. 2.4.2 Post-treatment After the deployment of the oil traps, D. mali infestations were assessed in five ways. Firstly, two sticky traps per plot with 2 μg pheromone lures were checked for adult males fortnightly between 24 November 2006 and 5 February 2007. Secondly, on 22 December 2006, 21 oil traps per plot were collected at 10 m intervals along two transects that bisected the centre of plots from the midpoints of each side. These traps were replaced with new ones. This trapping period from mid-November to late December encompassed the first midge generation. Based on the pattern of catches in the first assessment of oil traps, a more comprehensive set of traps (32 per plot) was sampled on 29 January 2007. The interval between mid-November and late January covered the first two midge generations. This set comprised each corner trap and those 10 m and 25 m away on either side of the two corners that had the highest catches in the December assessment. In addition, the first four traps at the midpoint of each side moving into the plot were also collected (these were adjacent to traps sampled in the December transects), plus four randomly selected central traps >20 m from the plot edges. In the laboratory, the oil trap contents were poured onto trays, and the numbers of D. mali were counted or estimated in clusters of 100. Thirdly, the number of leaves with D. mali damage on actively growing shoots were counted on a transect running along the central row of each plot. Assessments were conducted after the second and fourth peaks of male catches on 7–8 December 2006 and 7 March 2007 respectively. In December, starting at the first tree in the plot, 20 shoots were examined at 10 m intervals along the row (total 220 shoots). In March, when trees had largely stopped growing, ten shoots were examined at each of five locations: within 10 m of the plot edges, 20–30 m from the edges and within 5 m of the plot centre. Fourthly, the numbers of larvae inside rolled leaves were counted in December. Starting at the first tree in the plot and thereafter at 25 m intervals along the same row used for shoot assessments, ten infested shoots were sampled at each location (total 50 per plot). The oldest leaf with an unbroken roll on each shoot was soaked in
www.soci.org
a 10% alcohol solution with a drop of detergent to encourage the larvae out of the rolls, which were also broken open. The final assessment was an examination at harvest of 1000 apples per plot: 100 apples from each of ten picking bins. 2.5 Data analysis Pretreatment counts of sticky trap catches and shoot damage between treatments were compared by one-tailed t-tests. The proportions of damaged leaves were angular transformed before analysis to stabilise the variances. Trap catches were compared using percentage reduction compared with the control [100 − 100*(catch in treatment/catch in control). Shoot damage, larval counts and the estimated number of larvae per shoot were compared between trapped and untreated plots by fitting a Poisson generalised linear mixed model using a log link function, with treatment as a fixed effect, and site and site/treatment as random effects. No consistent trends in damage along transects were observed, so the location data for each plot were combined. The significant P-value was
