PEST MANAGEMENT AND SAMPLING
Trapping Adult Lissorhoptrus oryzophilus (Coleoptera: Curculionidae) with Aquatic Barrier Traps RAYMOND L. HIX,1 DONN T. JOHNSON,2
AND
JOHN L. BERNHARDT3
Environ. Entomol. 30(4): 770Ð775 (2001)
ABSTRACT The rice water weevil, Lissorhoptrus oryzophilus Kuschel, is a pest of rice in the seven rice producing states of the United States and several countries where it has been introduced. Carbofuran was used adequately for nearly 30 yr, but it is no longer registered for use in rice and has been replaced by -cyhalothrin (Karate) and dißubenzuron (Dimilin). The scouting methods for carbofuran were inadequate or too late in determining the need for insecticide applications of the newer insecticides. Therefore, a new population monitoring method was developed using a double-ended aquatic barrier trap. The trap function was based on weevil swimming behavior, and 16 barrier trap prototypes were tested on adult weevils, 22Ð23 July 1998, in a small bay of late planted rice. Adult means ⫾ SE were 73.9 ⫾ 9.4 per trap on 23 July 1998 and 54.4 ⫾ 6.4 per trap on 24 July 1998. The core sample mean for this plot was 72.9 ⫾ 7.0 larvae per core. The trap was tested in both commercial Þelds and early and late planted small bays in 1999. Larval weevil infestations ranged from moderate (near or slightly below the economic injury level (EIL) to high (62.8 per core). Regression analyses showed a signiÞcant correlation between adults captured in traps to subsequent larval density in Þeld edges (r2 ⫽ 0.98) and Þeld interiors (r2 ⫽ 0.92). For every 1.0 rice water weevil adult captured in barrier traps, a density of 1.2Ð2.8 larvae per core sample was predicted. The aquatic barrier traps caught adult weevils without lures. Trapping rice water weevil adults immediately after permanent ßood in drill-seeded Þelds can be used to aid decision making for insecticide application in rice. KEY WORDS Lissorhoptrus oryzophilus, rice water weevil, aquatic barrier trap, rice integrated pest management, weevil trapping, aquatic intercept trap
THE RICE WATER weevil, Lissorhoptrus oryzophilus Kuschel, is the main insect pest of rice in the United States (Way 1990, Hessler et al. 2000, Stout et al. 2000). This weevil is native to North America (OÕBrien and Wibmer 1982), but has been introduced into Japan, Korea, Taiwan, and China (Nagata 1990). Sexual reproduction occurs in most of its native range, and the weevil is parthenogenetic in California and areas where introduced (Takenouchi 1978). Although adult weevils feed on rice leaves causing characteristic longitudinal scars, larvae may cause economic yield reductions by feeding on the roots. Carbofuran has been used to manage rice water weevil larvae since the early 1970s. However, carbofuran is no longer registered for use in rice. The insecticides -cyhalothrin (Karate) targeted at the adults and dißubenzuron (Dimilin) targeted at the eggs are currently registered. For best results, applications of -cyhalothrin and dißubenzuron should be made within 10 d after a permanent ßood (Bernhardt 1997, 1998, 1999; Stout et al. 2000). The scouting methods used to determine the need for carbofuran application are the leaf scarring method 1 Department of Entomology, University of California, Riverside, CA 92521 (e-mail:
[email protected]). 2 Department of Entomology, University of Arkansas, 321 Agriculture Building, Fayetteville, AR 72701. 3 Rice Research and Extension Center, University of Arkansas, P.O. Box 351, Stuttgart, AR 72160.
which is inadequate and the larval core sampling method which is too late in determining the need of an insecticide application (Tugwell and Stephen 1981, Morgan et al. 1989, Hix et al. 2000). Therefore, a population monitoring method is urgently needed to predict when subsequent larval infestations will exceed the EIL. The EIL based on the cost of treatment with carbofuran was 10 larvae per core (Tugwell and Stephen 1981, Morgan et al. 1989). While evaluating ßoating cone traps on 5 June 1998, we observed that cone traps overturned by severe thunderstorms still caught weevils (Hix et al. 1999). Subsequently, a double-ended aquatic barrier trap was designed based on rice water weevil swimming behavior (Hix et al. 2000a, 2000b). Propulsion during swimming by adult weevils is provided by the mesothoracic legs with the pro- and metathoracic legs serving as diving planes. Furthermore, L. oryzophilus swims below the surface and was observed at depths of 18.0 cm in the laboratory with weevils spending 82.9% of their time swimming between the surface Þlm and a depth of 6.0 cm. The aquatic barrier trap functions as a swimming intercept trap much the same way a Malaise trap functions as a ßight intercept trap. This study evaluated the use of aquatic barrier traps to monitor adult rice water weevils. The objectives of the study were to determine if aquatic barrier traps would consistently catch adult weevils, and how trap means
0046-225X/01/0770Ð0775$02.00/0 䉷 2001 Entomological Society of America
August 2001
HIX ET AL.: TRAPPING ADULT RICE WATER WEEVILS
of adult weevils related to subsequent larval means from core samples. Materials and Methods Each aquatic barrier trap in 1999 was made from two boll weevil trap top assemblies by adding a vertical screen barrier that was 35 long by 11 cm wide and ßotation devices (Hix et al. 2000b). The trap prototypes used in 1998 did not have ßotation devices; therefore, they were clipped to the stake of two metal ßags (Hix et al. 1999). Larval core samples (10 cm diameter by 10 cm deep) consisted of plants, roots, and soil that were taken from rows within 2 m of a trap. Soil was washed from the core samples through a No. 40 sieve. The sieve was subsequently submerged in a saturated salt solution to ßoat the larvae to the surface. All larvae that were found ßoating or remaining in the debris in the bottom of the sieve were removed and counted. The formula EIL ⫽ C/Vxb (Pedigo et al. 1986) was used to calculate a revised EIL for Arkansas where C is the cost of treatment per hectare, V is the value of rough rice per cwt, and b is the yield loss per larva. The cost of treatment with -cyhalothrin was assumed to be $34.59 per hectare and the value of rough rice to be $5.75 per cwt. The yield loss of 0.289 per larva was calculated from the information provided by Tugwell and Stephen (1981). It should be noted that the current rice prices are well below the 10 yr average of $8.04 per cwt (Anonymous 1998). Prototype Test. On 22 July 1998, 16 aquatic barrier traps were placed next to rice plants at ⬇15 m intervals in late-planted ÔLemont,Õ PI312777, and mixed plots (1.1 by 6 m) consisting of alternating rows of the two rice lines. Additionally, 10 gray and 10 Saturn yellow painted ßoating cone traps were positioned at least 15 m apart in this bay and checked at the same time as the barrier traps (Hix et al. 1999). Adult weevils were removed from the traps on 23 and 24 July. At 29 d after ßood, one core sample was taken in eight plots of Lemont and eight plots of PI312777. Two core samples were collected, one from each line, in eight plots of the mixed planting. Procedures for removing larvae from the core samples were described above. Commercial Field Test. In June 1999, 16 traps (eight in Þeld edge and eight in Þeld interior) were placed at least 100 m apart immediately after permanent ßood in three commercial Þelds: (1) 15.4 ha of ÔBengalÕ in St. Francis County, AR, with trap placement on 1 June; (2) 28.3 ha of ÔDrewÕ in Cross County, AR, with trap placement 2 June; and (3) 7.7 ha of Drew in Arkansas County, AR (Foundation Seed Þelds at the RREC Stuttgart, AR) with trap placement on 9 June. Traps were checked daily for 6 d. The edge traps were placed at 100-m intervals in water adjacent to rice plants near levees. The Þeld edges between rice and potential overwintering sites were chosen. The interior traps were placed at 100-m intervals as well, with the nearest trap at least 100 m from the nearest Þeld edge. One larval core sample was taken within 2 m of each trap at 21 and 28 d after permanent ßood. Procedures for
771
larval removal were described above. Analysis of covariance (ANCOVA) was used on data from the three commercial Þelds to determine if there were differences in interior and edge adult trap means, and if there were differences for the larval means in the edges and interiors among the three Þelds (SAS Institute 1996). Reliability Test. Immediately after permanent ßood on 9 June 1999, eight traps were placed in each of four bays (10 by 142 m) of Bengal rice; one trap was placed at each end of the bays and three were placed on the sides of the bays at least 45 m apart. The traps were checked daily for 9 d between 0530 and 2000 hours. Larval core samples were taken at 21 and 26 d postßood and the larvae were removed and counted as previously described. This experiment was repeated in late planted rice consisting of four bays (10 by 142 m) of Bengal planted 17 June 1999. Traps were placed immediately after permanent ßood on 28 July 1999. The traps were checked daily between 0530 and 2000 hours for 9 d postßood. Larval core samples were taken at 21 and 26 d postßood with larvae removed and counted as described above. Both larval and adult voucher specimens were placed in the University of Arkansas Arthropod Museum (Fayetteville, AR). Results Prototype Test. Barrier trap means ⫾ SE were 73.9 ⫾ 9.4 adult weevils per trap on 23 July 1998 and 54.4 ⫾ 6.4 per trap on 24 July 1998, and were superior to the earlier ßoating cone trap design (Fig. 1A). The overall larval mean for this bay was 72.9 ⫾ 7.0 per core (Fig. 2B). Commercial Field Test. The Cross County Þeld had the largest barrier trap means, followed by St. Francis County, and Arkansas County had the least (Fig. 2). Edge and interior trap mean comparisons are reported in Table 1. Larval means in Cross County were above the revised Arkansas EIL based on the cost of treatment with -cyhalothrin of 20.8 larvae per core, St. Francis County was slightly below the revised EIL, and Arkansas County was below the revised EIL (Table 1). ANCOVA of all the commercial Þeld data indicated signiÞcant differences in the slopes (F ⫽ 78.79; df ⫽ 1, 44; P ⬍ 0.0001), but not the intercepts (F ⫽ 3.26; df ⫽ 1, 44; P ⬍ 0.0778) of the lines for the interior and edge traps (Fig. 3). For the values of the slopes b and intercepts a refer to Fig. 3. The differences in interior and edge data imply a regression for the interior data and another for the edge data. Additionally, ANCOVA indicated no signiÞcant differences for the slopes (F ⫽ 1.85; df ⫽ 2, 18; P ⫽ 0.1858) of edge trap data from the three Þelds or the slopes (F ⫽ 2.37; df ⫽ 2, 18; P ⫽ 0.1215) from the interior trap data (Fig. 3). Therefore, one line was calculated for the interior traps (Y ⫽ 0.94X ⫹ 4.2) and another line for the edge traps (Y ⫽ 2.4X ⫹ 4.8) for the three Þelds. Because the intercepts were not signiÞcantly different from each other or zero, the intercepts were forced through the origin (Fig. 4). Regression analyses indi-
772
ENVIRONMENTAL ENTOMOLOGY
Vol. 30, no. 4
Fig. 1. Evaluation of aquatic barrier trap prototypes in late planted rice 22Ð23 July 1998. (A) Adult rice water weevils/trap. Bars ⫽ SE (n ⫽ 16 for barrier traps, n ⫽ 10 cone traps). Bar ⫽ barrier trap, Yel ⫽ yellow ßoating cone trap, Gr ⫽ gray ßoating cone traps. (B) Mean weevil larvae/core sample for each variety and variety mix. Revised Arkansas EIL is 20.8 larvae/core. Bars ⫽ SE (n ⫽ 8). Overall mean for the bay was 72.9 ⫾ 7.0 larvae/core (n ⫽ 32).
cated that edge trap means were strongly correlated with subsequent edge larval infestations (F ⫽ 86.65; df ⫽ 1, 18; P ⬍ 0.0001; r2 ⫽ 0.98), and interior trap means were strongly correlated to subsequent interior larval infestations (F ⫽ 113.48; df ⫽ 1, 18; P ⬍ 0.0001; r2 ⫽ 0.93). Reliability Test. For each sample period, early June and late July 1999, the daily adult mean barrier trap captures and larvae per core among four bays were similar indicating good trap reliability (Fig. 5). For early June, the adult bay means ranged from 9 Ð14 per trap per day. Concurrently, bay larval means ⫾ SE at 21 d after ßood were as follows: Bay I 18.4 ⫾ 1.6, Bay II 20.8 ⫾ 2.6, Bay III 12.9 ⫾ 1.5, and Bay IV 12.2 ⫾ 1.9. A similar trend was observed in late July, but bay adult means were higher ranging from 29 to 37 per trap per day. Bay larval means ⫾ SE at 21 d after ßood were as
Fig. 2. Daily adult rice water weevil trap means for three commercial rice Þelds. Bars ⫽ SE (n ⫽ 8). (A) 28.3 ha of Drew in Cross County, AR. 2Ð8 June 1999. (B) 15.4 ha of Bengal in St. Francis County, AR, 1Ð7 June 1999. (C) 7.7 ha of Drew in Arkansas County, AR, 9Ð15 June 1999.
follows: Bay I 43.0 ⫾ 9.6, Bay II 44.2 ⫾ 3.4, Bay III 37.0 ⫾ 2.0, and Bay IV 41.3 ⫾ 4.2. The larval means per core were about the same among the early June small bays and among the late July small bays. Furthermore, 3,123 adult weevils were caught in the 32 traps over the 9-d period in the early small bay evaluation, and 9,416 weevils were caught in 32 traps over the 9 d period in the late small bay evaluation.
August 2001
HIX ET AL.: TRAPPING ADULT RICE WATER WEEVILS
773
Table 1. Adult rice water weevils per trap per day for interior and edge traps of three commercial fields (n ⴝ 8) and larval means from core samples at 21 and 28 d post flood for commercial fields (n ⴝ 8) County
Adults/trap/day
Cross-Edge Interior St. Francis-Edge Interior Arkansas-Edge Interior
22.0 ⫾ 3.2 23.5 ⫾ 3.6 10.1 ⫾ 1.0 13.5 ⫾ 2.7 6.3 ⫾ 1.8 4.3 ⫾ 0.9
Larvae 21 d
28 d
62.8 ⫾ 6.7 23.1 ⫾ 2.7 15.5 ⫾ 2.2 17.3 ⫾ 2.7 14.4 ⫾ 2.7 9.4 ⫾ 2.1
48.3 ⫾ 5.1 18.6 ⫾ 1.9 12.3 ⫾ 1.8 14.4 ⫾ 2.0 10.0 ⫾ 1.7 6.4 ⫾ 1.4
Shown are means ⫾ SE.
Discussion The aquatic barrier traps intercepted swimming adult weevils without chemical lures or baits, and were superior in catching adults than were the older ßoating cone traps (Hix et al. 1998, 1999, 2000) designed to be used with plant lures or possible weevil pheromones. A linear relationship was shown between the number of adult rice water weevils caught in the barrier traps and the subsequent number of larvae in core samples at 21 d after ßood. Data indicated that 1.0 adult weevil per trap subsequently resulted in between 1.2 (interior) and 2.8 (edge) larvae per core in the three commercial Þelds (Fig. 4). ANCOVA indicated a different line for the interior and edge trap data for the three Þelds. However, the numbers of adult weevils caught in the edge and interior traps were similar (Table 1). Therefore, it may be possible to use interior regressions to develop economic thresholds while using edge trap data to determine if adult populations exceed the threshold. A hypothetical economic threshold of 17.3 adults per trap per day was calculated from the commercial Þeld data of this study using the interior regression. There are two weaknesses in the revised EIL. First, the original EIL reported by Tugwell and Stephen (1981) was based on ÔStarbonnet,Õ and the action thresholds were reevaluated on ÔLebonnet,Õ ÔMarsÕ and Starbonnet (Morgan et al. 1989). Secondly, the current rough rice prices are at a record low. For example, if the revised EIL was based on the 10 yr average of $8.04 per cwt, it would be 14.9 larvae per core reducing the hypothetical economic threshold to 12.4 adults per trap per day. Nonetheless, the original EIL of Tugwell and Stephen (1981) has remained the benchmark for Arkansas. Similar numbers of adults were caught in nearly identical bays in the early small rice bay evaluations and again in the late small bay evaluations which substantiated barrier trap reliability (Fig. 5). Data from these studies suggest that aquatic barrier traps could be used to determine the need for insecticidal treatment against adult rice water weevils within 10 d after permanent ßood in drill-seeded rice. The aquatic barrier trap could have utility in other rice producing states, but the relationships between trap catches and larval numbers would need to be established. For example, females lay ⬇40 eggs in
Fig. 3. Regressions of interior (open circles) and edge (solid circles) adult rice water weevil trap data on Þeld by Þeld basis for three commercial Þelds. (A) Cross County, AR, 2Ð8 June 1999, b ⫽ 0.71 for interior and b ⫽ 1.8 for edge (interior r2 ⫽ 0.88; edge r2 ⫽ 0.92). (B) St. Francis County, AR, 1Ð7 June 1999, b ⫽ 0.97 for interior and b ⫽ 2.0 for edge (interior r2 ⫽ 0.91; edge r2 ⫽ 0.91). (C) Arkansas County, AR, 9Ð15 June 1999, b ⫽ 1.1 for interior and b ⫽ 3.1 for edge (interior r2 ⫽ 0.83; edge r2 ⫽ 0.97). ANCOVA indicated no signiÞcant difference for the slopes of interior traps (F ⫽ 2.38; df ⫽ 1, 18; P ⫽ 0.1215) among the three Þelds or the edge traps (F ⫽ 1.85; df ⫽ 1, 18; P ⫽ 0.18.58).
774
ENVIRONMENTAL ENTOMOLOGY
Vol. 30, no. 4
Fig. 4. Regressions for combined interior rice water weevil trap data (open circles; r2 ⫽ 0.93; b ⫽ 1.2) and combined edge weevil trap data (solid circles; r2 ⫽ 0.98; b ⫽ 2.4) for three commercial rice Þelds were strongly correlated. ANCOVA indicated that the slopes (F ⫽ 78.80; df ⫽ 1, 44; P ⬍ 0.0001) were signiÞcantly different and intercepts (F ⫽ 3.26; df ⫽ 1, 44; P ⬍ 0.0778) were not signiÞcantly different for the interior and edge trap data.
Arkansas (J.L.B., unpublished data), and parthenogenetic females in California lay ⬇200 eggs (Grigarick and Washino, 1993). Therefore, the relationship of the number of adults captured per trap and subsequent larval infestations would likely be different. Different cultural practices will likely require different procedure to use the barrier traps. The two major planting methods used in the United States are dry- (drill or broadcast) and water-seeded (pin-point or continuous ßood) (Helms, 1996). Water seeding involves the use of aircraft to apply seed to preßooded Þelds with pregerminated or dry rice seed. Drill seeding is the most common method in Arkansas. Water seeding is primarily the method used in California, and both methods are used in Louisiana. Additional grower Þeld studies need to be conducted to increase accuracy of the economic thresholds of adult captures per aquatic barrier trap and to develop a trap index like that for the boll weevil trap (Rummel et al. 1980). Economic thresholds may vary among rice varieties and would require further assessments of yield impact for several major varieties subjected to different larval densities. Studies are needed to assess yield impact from larval root pruning for the most commonly planted rice varieties. Overall, the aquatic barrier trap may aid in decision-making for pesticide use against adult rice water weevils. Acknowledgments We thank Barbara Lewis, Dan Baxter, Tahir Rashid, Clay McDaniel, Edward Gbur (statistician), Bond Farms, Shelby
Fig. 5. Small plot (10 by 142 m) evaluations of aquatic barrier trap in Bengal rice. Bars denote SE (n ⫽ 8) for daily adult rice water weevil means/trap. (A) Eight traps/bay were placed 9 June 1999 and monitored daily for 9 d postßood. (B) Traps placed 28 July 1999 monitored daily for 9 d postßood. Wright Farms, and Crossed-I Farms. We also thank the Arkansas Rice Research and Promotion board for funding this project.
References Cited Anonymous. 1998. Arkansas agricultural statistics. Univ. Ark. Coop. Ext. Serv. MP-409. Bernhardt, J. L. 1997. Control of rice water weevil with Karate 1996. Arthropod Manage. Tests 22: 289. Bernhardt, J. L. 1998. Control of rice water weevil with Dimilin,1997. Arthropod Manage. Tests 23: 258. Bernhardt, J. L. 1999. Control of rice water weevil with selected insecticides, 1998. Arthropod Manage. Tests 24: 270 Ð271. Grigarick, A. A., and R. K. Washino. 1993. Invertebrates. In M. L. Flint (ed.), Integrated pest management. Univ. Calif. Statewide IPM Proj. Div. Agric. Nat. Res. Publ. 3280. Helms, R. S. 1996. Rice production handbook. Univ. Ark. Coop. Ext. Serv. MP-192. Hessler, L. S., M. J. Oraze, A. A. Grigarick, and A. T. Palrang. 2000. Numbers of rice water weevil larvae (Coleoptera: Curculionidae) and rice plant growth relation to adult
August 2001
HIX ET AL.: TRAPPING ADULT RICE WATER WEEVILS
infestation levels and broadleaf herbicide applications. J. Agric. Urban Entomol. 17: 99 Ð108. Hix, R. L., D. T. Johnson, J. L. Bernhardt, T. L. Lavy, J. D. Mattice, and B. L. Lewis. 1998. Development of an IPM Monitoring Program for Rice Water Weevil Adults. In Rice research studies 1997. Univ. Ark. Agric. Exp. Stat. Res. Ser. 460: 95Ð102. Hix, R. L., D. T. Johnson, J. L. Bernhardt, J. D. Mattice, and B. L. Lewis. 1999. Trapping adult rice water weevils with ßoating cone and barrier traps. In Rice research studies 1998. Univ. Ark. Agric. Exp. Stat. Res. Ser. 468: 135Ð141. Hix, R. L., D. T. Johnson, and J. L. Bernhardt. 2000a. Swimming behavior of an aquatic weevil, Lissorhoptrus oryzophilus (Coleoptera: Curculionidae). Fla. Entomol. 83: 316 Ð324. Hix, R. L., D. T. Johnson, and J. L. Bernhardt. 2000b. An aquatic barrier trap for monitoring adult rice water weevils. Fla. Entomol. 83: 189 Ð192. Hix, R. L., D. T. Johnson, J. L. Bernhardt, J. D. Mattice, and B. L. Lewis. 2000. Trapping adult rice water weevils with aquatic barrier traps and ßoating cone traps with notes on chemical ecology. In Rice research studies 1999. Univ. Ark. Agric. Exp. Stat. Res. Ser. 476: 147Ð157. Morgan, D. R., N. P. Tugwell, and J. L. Bernhardt. 1989. Early rice Þeld drainage for control of rice water weevil (Coleoptera: Curculionidae) and evaluation of an action threshold based upon leaf-feeding scars of adults. J. Econ. Entomol. 82: 1757Ð1759. Nagata, T. 1990. JapanÕs unwelcome new arrival. Shell-Agri. 8: 8 Ð10.
775
O’Brien, C. W., and G. J. Wibmer. 1982. Annotated checklist of the weevils (Curculionidae sensu lato) of North America, Central America, and the West Indies (Coleoptera: Curculionoidea). Mem. Am. Entomol. Inst. 34. Pedigo, L. P., S. H. Hutchins, and L. G. Higley. 1986. Economic injury levels in theory and practice. Annu. Rev. Entomol. 31: 341Ð368. Rummel, D. R., J. R. White, S. C. Carrol, and G. R. Pruitt. 1980. Pheromone trap index system for predicting need for overwintered boll weevil control. J. Econ. Entomol. 73: 806 Ð 810. SAS Institute. 1996. A guide to statistics and data analysis using JMP and JMP IN software. Duxbury, Belmont, CA. Stout, M. J., W. C. Rice, R. M. Riggio, and D. R. Ring. 2000. The effects of four insecticides on the population dynamics of the rice water weevil, Lissorhoptrus oryzophilus Kuschel. J. Entomol. Sci. 35: 48 Ð 61. Takenouchi, Y. 1978. A chromosome study of the parthenogenetic rice water weevil, Lissorhoptrus oryzophilus Kuschel (Coleoptera: Curculionidae), in Japan. Experimentia 34: 444 Ð 445. Tugwell, N. P., and F. M. Stephen. 1981. Rice water weevil seasonal abundance, economic levels, and sequential sampling plans. Univ. Ark. Agric. Exp. Stn. Bull. 849. Way, M. O. 1990. Insect pest management in rice in the United States. In B. T. Grayson, M. B. Green, and L. G. Copping (eds.), Pest management in rice. Elsevier, Barking, UK. Received for publication 3 November 2000; accepted 22 May 2001.
