Using a Towed Array to Survey Red Drum Spawning Sites in the Gulf of Mexico

Scott A Holt

University of Texas at Austin Marine Science Institute
750 Channel View Drive, Port Aransas TX 78373 USA.
sholt@utmsi.utexas.edu

Introduction

The red drum (Sciaenops ocellatus) is an important recreational and, in some locations, commercial species throughout its range. Juveniles generally live in estuaries and move to nearshore oceanic waters as they reach maturity (Pearson 1929). Adults range widely over the nearshore continental shelf waters throughout the year but apparently move to coastal waters to spawn (Overstreet 1983). Spawning is generally thought to take place in coastal waters near inlets (Jannke 1971, Holt et al. 1985) although Lyczkowski-Shultz et al. (1988) found eggs and larvae out to 34 km from shore in the eastern Gulf of Mexico. There is also evidence of limited spawning activity within estuaries in Florida (Murphy and Taylor 1990, Johnson and Funicelli 1991) and in North Carolina (Luczkovich et al. 1999).

The location of spawning areas has typically been inferred through capture of fish with mature gonads or the distribution of eggs and larvae. Red drum make loud, characteristic sounds during spawning (Guest and Lasswell 1978). Listening for the characteristic sound production has recently been used to locate red drum spawning sites in Indian River Lagoon, Florida (Johnson and Funicelli 1991), and in Plamico Sound, North Carolina (Luczkovich et al. 1999), and at tidal inlets in South Carolina (personal communication, M. R. Collins, South Carolina Department of Natural Resources). These surveys have been done with both hand-held hydrophones and remotely placed sonobuoys.

Over a four-year period from 1998-2001, a hydrophone mounted on a pier in the Aransas Pass, Texas tidal inlet has been use to record sounds of red drum spawning activity every evening during the September through October spawning period. Recordings were made for 20 s every 15 m from 1700 to 0100 hours and spanned the 4-5 hour evening spawning period of red drum (Holt et al. 1985). Red drum produced characteristic spawning sounds from about one hour before sunset to about three hours after sunset with the most intense activity occurring during the two hours following sunset (S. Holt, unpublished data). These data, along with collections of red drum eggs and larvae at the site, confirmed that red drum spawn actively in the vicinity of the tidal inlet. The spatial extent of red drum spawning was still unknown but it was clear that surveying sound production during spawning was an effective means of locating spawning sites.

This paper reports on a survey of potential spawning sites in the nearshore western Gulf of Mexico using a towed hydrophone array.

Study Area and Methods
The survey was conducted in the northwestern Gulf of Mexico along the central portion of the Texas, USA, coast. Preliminary surveys with a hand-held hydrophone in the area revealed that red drum spawning sounds were more commonly observed along the 10 m contour than in either shallow water near the surf zone or farther offshore in deeper water. Hence, for this initial survey, three transects were established roughly parallel to the coastline along the 10 m contour. Transects were sampled on three consecutive nights (one transect per night) in late September 2000. Sampling commenced about 30 - 45 min before sunset, which was about 1925, and ran for about 3.5 hours.

The towed array was composed of eight hydrophones in an 80 meter cable connected to a 200 meter towing cable and was towed at approximately 4.5 kts from a 105 foot stern trawler. The array is spectrally flat (i.e. no peaks in sensitivity) from 6Hz to 18 kHz, with a sensitivity of approximately -191 dB re 1 volt per m Pa at 7.2 kHz. The signals from each of the eight separate hydrophones were saved to an eight-track digital recorder (Tascam DA-88) sampling at 44 kHz. The combination of a temporal window of spawning vocalizations (about 3.5 hours) and optimum towing speed for the array of (4.5 kts) limited each nightly transect to about 20 km.

Red drum produce low frequency sounds described as knocks (Fish and Mowbray 1970) or drumming (Guest and Lasswell 1978). Although Guest and Lasswell (1978) found the "dominant energy" of their recordings from a tank was around 240 Hz — 1000 Hz, I have found the fundamental frequency of red drum calls obtained from unconstrained fish in the field to consistently be around 140 Hz — 160 Hz (Fig. 1).

Figure 1. Sonogram of a red drum call from an unconstrained individual in the field. This particular call consists of three widely spaced knocks followed by two pairs of closely spaced knocks.

Each call consists of a variable number of pulses, or knocks, that are repeated at a range of pulse repetition rates (Guest and Lasswell 1978, laboratory observations; S Holt unpublished data, field observations). Whether there are specific behaviors associated with specific call types is unknown but the existence of numerous variants in call pattern suggests individual variability. Despite variation in call duration and pulse repetition rate, the consistency in fundamental frequency and general character of the call pattern make recognition by ear relatively easy.

Recorded signals from the array were analyzed by listening to the tapes while observing the real-time power spectra and real-time sonogram on a computer screen (SpectraPro 3.32, Sound Technology Inc.). Two classes of red drum sounds could be distinguished. One was a low frequency rumble with a prominent energy peak in the 150 Hz range. This was presumed to be from large numbers of red drum producing sounds simultaneously but at some distance from the hydrophone. (The sound produced by the ship and the hydrophone itself was determined to have dominant energy in the range of 250 Hz — 300 Hz.) The other class of sounds was clearly distinguishable calls made by an individual or small group of red drum.

Figure 2. Location San Jose "A" and "B" hydrophone transects. The line indicates the cruise track. Bars above the line indicate low one-minute drumming rates at that location. Bars below the line indicate high one-minute drumming rates. Sampling time is indicated randomly along the track.

The occurrence of background rumble indicates spawning activity in the vicinity of the hydrophone but more work is needed before the spatial scale over which those sounds travel can be meaningfully interpreted. For this paper, I will describe only the distribution of individual or small-group calls. From our observations and the work of Luczkovich et al. (1999), it appears that the drumming of an individual red drum can be distinguished over a distance of about 100 m. Thus, we can roughly define the spatial distribution of individual red drum detected by the hydrophones as a 200 m swath along the transect. The physical location of each observation was determined by comparing the underway data recorded from the ship’s SAIL system (which included time and latitude/longitude as well as several physical parameters) and the clock time on the digital recorder which was carefully synchronized with the ships clock. The data set was initially constructed by recording the hour/minute/second of each identifiable call. The data was then summarized by counting the number of calls heard in each one-minute segment (the ships location was recorded once per minute so that was our finest scale of spatial resolution). The number of calls/minute was arbitrarily divided in two groups: <16 per minute and 16 or more per minute. This division was set to separate the typically lower occurrence of drumming (5-10 per minute was typical) from the relatively rarer higher rate (we rarely heard more than 20-30 per minute). Finally the drumming rate (i.e. none, low, or high) was plotted on the cruise track.

Results
Red drum calls were detected along most sections of the three transects (Figs. 2 & 3). Calls were detected both in extensive clusters and in isolated occurrences along the transects. For example, on the San Jose "A" transect (Fig. 3), there are two occurrences of near continuous calling that extend over several kilometers. On the same transect, there are several isolated occurrences of red drum calls and extensive segments (up to 4 km) where there are no calls. Transect segments were dominated by the absence of red drum calls. There was a total of 474 minutes of observations over all transects. Of those, 330 minutes (70%) had no red drum calls, 109 minutes (23%) had low drumming rates (<16 ^{per min}), and 35 minutes (7%) had high drumming rates (>15 ^{per min}). High drumming activity was concentrated in two segments along the San Jose "A" transect and in one segment of the Matagorda transect. One segment, on the east end of the transect, spanned 5 minutes of towing time and covered 600 m. The other, farther to the west on that transect, spanned 14 minutes of towing time and covered 2.2 km. Only 4 of the 14 minutes in this segment were low level drumming and none were without drumming.

Figure 3. Location of the Matagorda hydrophone transect. See Fig.2 legend for details.

The most intense drumming activity occurred between 1830 and 2130. Little drumming was heard after 2130 on the Matagorda or San Jose "A" transects (data for the later part of the San Jose "B" transect was lost due to an audio tape malfunction). Low and high drumming rates were distributed throughout this time period without any temporal pattern.

Discussion
Based on the distribution of sound production, red drum appear to spawn all along the nearshore region of the central Texas coast. This survey was not spatially comprehensive enough to fully delineate the spawning area, but it is clear from this initial survey that spawning activity is widespread. Spawning was not concentrated at inlets as suggested by earlier authors (Simmons and Breuer (1962), Jannke (1971). Areas of the coastline far removed from the inlets had relatively intense drumming activity and confirms suggestions of Murphy and Taylor (1990) that spawning also occurs over the nearshore continental shelf.

It is still not exactly clear how drumming by male red drum should be interpreted. There are at least three possibilities: 1) the drumming male will engage in spawning at that location on that evening; 2) the drumming male is calling from a potential spawning site but will spawn at that site on that day only if joined (or selected) by a cooperative female; or 3) the drumming male may move to another place before engaging in spawning. Luczkovich et al. (1999) observed instances of red drum drumming without finding eggs and Johnson and Funicelli (1991) found red drum eggs without hearing drumming. In both instances, short-term observations were made in shallow water with a hand held hydrophone and the observers may have disturbed the fish or missed part of the spawning process. At this point, it is assumed that drumming roughly equates to spawning but the issue needs more investigation.

The distribution of drumming male red drum suggest that some, if not most, of the spawning takes place among widely distributed individuals as opposed to highly aggregated groups. Only 7% of the one-minute summaries recorded high drumming rates of more than 15 calls per minute. Guest and Lasswell (1978) reported a call rate of about 2-16 calls per minute for captive red drum in courtship. Our subjective impression from listening to the tapes was that many of the low drumming rates were produced by a single fish. There were, however, at least two large aggregations of drumming fish. Both were in the vicinity of Cedar Bayou, a relatively small but historically persistent tidal inlet. One of these aggregations spanned a linear distance of over 2 km and its breath was undetermined. The number of calls per minute (up to 40) indicates that several red drum were calling simultaneously within the roughly 100 meter detection range of the hydrophones and this "density" was consistent over most of the 2 km stretch.
The full extent of the offshore spawning area of red drum is yet to be determined and much remains to be learned about their reproductive strategies, but the use of towed hydrophone arrays offers promise of an efficient means to achieve those goals.

Acknowledgments
I thank John Keller for processing the audio data and Cameron Pratt for preparing the figures. The crew of the R/V Longhorn was instrumental in acquiring the recordings. This work was funded through a grant from the Sid W. Richardson Foundation.

References
Fish, M. P., and W. H. Mowbray. 1970. Sounds of western North Atlantic fishes; a reference file of biological underwater sounds. Johns Hopkins Press, Baltimore.

Guest, W. C., and J. L. Lasswell. 1978. A note on courtship behavior and sound production of red drum. Copeia 1978: 337-338.

Holt, G. J., S. A. Holt, and C. R. Arnold. 1985. Diel periodicity of spawning in sciaenids. Marine Ecology Progress Series 27: 1-7.

Jannke, T. E. 1971. Abundance of young sciaenid fishes in Everglades National Park, Florida, in relation to season and other variables. University of Miami Sea Grant Technical Bulletin 11: 128p.

Johnson, D. R., and N. A. Funicelli. 1991. Spawning of the red drum in Mosquito Lagoon, East-Central Florida. Estuaries 14: 74-79.

Luczkovich, J. J., H. J. Daniel, III, and M. W. Sprague. 1999. Characterization of critical spawning habitats of weakfish, spotted seatrout and red drum in Pamlico Sound using hydrophone surveys. Pages 128. North Carolina Department of Environment and Natural Resources, Morehead City, NC.

Lyczkowski-Shultz, J., J. P. Steen, Jr., and B. H. Comyns. 1988. Early life history of red drum (Sciaenops ocellatus) in the northcentral Gulf of Mexico. Pages 148. Mississippi-Alabama Sea Grant Consortium.

Murphy, M. D., and R. G. Taylor. 1990. Reproduction, growth, and mortality of red drum Sciaenops ocellatus in Florida waters. Fishery Bulletin, US 88: 531-542.

Overstreet, R. M. 1983. Aspects of the biology of the red drum, Sciaenops ocellatus, in Mississippi. Gulf Research Reports Supplement 1: 45-68.

Simmons, E. G., and J. P. Breuer. 1962. A study of redfish, Sciaenops ocellata Linnaeus and black drum, Pogonias cromis Linnaeus. Publications of the Institute of Marine Science 8: 184-211.

Return to Top | Workshop Proceedings: Short Papers