Evaluation of an Integrated Management and Compliance Program for ...

FORUM

Evaluation of an Integrated Management and Compliance Program for Tomicus piniperda (Coleoptera: Scolytidae) in Pine Christmas Tree Fields DEBORAH G. MCCULLOUGH1

AND

CLIFFORD S. SADOF2

J. Econ. Entomo\.91(4): 785-795 (1998) ABSTRACT The pine shoot beetle, Tomicus piniperda L., is an exotic pest that breeds in phloem of recently cut or dying pine logs, stumps, and slash. Adults feed in shoots of live pines to complete maturation. After T. piniperda was discovered in North America in 1992, federal and state quarantines were imposed to regulate movement of pine Christmas trees, logs, and nursery stock out of infested counties. Currently, pine Christmas tree fields must be inspected in the fall and certified as free of T. piniperda before trees can be shipped out of quarantined areas. We developed an integrated program to manage T. piniperda in pine Christmas tree fields. Components of the program include sanitation to reduce availability of cut trees and stumps that serve as brood material, use of trap logs to attract parent beetles into cut trees which are subsequently destroyed, and an insecticide spray to control shoot-feeding beetles. This management program was formalized into a Compliance Program in cooperation with regulatory agencies and tested in 1995 and 1996 in a total of 48 Christmas tree fields in Michigan and Indiana. We periodically surveyed fields and growers to assess compliance with program requirements, levels of T. piniperda shoot damage, and feasibility of the management activities. In fields where all program requirements were completed, percentage of trees with T. piniperda shoot-feeding damage ranged from 0 to 4%. In comparison, shoot-feeding damage was observed on 28-67% of trees in unmanaged fields where brood material was available for colonization by parent beetles. Information provided by cooperating growers indicated that Compliance Program activities could be readily integrated into standard production practices. Adoption of a regulatory policy that inspects compliance with the year-round management program rather than inspecting fields just before harvest is likely to result in sustained T. piniperda control and a more unif0n11 application of regulatory standards to reduce the risk of spread of an exotic pest. KEY WORDS

pine shoot beetle, regulatory entomology, quarantine, exotic pest, bark beetle, Christmas tree production

THEPlNE SHOOT BEETLE, Tomicus piniperda L., is a univoltine pest of pines native to Europe and Asia. Parent adults become active early in spring after temperatures exceed 12°C (Bakke 1968;Haack and Lawrence 1995a,b). Adults utilize host volatiles to locate suitable brood material such as recently killed or cut pine trees, logs, stumps or slash, or severely stressed or dying pines (Byers et at. 1985, Schroeder 1988, Byers 1991, Tunset et al. 1993). Parent adults colonize the inner bark of brood material and oviposit. Larvae feed in the phloem for 6-10 wk, and then pupate. Progeny adults emerge in early summer and feed in shoots oflive pine trees throughout the summer to complete maturation. Each beetle may tunnel into 2-6 shoots, which eventually die and break off (Ulngstrom and Hellqvist 1991, McCullough and Smitley 1995). Maturation feeding in shoots continues until autumn, when beetles move down to the base of trees to overwinter. High populations of T. piniperda have been reported I l)t'partn1l'nt of Entomologyand Department of Forestry, 243 NnturalScienct'Building,Michig,mStnteUniversity,EastLansing,MI

48824··1115. C Department of Entomology,1158 EntomologyHall,Purdue University,Wl'st Lafayette,L'Il 47907-1158.

in Scandinavia and China after severe drought or when brood material such as pine logging slash has been abundant (Eidmann 1992, Hui 1991, Ulngstrom and Hellqvist 1993). Reduced tree growth caused by shoot-feeding can occur when brood material is consistently available, such as in forested areas surrounding pine sawmills (Ulngstrom 1980, Ulngstrom and Hellqvist 1991). Tomicus piniperda was 1st detected in North America in 1913, in a commercial Scotch pine, (Pinus sylvestris L.), nursery in New Jersey (Headlee 1914). This infestation was apparently eradicated because no further mention of this infestation or the beetle was reported in subsequent New Jersey state reports. T. piniperda was not detected again despite periodic scoIytid surveys (e.g. Deyrup 1981) until it was discovered in a Scotch pine plantation in Ohio in 1992 (Haack et al.1997). Results of subsequent delimitation surveys indicate that as of October 1997, T. piniperda was established in more than 220 counties in 9 states in the north central and northeastern United States, and in 18 counties in Ontario, Canada (NAPIS Database 1997). Recent research indicates that T. piniperda can successfully develop and shoot-feed in most North

0022-0493/98/0785-0795$02.00/0

© 1998 EntomologicalSocietyof America

786

JOURNAL

OF ECONOMIC

American pine species (Sadof et al. 1994, Lawrence and Haack 1995, Haack and Lawrence 1997). Federal and parallel state quarantines were imposed in 1992 on all infested counties to regulate movement of pine Christmas trees, logs, and nursery stock out of the quarantined area (USDA, APHIS, PPQ 1992). Regulations stipulate that pine Christmas tree fields must be inspected in autumn, if any trees from these fields will be shipped out of the quarantined area. If no beetles or evidence of T. piniperda are found during inspections, the field is certified and trees can be exported. If a single T. piniperda adult or tunnelled shoot is found during the inspection, however, the entire field is restricted. Trees from restricted fields can be sold only within the quarantined area. Although T. piniperda populations cause little physical or aesthetic damage to trees in managed fields, the quarantine has seriously affected the pine Christmas tree industry in Michigan and other north central states. Growers who wholesale large numbers oftrees typically contract with customers many weeks before regulatory inspections occur. If fields are restricted, growers may lose income and long-standing customers or face legal challenges if they are unable to meet contracts and deliver trees (Riessen 1997). We recently developed an integrated approach to T. piniperda management in pine Christmas tree production fields (McCullough and Sadof 1996). Past studies reported that several insecticides effectively reduced T.piniperda survival, but none provided 100% control (McCullough and Smitley 1995, McCullough et al. 1998). Therefore, we used a combination of cultural and chemical tactics to reduce the likelihood of T. piniperda infestation in Christmas tree fields including (1) destruction of potential brood material before T. piniperda progeny adults could complete development, (2) deployment of "trap logs" to attract reproductively active parent adults into cut trees that were then destroyed, and (3) 1 insecticide application to foliage to control progeny adult beetles beginning maturation feeding. This integrated management plan was formalized into a Compliance Program for T. piniperda (USDA, APHIS, PPQ 1996) that provided Christmas tree growers with deadlines and instructions for completing each management activity (McCullough and Sadof 1996). Federal and state plant regulatory officials agreed that if this management program effectively reduced T. piniperda populations to low or undetectable levels, then trees from these fields would represent little risk of introducing T. piniperda into new areas. Trees from fields known to be in compliance (i.e., meeting all requirements of the integrated management program) could then be shipped out of quarantined areas without requiring an additional inspection and certification in autumn before harvest (USDA, APHIS, PPQ 1996). The Compliance Program approach could provide regulatory officials with an alternative option for T. piniperda management. A similar approach might also be applicable to management of other exotic pests in the future.

Vol. 91, no. 4

ENTOMOLOCY

In cooperation with state and federal agencies, we evaluated the T. piniperda Compliance Program approach in a pilot study conducted in Scotch pine Christmas tree fields in 1995 and 1996. Fields in Indiana and Michigan were selected for the pilot study to ensure that a wide range of management scenarios was represented. Michigan is one of the major producers of Christmas trees in the United States. Each year, 2.8-3.8 million Christmas trees are shipped out of state, and roughly 60% of these trees are Scotch pine (Koelling et al. 1992, Michigan Agricultural Statistics Service 1994). Many trees are also sold locally, often on small "choose-and-cut" tree farms, where customers select their own trees. In Indiana, Christmas tree production is generally on a smaller scale. Most trees are sold to local customers or in large cities within the state. Although the management program and general sampling methods were developed jointly each year, implementation of the project was conducted by individual investigators and cooperating growers. Our 1st objective was to evaluate the effectiveness of the Compliance Program in reducing or maintaining T. piniperda populations at very low levels in Scotch pine fields. Our 2nd objective was to determine if the program could be incorporated into standard Christmas tree production practices and would be acceptable to growers. Each grower was asked to execute the same management program and maintain detailed records. We periodically surveyed the T piniperda population in each field and evaluated the timing and methods used by the growers to comply with the program requirements. Results of the pilot project are reported here, along with results of a preliminary study in Indiana designed to evaluate the relative effectiveness of trap logs and insecticide sprays. Materials and Methods Integrated Management Program. Growers participating in the pilot project in Michigan and Indiana were asked to follow the same 3-part T. piniperda management program. Sanitation, the 1st component of the program, included destruction of cut trees and other potential brood material by 20 May to prevent development of progeny beetles. Potential brood material was destroyed by chipping in most wholesale fields and was burned in 2 of the small choose-and-cut fields. Stumps cut within the previous 10 mo had to be 500, depending on field size and number of trees remaining in partially harvested blocks. Growers deployed 26-99 trap logs per hectare in wholesale fields and 18-31 trap logs per hectare in choose-and-cut fields each year. In wholesale fields, trap logs were correctly set, collected, and destroyed by chipping or burning, except 1 field where trap IOl!;s were destroyed early (Tables 1 and 2). In chooseand-cut fields, trap logs were occasionally set or collected later than specified, and a few logs were inadvertently left out in 1 field (Tables 1 and 2). Observers surveyed trap logs at 20- to 50-m intervals in late March and April for boring dust or other signs of T. piniperda colonization in 1995and 1996.Five trap logs collected from each field in early May 1996 were transported to a rearing facility in Ingham County and caged until progeny beetles emerged. Emerged beetles were identified to confirm that trap logs were colonized by T. piniperda and not by native scolytid species. Fields were monitored periodically during the summer in 1995 and 1996 for compliance with sanitation requirements. In 1996, we also recorded height and evidence of T. piniperda colonization on the first 30 newly harvested stumps encountered at 60-m intervals

JOURNALOF ECONOMICENTOMOLOCY

788

Vol. 91, no. 4

Table 2. Michigllndalll from wholesllle(WH), ehoose-and-cul(CC), and urunlll1llged(UM)Scolch pine Christmastree fieldsthaI pllrtidpllted in T. pu.iperda piiOIstudyin 1996 Trap logs

Field

ha

WH-2 21.0 WH-3 24.3 WH-5 17.0 WH-6" 31.6 WH_7 8.1 WH-8" 26.3 WH_9"·d 4.0 CC_2c•d 3.6 CC-3c.d 3.2 C

Stumps Fall survey Regulatory Compliance Programelementsnot No. cover Height No. No. % % % spraysU rank inspection completed (em) checked injured checked infested :!:SE infested 169 161 115 238 33 172 63 47 25

5 0 4 3 9 2 5 30 8

7.1:!:0.18 24.0:!:1.05 5.0:!:0.10 6.8:!:0.13 9.4:!:0.15 4.5:!:0.26 6.5:!:0.15 2.1:!:0.06 4.7:!:0.16

0 0 0 6 7 6 9 0 0

3 3 4 2 2 3 2 0 0

282 207 252 462 65 400 100 58 65

1 0 0 0 5 0 4 0 5

Passed Passed Passed Passed Restricted Passed Restricted Passed Restricted

UM-2

2.8

20

30

NA

0

77

53

NA

UM-3

0.8

20

40

NA

0

34

56

NA

UM-4b

2.4

20

60

NA

0

69

67

NA

4 3 4 4 3 4 4 3 2

Stumpstoo hi!!:h Trap lo!!:.s collectedearly No coverspray No spray;trap logsset and collectedlate No sanitation,trap lo!!:s,or spray No sanitation,trap lo!!:s,or spray No sanitation,trap lo!!:s,or spray

Numberof trap logsand trees surveyedper fielddependedon fieldsizeand numberof trees remainingin fields.Thirty-fivestumpswere checkedin each field.NA,not applicable. a Onlyspraysappliedto foliageduringT.piniperda shoot-feedingperiod are included. /,Fieldsadded to pilotproject in 1996. Regulatoryinspectionfollowedmoreintensivenurseryprotocol. d Field restrictedin 1995due to T. piniperda injury. C

along diagonal transects through blocks in each field in 1996. Growers cooperating in the pilot study completed questionnaires in January for each field enrolled in the project during the previous year, providing information on methods and timing of activities such as destruction of brood material, management of stumps and trap logs, and timing and target pest of insecticide applications. Each field was visited in late September or October in 1995 and 1996 by a Michigan Department of Agriculture field inspector. Fields were inspected for evidence of T. piniperda following the federal regulatory protocols for Christmas tree fields (USDA, APHIS, PPQ 1993b). In the small choose-and-cut fields, inspectors followed the more intensive protocol designed for nursery operations (USDA, APHIS, PPQ 1993b). Any evidence of current or recent T. piniperda shoot-feeding resulted in restriction of the field, regardless of the inspection method used. Three additional Scotch pine fields were designated as untreated controls in 1995. These fields ranged in size from 0.8 to 4 ha, and no efforts were made to manage T. piniperda, In unmanaged field UM-l, we observed T. piniperda adults feeding in shoots of trees in 1994, and surveys in March 1995 indicated that 3% of the trees had damaged shoots left from 1994. However, no harvesting occurred in UM-l in 1994, and there was no evidence that cut trees, stumps, or other suitable brood material were present for T. piniperda colonization in 1995. This field was sold in 1995 and was no longer available for our use. Another unmanaged Scotch pine field, UM-4, was selected as a replacement in 1996. Abundant brood material including fresh stumps and piles of cut trees was present in

the UM-4 field in spring 1996. The UM-2 and UM-3 fields had freshly cut trees, stumps, and logs present during the spring T. piniperda flight in both 1995 and 1996. We surveyed 34-172 trees per field in April 1996 and September 1995 and 1996, to estimate percentage of trees with T. piniperda shoot damage. In 1996, at least 20 cut trees or logs scattered throughout the 3 control fields were examined in April and found to be colonized by T. piniperda (Table 2). Indiana: Preliminary Trap Log and Cover Spray Study. A 12-hectare field of 7-yr-old Scotch pine located in Starke County, IN, was used in 1994 to evaluate effectiveness of trap logs and a foliar insecticide spray in reducing T. piniperda shoot-feeding injury. This field was quarantined in September 1993 when shoots injured by T. piniperda were detected by state inspectors. Because trees had not been previously harvested from this field, stumps that could serve as breeding sites for T. piniperda in 1994 were not present. All dead and broken trees that could serve as T. piniperda brood material were removed from the field 18 February 1994. Trees were planted in blocks 002 rows 0000 trees, separated by a3.7-m wide grass strip. Blocks of trees were arranged so that 6 blocks were on the western half of the field and 6 blocks were on the eastern half. Twelve experimental plots, each consisting of 12 rows of 70 trees, were located on the western and eastern sides of the field. The 6 western and 6 eastern plots were separated by 150 m, encompassing the remaining 40 trees of each block and the grass strip that ran down the center of the field. We used a split-plot design to distinguish between the effects of trap logs and insecticide treatments. The whole plot treatment of trap logs was created by de-

August 1998

MCCULLOUCH

AND SADOF: COMPUANCE

ploying trap logs in the 6 western plots, leaving the 6 eastern plots as check plots. Seventy live Scotch pine trees were cut at ground level on 18 February for use as trap logs. Trap logs were deployed at 18-m intervals along the grassy lanes that bordered each of the 6 treated plots on the western side of the field (30 trap logs per hectare). Within each whole plot, 3 subplots, 70 by 100 trees each, were sprayed by the cooperating grower on 16 June 1994 with chlorpyrifos 4 EC (emulsifiable concentrate) at a rate of 184 g (AI) /ha (0.5 g [AI] per tree) with an AgTech 3004 air blast sprayer (AgChem, Niles, MI) that delivered a total volume of 234 liter per hectare. Three unsprayed subplots within each whole plot were left as controls. Inspection of pine trap logs located in a field 5 km from this study site indicated that spring £light of parent beetles began on 21 March 1994. Fourteen of the trap logs in the study field were collected on 19 April and taken to a rearing facility in West Lafayette, IN. The remaining trap logs were moved to West Lafayette on 5 May. Effectiveness of treatments in each plot was assessed on 30 June and on 7 September by visually inspecting trees for the presence or absence of T. piniperda shoot-feeding injury. Each tree was examined for 3 min. Fifteen trees were examined for shoot injury in each plot in June, and 30 trees were inspected in September. Sanlpled trees were located along transects through each subplot at 10 trees from the edge in June and 15 and 40 trees from the edge in September. These 540 trees represented =4.5% of all trees in the 12 subplots. Effects of the whole plot (trap logs) and those of split plots (insecticide spray) on the percentage of injured trees in each sample were compared using a repeated measures split-plot analysis of variance (ANOVA) (SASInstitute 1989). Proportions of infested trees were arcsine square root transformed before analysis to meet assumptions of normality. To determine if trap trees attracted breeding beetles after they were cut and placed in the field, 2 trap logs placed between 30 and 60 trees from the western edge of the field were collected from 7 of the grassy lanes between blocks and brought to the rearing facility. Tnmk diameter at the 1st branch whorl, stem length, T. piniperda entrance holes, and number of progeny adults that emerged from each tree were recorded in the laboratory. Indiana Pilot Project. In 1995, 4 growers with 9 fields of 7- to 9-yr-old P. sylvestris known to be infested with T. piniperda participated in the Compliance Program pilot study. In 1996, 6 growers with 12 fields of 7- to 9-yr-old infested P. sylvestris participated. All fields were within 5 km of a stand of either P. sylvestris or Pinus resinosa where T. piniperda shoot feeding had been detected. Untreated control fields were not readily available because growers were reluctant to risk restricting sales of their crop and because most commercial growers implement 1 or more components of the Compliance Program in normal production practices. Establishment of controls was attempted each year by identifying 3 fields where growers were asked not to implement all parts of the

OF

T. piniperda

IN PINE FIELDS

789

integrated management program. Managed fields in the program ranged from 1.6 to 16.2 ha, and control fields ranged from 3.2 to 6.1 hectare. Blocks of trees within fields ranged from 100 to 300 trees long and 5 to 10trees wide. Although 2 growers participated both years, no field was in the program for> 1 yr. Fields were surveyed in both 1995 and 1996 to determine the levels ofT. piniperda shoot injury in spring and fall, the percentage of trees with live adult beetles, and grower compliance with the management program requirements. During February and March, 30 trees in each block of each field were examined for up to 3 min or until an injured shoot was found to estimate the amount of shoot feeding by T. piniperda in the previous year. Fifteen trees were sampled at 20-m intervals along transects through the width of a block, = 10 trees and 40 trees from the block edge. After the spring £light of parent beetles, the proportion of trap logs colonized by T. piniperda was estimated by sampling 3 trap logs in the grassy lanes between blocks. In 1995, these logs were collected on 15 April, =1 mo after the onset of the spring adult £light, and brought to a rearing facility. In 1996,logs were dissected in the field during May. Levels of sanitation were assessed in each field to determine the availability of breeding material other than trap logs. Also, in 1996the number of recently cut stumps suitable for T. piniperda colonization was estimated in all fields by randomly selecting 1 row in each block and counting stumps present in the first 100 tree spaces. Heights and numbers of stumps infested with T. piniperda were measured in May for the first 30 stumps encountered. In 1995, stumps were examined only in fields where trap logs were used. Mean height of stumps in each field was compared using a I-tailed t-test to determine if it was