Translator Disclaimer
1 March 2009 Acoustic Characteristics of Dynastid Beetle Stridulations
Author Affiliations +
Abstract

Oryctes rhinoceros (L.) (Coleoptera: Scarabaeidae: Dynastinae) causes economically important damage to ornamental and commercial coconut palm trees in the western Pacific region that could be mitigated significantly by early detection and treatment. Adults are difficult to detect visually because they attack the crowns of the trees and feed internally before mating and dispersing to new hosts. Visual inspection is nevertheless the most widely used detection method, augmented with pheromone traps. This species is an ideal candidate for acoustic detection because the adults are large, active borers that produce stridulations during courtship and mating. The stridulations have distinct, easily recognized temporal patterns. Larvae and pupae also produce stridulations. To assist in development of new detection methods, acoustic characteristics of the adult and larval stridulations have been quantified and compared with stridulations produced by larvae of other dynastids recorded in the western Pacific region.

Stridulation has been identified in many coleopteran families (Wessel 2006) but relatively more is known about the morphology of the stridulatory organs in most of these species than about the acoustic characteristics and behavioral roles of the signals. It has long been known, for example, that adults, pupae, and larvae of the coconut rhinoceros beetle, Oryctes rhinoceros (L.), produce audible stridulations (Darwin 1871; Gressitt 1953). Mini & Prabhu (1990) described the stridulatory structures of adult O. rhinoceros and observed males that stridulated during courtship and mating, as well as during aggressive encounters with other males. It was not specified whether the stridulations produced in the different contexts were similar or different. Mated females stridulated occasionally, usually during interactions with males, but the function of these stridulations was unknown. Larvae and pupae of other beetle species also produce stridulations in response to disturbance (Palestrini et al. 1990), and stridulations have been shown to repel predators (Masters 1979; Hill 2007). However, the acoustic characteristics and much of the behavioral role of stridulations in O. rhinoceros and other beetles remain to be analyzed.

Because O. rhinoceros is a large insect that causes extensive damage to coconut (Cocos nucifera L.) and other palms (Gressitt 1953; Hinckley 1973; Bedford 1980), acoustic characterization of its stridulations is of considerable practical use above and beyond any behavioral, physiological, or evolutionary considerations. Adults bore into and feed on the folded emerging fronds at the crowns of the palm trees, and frequent attacks by several adults eventually kill the tree (Gressitt 1953). It would be ideal to remove the insects, treat with pesticide or viruses, or cut down a tree soon after the first attacker enters the tree (Jackson & Klein 2006), but beetle attacks cannot be easily detected from the ground until well after significant damage has occurred. Visual inspection of frond and trunk damage is nevertheless the most commonly method of detection, augmented by traps baited with aggregation pheromone (Hallett et al. 1995; Ramle et al. 2005). Acoustic techniques have potential to detect adults that are moving and feeding in the crowns, as well as larvae that feed in standing dead trunks of previously killed palms (Mankin et al. 2000, 2002, 2008a, b). The reliability of acoustic methods improves considerably when stridulations and other behaviors can be detected and distinguished from background noise by automated digital signal processing tools. Here we describe stridulations recorded during acoustic detection studies conducted in Guam and Australia to provide basic information about stridulation behavior in O. rhinoceros and other dynastids, and also to assist efforts to detect O. rhinoceros in coconut trees in urban Guam and other areas with high background noise.

MATERIALS AND METHODS

Insects and Recording Procedures

In laboratory studies, O. rhinoceros adults and larvae were obtained from a rearing facility at the University of Guam. Recordings were done on different days from 3 separate groups of 10, 3, and 5 adults, respectively, held in different 30 × 24 × 24-cm plastic rearing cages containing short pieces of coconut palm trunk and fronds, and on 3 separate days from multiple larvae (numbers unknown) in other cages containing decaying pieces of palm trunk and detritus. Temperatures were ca. 24–26°C. Field re-cordings of larval activity were obtained from dead coconut palm trees in the Tumon Bay hotel district of Guam at temperatures of 28–31°C (stridulations were detected at 2 of the 3 sites where larvae were recovered). The numbers of larvae recovered from the trees after recordings were obtained ranged from 3 to >50. In Australia, recordings of dynastids other than O. rhinoceros were obtained from sugarcane fields near the BSES laboratories in Mackay and Bundaberg, Queensland. Temperatures were 27–32°C.

Fig. 1.

Two series of 4 chirps recorded from a microphone placed over a plastic rearing cage holding 10 adult O. rhinoceros. Positions of chirps within a series are noted as first, intermediate, and last.

f01_123.eps

Fig. 2.

Chirp recorded from an accelerometer attached to a second rearing cage holding 5 adults. As in most chirps, the impulses were variable in amplitude and temporal pattern, but generally highest in amplitude at the beginning and end. The protraction and retraction phases of the chirp were separated by a brief pause.

f02_123.eps

Adult- and larval-produced vibrations were recorded by inserting a 30-cm-long steel probe into the substrate, usually a trunk or a section of a log, and magnetically attaching an accelerometer (model 4370, Brüel and Kjær [B & K], Nærum Denmark) connected to a charge amplifier (model 2365, B & K), as in Mankin et al. (2000, 2001, 2002). For detection of white grubs in Australian sugarcane fields, probes were inserted into the soil next to a sugarcane stool. Signals were stored on a digital audio recorder (model HD-P2, Tascam, Montebello, CA) at a 44.1 kHz-, 24-bit sampling rate and subsequently copied to a computer.

Adult stridulations of O. rhinoceros also were recorded from a rearing cage containing 10 adults by setting a MP3 recorder with an internal microphone on top of the cage.

Signal Processing

Digitized signals were analyzed with Raven 1.2 software (Cornell Lab of Ornithology, Ithaca, NY) and other dedicated signal sampling software developed in Matlab (MathWorks, Inc., Natick, MA), or with the customized software program, DAVIS (Mankin 1994; Mankin et al. 2000, 2008b). Signals were band-pass filtered between 0.5 and 5 kHz to eliminate low- and high-frequency background noise. Fast Fourier Transforms for spectrogram displays were calculated on 64-point (1.5 ms-) time-slices of the waveforms in a Hamming window, and spectrograms were calculated by sections with 90 per cent overlap. Transforms for spectral comparisons were calculated on 512-point (11.6 ms-) time-slices, and all spectrum levels were referenced to the maximum acceleration, Amax, in the 0.5–5 kHz reference range, i.e., dB = 20 log10(A/Amax) (Mankin & Benshemesh 2006).

Fig. 3.

Two consecutive impulses sampled from chirp in previous figure, showing typical variations of shape and duration.

f03_123.eps

Table 1

Mean durations and standard errors of the protraction, pause, and retraction phases of the first, intermediate, and last chirp in series recorded from caged adult O. rhinoceros

t01_123.gif

Table 2

Mean and standard errors of numbers of impulses in the protraction, retraction, and total phases of the first, intermediate, and last chirp in series recorded from caged adult O. rhinoceros.

t02_123.gif

Fig. 4.

Spectra of 0.1-s signals in chirps from 3 different adults in rearing cage containing various pieces of coconut palm tree trunks. The vertical axis indicates spectrum level relative to the highest level between 500 and 5000 Hz and the horizontal axis indicates frequency.

f04_123.eps

The DAVIS program identified and timed groups (trains) of impulses with interpulse intervals less than a preset duration, Ii, storing the beginning and end times of these trains in a spreadsheet along with the times of individual impulses and the numbers of impulses in each train. Impulses that failed to match 1 or more specified insect sound profiles (Mankin et al. 2000, 2007, 2008b) were discarded. The beginning of a train was set as the beginning of the first impulse after a period where the interpulse interval was > Ii, and the end was set as the end of the last impulse whose interpulse interval was < Ii. For this study, we set Ii 250 ms, which resulted in trains that closely matched the timing of distinct sounds identified when the recordings were played back.

Fig. 5.

Oscillogram (A) and spectrogram (B) of a chirp recorded from an O. rhinoceros larva in a palm tree trunk. The first 5 impulses in the chirp are marked by arrows. Darker shading in spectrogram indicates higher relative spectrum level.

f05_123.eps

Comparisons among temporal and signal patterns of trains and impulses were conducted with nonparametric (NPAR1WAY) analyses in SAS (2004).

RESULTS

Adult Stridulations

Adult O. rhinoceros in rearing cages produced frequent, audible stridulations in multiple series of 3 to 6 (most frequently 4) chirps separated by variable intervals (Fig. 1). Individual chirps within a series contained a 3-phase sequence of protraction, pause, and retraction. The protraction phase (Fig. 2) contained high-amplitude, irregularly-timed impulses. As the protraction phase proceeded, the impulse amplitudes usually decreased. After a brief pause, the impulses then increased in amplitude until the end of the chirp (retraction phase, Fig. 2). In the example of Fig. 2, the rise in amplitude at the end of the chirp is not as pronounced as in most of the chirps in Fig. 1, and considerable variation was observed in amplitudes overall. Individual impulses within a chirp were variable in shape as well as amplitude, and ranged typically from 1 to 4 ms in duration (Fig. 3).

As observed in other studies of insect vibrations in different substrates (Markl 1983; Hirschberger 2001; Mankin et al. 2008a, b; Joyce et al. 2008), the spectral characteristics of the signals recorded in the rearing cages depended on the resonance properties of the palm tree trunks or other substrates in which the adults were moving and feeding. Stridulations from males recorded on different substrates in the cage were distinctively different to the human ear and also had distinct frequency spectra, examples of which are shown in Fig. 4. These differences enabled listeners (and possibly the beetles themselves) to establish that different adults were stridulating at different times. It should be noted, however, that this is not a totally unambiguous method of identifying different individuals. Two adults stridulating within the same substrate might not produce distinctively different spectra. Also, they may not be distinguishable on the basis of chirp impulse rate. The 3 different adults whose stridulation spectra are shown in Fig. 4, for example, produced impulses at intervals whose medians were not significantly different under the Kruskal-Wallis Htest (χ2 = 2.71, P = 0.25, df = 2). The means and standard errors of the interpulse intervals of the second chirp (i.e., the first intermediate chirp) in the first series recorded from each adult were 11.2 ± 0.8 (n = 32), 11.9 ±0.8 (n = 30), and 10.6 ±1.3 ms (n = 29).

Fig. 6.

Oscillograms of 2 short chirps (A) and a longer chirp (B) recorded from multiple O. rhinoceros larvae in a rearing cage. The chirps are embedded in other larval movement and feeding signals of similar amplitude.

f06_123.eps

To consider the range of chirp durations and interpulse intervals that can be expected in different O. rhinoceros recordings, we examined chirps collected over approximately 1 h from 3 different rearing cages containing a total of 18 adults. When possible, we selected signals from separate adults, based on audible and spectral differences between chirps evaluated by playback of the recordings in the Raven signal analysis program. Comparisons of the durations of chirps in different series suggested that the durations of the first and subsequent chirps in a series were variable (Table 1, Fig. 1). However, the difference was statistically significant only for the protraction phase of the chirp. As with the durations, there were statistically significant differences in the mean numbers of impulses occurring in the protraction phases of the first and subsequent chirps in a series (Table 2). The mean interval between chirps within a series was 77.2 ±6.0 ms, similar to the mean pause between the protraction and retraction phases of the chirps.

Fig. 7.

Short, unpatterned chirp recorded from a dynastid larva in sugarcane field near Bundaberg, Australia.

f07_123.eps

Larval Stridulations

Stridulations from O. rhinoceros larvae were detected in field recordings from palm trees (Fig. 5) and in 1 of 6 recordings from rearing cages containing multiple larvae feeding on pieces of palm tree fronds and trunk (Fig. 6). Unlike the chirps produced by adults, the larval chirps did not occur in a series and had no discernible pattern of amplitudes. The chirp durations were highly variable, and the amplitudes were lower than for adults.

The O. rhinoceros signals were similar in many respects to larval stridulations of other dynastids recorded in a recent survey conducted in Australia to detect root-feeding melolonthines (Agnew 1997; Miller & Allsopp 2000) in sugarcane. Stridulations were recorded at a site near Bundaberg, AU (Fig. 7), from which a dynastid larva was excavated. Stridulations also were recorded at a site near Mackay, AU (Figs. 8 and 9), from which a larva subsequently identified as Dasygnathus sp. was recovered along with several nonstridulating invertebrates, including 2 earthworms, a cockroach, a tenebrionid larva, and a larva of an economically important root-feeding melolonthine scarab, Dermolepida albohirtum (Waterhouse). Some of the chirps (Figs. 8 and 9A) had an amplitude pattern similar to that seen for adult O. rhinoceros in Fig. 1. Other chirps appeared to be without pattern (Fig 9B) similar to those detected from O. rhinoceros larvae (Figs. 5 and 6). The mean interpulse in a randomly sampled Dynastinae chirp recorded at Mackay was 4.0 ±0.4 ms (n = 27), and was similar to the mean interpulse interval of a chirp sampled at random from a cage with O. rhinoceros larvae in Guam (3.44 ±0.4 ms, n = 16). The median interpulse intervals were not different in a Mann Whitney U test (U = 299, Z = -1.38, P = 0.17). In general, the amplitudes of impulses in chirps were smaller and the intervals between impulses were shorter for the Australian dynastid larvae and the O. rhinoceros larvae than for the adult O. rhinoceros.

Fig. 8.

Sample series of chirps recorded from a Dasygnathus sp. rhinoceros beetle larva in a sugarcane field near Mackay, Australia.

f08_123.eps

DISCUSSION

The occurrence in this study of several periods when a series of chirps produced by 1 adult was followed immediately by series produced by other adults supports previous hypotheses that stridulation may play an important role in intraspecific communication of this and other scarabaeoids in hidden environments (Mini & Prabhu 1990; Hirschberger 2001; Carisio et al. 2004; Kasper & Hirschberger 2006). Also, the similarities of disturbance stridulations among rhinoceros beetle larvae in Guam and Australia suggest that similarities may be found as well among stridulations produced by other dynastid larvae. It is likely that O. rhinoceros stridulations will remain of interest in future basic studies of insect communication behavior.

In addition, the relative uniformity of interpulse intervals and chirp amplitude patterns observed during stridulatory behavior may be of practical interest in future acoustic monitoring of O. rhinoceros in coconut palm trees. If stridulations similar to those recorded here from adults were detected at the base of a live tree, it could be predicted that adults (most likely males, Mini & Prabhu 1990) were present in the crown. Similarly, if stridulations similar to those recorded here from larvae were detected at the base of a standing dead tree, it would predict that larvae were present in the rotten part of the upper trunk. However, the larval stridulations would not be expected to occur as frequently or as loudly as with adults in the crown of a live tree. Either adult or larval stridulations would be easier to detect than unpatterned movement and feeding sounds because the occurrence of a pattern in the target signal enables unpatterned nontarget signals to be averaged out over time (Mankin et al. 2000, 2008b).

Fig. 9A.

Expanded version of a single chirp from Figs. 8 and 9B. Sample chirp of a different pattern appearing at a later time in recording from same site in a sugarcane field near Mackay, Australia.

f09a_123.eps

We noted with interest the numbers of impulses that occurred within adult O. rhinoceros chirps and the durations of intervals between chirps within series. In previous acoustic studies of larvae of the red palm weevil, Rhynchophorus ferrugineua (Olivier) (Mankin et al. 2008a) and the Asian longhorned beetle, Anoplophora glabripennis (Motschulsky) (Mankin et al. 2008b), it had been determined that particular classes of impulse trains, denoted as bursts, were good predictors of the presence of insect infestation at a recording site. For R. ferrugineus, the bursts were defined as trains with >6, but <200 impulses, separated by intervals <500 ms. For A. glabripennis, bursts were defined as trains with >6, but <50 impulses, separated by intervals <250 ms. For this study, the results in Table 2 suggest that a series of chirps could be defined as a burst, in which case the minimum number of impulses could be as few as 6, as in the examples above, and the maximum could be the number of impulses in a series of 5 chirps of 40 impulses, or ca. 200 impulses, as with R. ferrugineus. In this case, and perhaps also with R. ferrugineus and A. glabripennis, the acoustic feature defined as a burst may be interpretable as a sound produced by stereotyped behaviors such as abdominal movements or the movement of legs or other body parts.

The stridulating dynastids detected in sugarcane fields near Bundaberg and Mackay were not considered pests because they fed primarily on decaying vegetation and detritus. At the Mackay site, however, the dynastid was present with a root-feeding larval pest, D. albohirtum. To unambiguously identify both insects in this case would require knowledge also of a signal feature that is associated with the presence of D. albohirtum, perhaps a specific behavior performed only by this pest. Indeed, a grand challenge of acoustic detection surveys is to identify and characterize multiple features of pest and nonpest signals, like the dynastid larval stridulations, or previously described termite head-bangs and ant stridulations (Connétable et al. 1999; Rauth & Vinson 2006; Mankin & Benshemesh 2006), and use these features to identify and discriminate target pests from nontarget organisms.

ACKNOWLEDGMENTS

The use of trade, firm, or corporation names in this publication does not constitute an official endorsement or approval by the United States Department of Agriculture, Agricultural Research Service of any product or service to the exclusion of others that may be suitable. Robert Bourgeois (University of Guam), Roland Quitugua (Guam Coconut Rhinoceros Beetle Eradication Project Operations Chief), Tim Staier and Ben Allen (BSES), and Everett Foreman and Betty Weaver (ARS) provided technical support. Financial support was provided in part by the Sugar Research and Development Corporation of Australia, BSES Limited, and a Western Integrated Pest Management Center Special Issues Grant.

REFERENCES CITED

1.

J. R. Agnew 1997. Australian sugarcane pests. Bureau of Sugar Experiment Stations, Indooroopilly, AU. Google Scholar

2.

G. O. Bedford 1980. Biology, ecology, and control of palm rhinoceros beetles. Annu. Rev. Entomol. 26: 309–339. Google Scholar

3.

L. Carisio , C. Palestrini , and A. Rolando 2004. Stridulation variability and morphology: an examination in dung beetles of the genus Trypocopris (Coleoptera: Geotrupidae). Popul. Ecol. 46: 27–37. Google Scholar

4.

S. Connétable , A Robert,. F. Bouffault , and C. Bordereau 1999. Vibratory alarm signals in two sympatric higher termite species: Pseudacanthotermes spiniger and P. militaris (Termitidae, Macrotermitinae). J. Insect. Behav. 12: 329–342. Google Scholar

5.

C. Darwin 1871. The ascent of man, and selection in relation to sex. Vol. 1. John Murray, London. Google Scholar

6.

J. L. Gressitt 1953. The coconut rhinoceros beetle (Oryctes rhinoceros) with particular reference to the Palau Islands. Bulletin 212, Bishop Museum, Honolulu. Google Scholar

7.

R. H. Hallett , A. L. Perez , G. Gries , R. Gries , H. D. Pierce Jr. , J. Yue , A. C. Oehlschlager , L. M. Gonzalez , and J. H. Borden 1995. Aggregation pheromone of coconut rhinoceros beetle, Oryctes rhinoceros (L.) (Coleoptera: Scarabaeidae). J. Chem, Ecol. 21: 1549–1570. Google Scholar

8.

S. A. Hill 2007. Sound generation in (Mantodea: Mantidae): stridulatory structures and acoustic signals. J. Orthopt. Res. 16: 35–49. Google Scholar

9.

A. D. Hinckley 1973. Ecology of the coconut rhinoceros beetle, Oryctes rhinoceros (L.) Coleoptera: Dynastidae). Biotropica 5: 111–116. Google Scholar

10.

P. Hirschberoer 2001. Stridulation in Aphodius dung beetles: Behavioral context and intraspecific variability of song patterns in Aphodius ater (Scarebaeidae). J Insect Behav. 14: 69–88. Google Scholar

11.

T. A. Jackson , and M. G. Klein 2006. Scarabs as pests: A continuing problem. Coleopterists Bull. 60: 102–119. Google Scholar

12.

A. L. Joyce , R. E. Hunt , J. S. Bernal , and S. B. Vinson 2008. Substrate influences mating success and transmission of courtship vibrations for the parasitoid Cotesia marginiventris. Ent. Exp. Appl. 127: 39–47. Google Scholar

13.

J. Kasper, and P. Hirschberger 2006. Vibratory communication in dung beetles (Scarabaeidae, Coleoptera), pp. 405–419 In S. Drosopoulos, and M. F. Claridge [eds.], Insect sounds and communication: physiology, behaviour, ecology, and evolution. Taylor and Francis, Boca Raton, FL. Google Scholar

14.

R. W. Mankin 1994. Acoustical detection of Aedes taeniorhynchus swarms and emergence exoduses in remote salt marshes. J. American Mosq. Cont. Assoc. 10: 302–308. Google Scholar

15.

R. W. Mankin , and J. Benshemesh 2006. Geophone detection of subterranean termite and ant activity. J. Econ. Entomol. 99: 244–250. Google Scholar

16.

R. W. Mankin , J. Brandhorst-Hubbard , K. L. Flanders , M. Zhang , R. L. Crocker , S. L. Lapointe , C. W. Mccoy , J. R. Fisher , and D. K. Weaver 2000. Eavesdropping on insects hidden in soil and interior structures of plants. J. Econ. Entomol. 93: 1173–1182. Google Scholar

17.

R. W. Mankin , J. L. Hubbard , and K. L. Flanders 2007. Acoustic indicators for mapping infestation probabilities of soil invertebrates. J. Econ. Entomol. 100: 790–800. Google Scholar

18.

R. W. Mankin , S. L. Lapointe , and R. A. Franqui 2001. Acoustic surveying of subterranean insect populations in citrus groves. J. Econ. Entomol. 94: 853–859. Google Scholar

19.

R. W. Mankin, A. Mizrach, A. Hetzroni, S. Levsky, Y. Nakache, and V. Soroker 2008a. Temporal and spectral features of sounds of wood-boring beetle larvae: identifiable patterns of activity enable improved discrimination from background noise. Florida Entomologist 91: 241–248. Google Scholar

20.

R. W. Mankin , W. L. Osbrink , F. M. Oi , and J. B. Anderson 2002. Acoustic detection of termite infestations in urban trees. J. Econ. Entomol. 95: 981–988. Google Scholar

21.

R. W. Mankin, M. T. Smith, J. M. Tropp, E. B. Atkinson, and D. Y. Jong 2008b. Detection of Anoplophora glabripennis (Coleoptera: Cerambycidae) larvae in different host trees and tissues by automated analyses of sound-impulse frequency and temporal patterns. J. Econ. Entomol. 101: 836–849. Google Scholar

22.

H. Markl 1983. Vibrational communication, pp. 332–353, In F. Huber and H. Markl [eds.], Neuroethology and Behavioral Physiology, Springer-Verlag, Berlin, Germany. Google Scholar

23.

W. M. Masters 1979. Insect disturbance stridulation. Its defensive role. Behav. Ecol. Sociobiol. 5: 187–200. Google Scholar

24.

L. J. Miller , and P. G. Allsopp 2000. Identification of Australian canegrubs (Coleoptera: Scarabaeidae: Melonthini). Invert. Taxon. 14: 377–409. Google Scholar

25.

A. Mini , and V. K. K. Prabhu 1990. Stridulation in the coconut rhinoceros beetle Oryctes rhinoceros (Coleoptera: Scarabaeidae). Proc. Indian Acad. Sci.., Animal Sciences 99: 447–455. Google Scholar

26.

Palestrini C. , M. Zunino , and M. Zucchelli 1990. Sound production in the larvae of Geotrupes spiniger (Marsham) (Coleoptera: Geotrupidae). Bioacoustics 2: 209–216. Google Scholar

27.

M. Ramle , M. B. Wahid , K. Norman , T. R. Glare , and T. A. Jackson 2005. The incidence and use of Oryctes virus for control of rhinoceros beetle in oil palm plantations in Malaysia. J. Inv. Pathol. 89: 85–90. Google Scholar

28.

S. J. Rauth , and S. B. Vinson 2006. Colony wide behavioral contexts of stridulation in imported fire ants (Solenopsis invicta Buren). J. Insect Behav. 19: 293–304. Google Scholar

29.

SAS Institute. 2004. SAS/STAT 9.1 Users guide. SAS Institute Inc., Carey, NC. Google Scholar

30.

A. Wessel 2006. Stridulation in the Coleoptera-an overview, pp. 397–403 In S. Drosopoulos and M. F. Claridge [eds.], Insect Sounds and Communication: Physiology, Behaviour, Ecology, and Evolution. Taylor and Francis, Boca Raton, FL. Google Scholar
R. W. Mankin, A. Moore, P. R. Samson, and K. J. Chandler "Acoustic Characteristics of Dynastid Beetle Stridulations," Florida Entomologist 92(1), 123-133, (1 March 2009). https://doi.org/10.1653/024.092.0119
Published: 1 March 2009
JOURNAL ARTICLE
11 PAGES


SHARE
ARTICLE IMPACT
Back to Top