F. Observing and Documenting Courtship and Spawning Behavior

We are always faced with the question of whether some behavior seen in fish is actually courting and spawning and how to "prove" this is the case? While documenting spawning is often an objective of a study, it is sometimes uncertain whether some reports of spawning truly reflect that spawning indeed occurred. In most cases experience with the species of interest and with other fishes in general helps. If you have seen reproductive behavior in a number of fishes, particularly ones closely related to those now being watched, this provides background to interpret what is seen in another species. Many observers, including all of the present authors, have at one time or another believed they were watching courtship or spawning behavior when in reality they were seeing something very different. In our cases, we were able to figure out what we were seeing was not courtship and spawning (by applying some basic criteria to verify spawning) before we prematurely published observations as reproductive when they were not.

Similar care is also needed in linking sex and coloration. In many cases, the relationship between differing coloration and sex can be easily observed, but this does not mean that under different conditions the coloration is constant to one sex. A good example of this is the "bicolor" phase of the Nassau grouper. When a small group of fish is present at an aggregation site, preliminary courtship (an hour or more before the spawning at sunset) often has a male in bicolor phase courting females in a somewhat normal phase (see Fig. 30). This might suggest that males are bicolor and females are not, under such circumstances. In reality though, the bicolor phase is believed to represent a "submissive" or non-aggressive signal, not limited to a single sex. In group spawns, females are believed to also display the bicolor pattern, and during non- aggregation, non-spawning season encounters, either sex can flash the bicolor pattern in response to an approach by a larger fish, to indicate it is being submissive to the larger individual.

Courtship and Agonistic Behavior: How to Tell the Difference

More than one observer has been fooled into thinking that some type of agonistic behavior represents courtship in groupers and other reef fishes. In many groupers, males hold territories in the area of the spawning aggregation and often interact at the edges of their territories with other males (Fig. 31). There are a number of things to assess when you see something that might be either courtship or aggression. These include 1) do the two fish differ in color pattern, 2) do they repeatedly tend to threaten each other with a direct head on approach, often with the mouth open and gill covers flared, 3) after an encounter does one fish tend to swim off while the other remains near where it was. Is one fish swollen with eggs while the other is not? Generally there will be some color differences between males and females, although color differences can also

Figure 30. Pair of Nassau groupers, with male (bicolor phase) below, female (‘normal’ phase) above with big belly, engaged in courtship behavior in mid-water, Long Island, Bahamas (PLC).

Figure 31. Two male Epinephelus polyphekadion engaged in a boundary dispute during an aggregation period, Palau (PLC)

occur in same sex fish if they are in different behavioral states or may only last for a short period of time. Where there are alternating aggressive thrusts, during which fish appear to be assessing the other individual’s strength, these are most likely males engaging in a territorial assessment. Females generally seem to be less likely to be aggressive, but not in all cases. In territorial disputes, the loser will usually retreat and the victor remains at the edge of the territory claimed.

Aggressive encounters may be related to spawning territories or have nothing to do with spawning at all. Usually the best way to separate such activities is to gain some knowledge of the behavior of the fish in question by observing their activities over a period of time. Without knowing the typical (day to day) behavior of a fish, it is difficult to determine when it is doing something different, like preparing to spawn (Fig. 32). Aggregation workers first need to be good fish observers.

Figure 32. Probable courtship among Carangoides ferdau at Chuuk Atoll, Micronesia. The probable males have dark areas, typical of courting male carangids. Also in carangids males tend to trail behind females and attempt to keep other males from following the same female. (PLC)

Determining Duration and Timing of Spawning Behavior

In most cases, observers have simply made a series of diving observations which over time (days to weeks) allow a general idea of spawning duration and timing. This can provide significant information and at present is the most feasible method of gathering spawning occurrence data. The method of starting a video camera at a known time and continually taping all activity at the spawning site is quite useful, since it allows the exact timing of documented events to be compiled later (Fig. 33). This technique is discussed in more detail under video documentation.

There are several methods to improve the gathering of this type of data. Most would involve some sort of electronic monitoring of spawning, once spawning behavior has been

Figure 33. Timing of spawning relative to sunset for Epinephelus striatus at Little Harbor aggregation site, Long Island, Bahamas. Data from Colin, 1992

observed and positively identified. For example, Lobel (1992) reported group spawning Scarus iserti to make hydrodynamic noise when spawning which could be detected remotely. This allowed for counting and acoustic mapping of mating events using spawning-associated noises. Such sounds can often be heard by divers, if they are quiet and listen, as a "whooshing" as a group of fish ascends to spawn. Such information would allow the monitoring of spawning activity acoustically by remote stations not unlike those that presently are used to track sonically tagged fishes. Video techniques have great potential to monitor spawning activity, particularly in remote areas where divers can not easily visit study sites on a regular basis. An AUVS system (described subsequently) might find utility in this regard, although the biology of the species must be understood to determine whether sound and spawning are always closely associated.

How to Prevent Disturbance of Spawning Fish

Observers quickly learn that many spawning fishes are disturbed by the presence of divers. The physical presence and movement of divers is disturbing, coupled with the desire of the divers to approach activity closely. SCUBA gear generates noise and bubbles as does associated equipment used by the divers, such as dive scooters, lights, etc.

Figure 34. Probable courtship in the jewfish, Epinephelus itajara. The fish on the left is believed to be the male, with a dark body and light head, while the female is on the right in "normal" coloration. The presumed male approaches the female and displays, often producing a booming sound with its swim bladder. Spawning has not yet been observed in this species. (Photograph copyright by Doug Perrine)

There are strategies for dealing with all of these problems. Disturbance from the physical presence of divers and their movements can sometimes be reduced by 1) moving slowly, 2) staying close to the bottom, 3) reducing the number of divers present, 4) reducing the motion of arms and legs while holding station, 5) not moving directly towards fish, and 6) approaching from a down current direction. As a general rule, try to avoid approaching too close by gradually moving towards the fish and if behavior is interrupted, stop approaching for awhile or even back off a bit.

One particularly useful technique for both short and long-term observations of aggregations is "tethering" (Colin and Clavijo, 1988). The idea of tethering is to make the observer into a stationary object floating above the bottom on a line attached to the bottom. SCUBA divers should be equipped with a buoyancy control device, such as a buoyancy compensator. A light nylon line of suitable length is attached to the sea bottom and a small clip on the upper end is used to attach to some point on the diver. The diver inflates the BC sufficiently to achieve moderate positive buoyancy, so that the diver in essence floats above the bottom like a tethered balloon. Since no swimming is needed to remain above the bottom and the tether prevents the diver from drifting away, the observer can remain stationed above the bottom without moving hands or feet. This greatly reduces the exercise load of the diver and reduces the breathing rate, distracting diver movements, and noise from inhalation and exhalation. In this manner, a tank of air lasts longer and the diver produces minimal disturbance. This also reduces the depth of the diver, compared to being on the bottom, and at depths below about 10 m can mean a significant increase in observation time without decompression being required.

For various types of long-term (day after day) observations, the use of the same tether point allows a directly comparable area (assuming stable water visibility) to be observed each day. At their option, two divers could be stationed sufficiently apart to separately survey two different areas, while remaining in contact visually or by sonic signals, such as banging on tanks, etc. Tethering was used in Puerto Rico to document spawning activity rates per minute for aggregations of two surgeonfishes, Acanthurus bahianus and A. coeruleus (Fig. 35) (Colin, 1985). Permanent tether lines were attached to the reef in 60 feet of water and occupied by an observer for an hour or more each afternoon.

Figure 35. Daily spawning activity of an aggregation of Acanthurus bahianus off La Parguera, Puerto Rico. Data were taken at 1-minute intervals from within a consistent field of view monitored by an observer tethered above the bottom. (After Colin, 1985)

Tethering can also be done while on snorkel at the surface. Many times it is advantageous to be at the surface motionless, able to hold place in a current, for observations. This was true, for example, for the humphead wrasse, Cheilinus undulatus, which is very wary, despite it size, and readily disturbed by SCUBA divers. Since it spawns near the surface, in Palau it could be easily observed while snorkeling and a long line to the bottom allowed the observers to hold position over the drop off without swimming actively.

Where tethering isn't feasible, it might be useful to pick a single point from which to make observations (Colin, 1978). Fishes tend to become habituated to a diver's presence over a period of days. Look for a location where there might be a rock or coral head you can hide behind to reduce your visual presence.

Divers using open-circuit SCUBA find that the noise and bubbles of exhalation are particularly disturbing to fish. While no longer commonly available, some observers prefer the old-style "two-hose" regulators, which exhaust the bubbles behind the head and are somewhat quieter than single hose regulators. The benefits of diving rebreathers are an unknown factor when considering spawning observations. It might seem that having no bubbles or SCUBA noise would be a positive factor in reducing disturbance to fish, but no one has actually used such equipment on any major spawning aggregation study, as far as we are aware.

Diving Techniques and Dive Safety

Many aggregations are located in remote areas, far from shore, and it is often necessary to dive at or near the time of sunset. Such dives may well last until well after sunset, particularly if the divers need to do safety stops or decompression at the end of the dive. Also it is often rough in areas of spawning aggregations, making it difficult for boats to anchor while divers are in the water, or currents may make it difficult to work from an anchored boat. Techniques need to be used that allow divers to work safely in such conditions.

One example we can cite, where it was difficult to work from boats was the Vieques tiger grouper aggregation (Sadovy et al., 1994b). There the aggregation was deep (33-35 m) with no shallow water for anchoring nearby. If the boat did anchor, the bottom was such that the anchor may easily foul. Recommended approaches to dealing with this were to either install a temporary mooring for the dive boat or else dive around a temporary shot line. The shot line should have an adequate lead weight, a line length perhaps one third more than the water depth (i.e. for a 30 m depth, the line should be at least 40 m long) and a large, visible surface float. The float should have a light or at the very least reflective tape, so the boat on the surface can find it easily in the dark. The shot line is dropped at the correct area and the divers are dropped by the boat at the shot line when ready to dive. The divers descend the shot line to the bottom and ideally work around it and ascend up the shot line at the end of the dive. For maximal safety, the divers could move out from the anchor weight using reels and lines, to prevent losing the location of the ascent line, or, a small chemical light could be attached at the anchor weight to help guide divers back to it in the dark after sunset.

The importance of divers not losing track of the ascent line or anchor line of the boat in the dark can not be overemphasized. In the case of an anchored boat, a diver surfacing some distance away may not be easily seen or recovered, since the boat probably must wait for other divers to ascend first. There is the possibility that currents may prevent a surfaced diver from swimming to the boat. Obviously each diver must have devices to indicate their location in the dark, such as lights and whistles (or other sonic device).

For any work on spawning aggregations at depths below about 10 m, it is almost essential to use decompression computers to track bottom time and the need for decompression. The dive computer will allow the observer to focus on the behavior of the fish, rather than constantly be scanning depth gauges to note any change in depth that would affect no-decompression limits when using dive tables. Spawning aggregation dives often involve multiple depth levels, which are well-suited to dive computers.

Nitrox (compressed air with additional oxygen added to reduce the percentage of nitrogen in the gas mix) diving has great potential for use in spawning aggregation work. In many cases, spawning aggregations occur over water depths where nitrox diving is most advantageous, generally about 20-40 m depth. For example, if a 32% oxygen mixture (rather than the 20.8% found in normal air) is used to dive to 30 m, this makes the effective air diving depth of about 24 m. With a no decompression bottom time at 24 m of nearly 30 minutes, this almost doubles the amount of no decompression bottom available on a first dive to a depth of 30 m. Repetitive dives on nitrox gain an increasing advantage, compared to air, so that more effective bottom time at a given depth is achieved using nitrox. The down-side of this is that the lower depth limit for nitrox is much less than for air. Workers should receive specialized training in using and handling nitrox before attempting to dive with such gas mixtures, as the dangers of oxygen toxicity and other factors are very real.

Photographic and Videographic Documentation
Still photography

Still photography using either film cameras or digital cameras can provide extremely useful information on both aggregations and spawning. In general if you wish to capture a large area with many aggregated fish, it is usually necessary to shoot in available light (e.g., Fig. 36). Wide-angle lenses are useful for getting the overall scene. Ideally some ‘fee ‘for the bottom communities can also be obtained from wide-angle photography. For more detailed documentation of color patterns of fish, it is usually beneficial to use electronic strobe illumination for photography.

In the first study to publish any photographs of reef fishes spawning planktonic eggs Randall and Randall (1963) included stills taken from 16 mm motion picture shots of spawning ascents by groups of Sparisoma rubripinne. Myrberg et al. (1988) contains excellent photographs of spawning groups of surgeonfishes taken with Nikonos and super-8 movie camera. Colin (1978) had spawning sequences of Scarus iserti (=croicensis) taken from motion picture footage. Gilmore and Jones (1992) obtained excellent photographs of color patterns of deep- water groupers in Florida aggregations using the external still camera on a submersible. Such photographs, along with notes and sketches made of color patterns, can be used to prepare drawings showing color patterns associated with various behaviors (see Gilmore and Jones, 1992 for examples), illustrate spawning behavior or sequences, or directly show color changes associated with courtship (Fig. 37, 38).

In many instances photographing upward against lighter surface layers, particularly for aggregations in deep water where the fish move off the bottom at the time of spawning, such as Nassau groupers and some snappers, can provide a useful exposure whereas photographing downward would not have sufficient light for a decent exposure. Upward oriented photos can provide useful information on fish numbers, since they generally appear as silhouettes against a lighter background.

Figure 36. Spawning sequence of Nassau grouper, Epinephelus striatus, as recorded by video (left) and still photography (right). The left panel has stills taken from an 8 mm analogue video tape while the right photos of the same spawning sequence were taken using a Nikonos camera with 15 mm lens on Tri-X black and white film and available light. (PLC)

Figure 37. Normal color pattern in the adult tiger grouper, Mycteroperca tigris.

Figure 38. Male tiger grouper, Mycteroperca tigris, in courtship coloration (head becomes yellow/bronze, ventro-posterior area becomes white) near Vieques Island, Puerto Rico, February 1992 (from Sadovy et al. 1994b) copyright M.L. Domeier.

Color film can be used in low light situations, with ASA ratings being able to be push developed to over 1600 or so. However, when film is pushed, graininess increases and detail is lost. A fast black and white film, such as Tri-X, is perhaps preferable to color slide film for ambient light photos during late afternoon. During the period near dusk, it is almost always necessary to shoot with the camera aperture wide open and a slow shutter speed. This requires a steady hand on the camera. We feel the Nikonos series cameras with wide angle to normal angle lenses and high speed B&W film are the best way to get still photos of low light spawning sequences. Often still photographs and video are used to prepare a drawing showing the typical spawning behavior of a species (Fig. 39). This allows all the aspects of a spawning act to be shown in one picture.

Figure 39. Generalized pattern of a spawning rush by the Nassau grouper (from Sadovy and Eklund, 1999 - NOAA Tech. Rpt. NMFS 146) indicating the color patterns, motion and gamete release point for the spawning. Numbers represent different stages of the spawning rush: 1. fish begin to move into the water column; 2. small group of fish rising; 3. sperm and eggs are released, and 4. fish return rapidly to the substrate.

Digital Still Photography

Digital photography is becoming increasingly common underwater and would seem to have great potential for documenting spawning aggregations. The light sensitivity of such cameras is generally better than high speed film, and the ability to quickly download the images taken, and manipulate their contrast and brightness, allows rapid use of such photos in the field. There are many underwater housings for the popular digital cameras, however, at this point we can not make any specific recommendations since our experience in this new field of photography is limited.

Useful still images can be obtained from digital video cameras by "grabbing" individual frames. This would be useful in analyzing a spawning sequence (see below).

Artificial Illumination

To obtain the details of coloration in aggregated or spawning fish it is usually necessary to provide strobe illumination for still photos. Artificial lighting is generally not useful in photographing overall aggregations as it can only illuminate a small area and disturbs the behavior and coloration of the fish.

Video Cameras and Underwater Housings

Underwater motion pictures for aggregation and spawning have pretty much been superceded by video camcorders, and only the most unusual circumstances would warrant shooting motion picture film of aggregations for scientific study.

The underwater video camera is perhaps the single most useful tool for documenting aggregations (Fig. 40), including the species present, numbers of fish present, and their behavior, such as courtship and spawning. Before the mid-1980s, scientists were limited to still photographs and occasionally, super-8 or 16 mm film, to visually document aggregations. The range of video camera and housings available today is great and the selection of the right equipment can be important for optimal results.

Today the major choice is between digital and analogue video cameras. Each has its advantages and neither type is ideal for all situations. Digital video cameras provide better resolution, the ability to stop-frame with high resolution, the ability to make full resolution copies and export digital images. Their one drawback, however, is a lesser light sensitivity, compared to analogue video, and given that much spawning and courtship activity occurs in low light conditions around sunset, and in some cases sunrise, this can be a major disadvantage. Ideally a study should have both types of video cameras available, with the analog camera specifically capable of imaging in low light conditions.

Figure 40. Typical underwater video camera with the housing on the left and camcorder on the right. This particular model, a Sony VX-1000 is a high-end three chip digital camera. It produces excellent images, but is limited in its low-light level capabilities.(PLC)

With zoom lenses, a single video camera is capable of capturing the broad picture of an aggregation, then zooming in on specific behavior that might have to be recorded from some distance away (assuming clear water). Tape length becomes a factor if extensive recording needs to be made on a single dive. Digital cameras generally having a 60 min tape length while analogue cameras have up to 120 min tape length. Battery life also is important, as earlier video cameras often did not have the battery power to record an entire video-tape without recharging the battery. More recent video cameras have improved batteries and lower power consumption, so an entire tape could be recorded on one dive. However, if using or purchasing an earlier video camera, the factor of recording time on one battery charge is important. Also, batteries used for underwater videos need to be in good condition and their recharging carefully undertaken to retain maximum battery length. There is nothing worse than having your video camera die in the middle of recording seldom seen and critical activity.

It is possible to do analogue to digital transfers, that can then be used for stop frame analysis which would not be possible by directly viewing the analogue images.

Video Lights

The use of external video lights is a subject that needs some consideration. In general, it is not advisable to use external video lights to try to document aggregations and spawning behavior. Lights usually constitute a disturbance beyond that of observers that can interrupt behavior or drive the fish away from observers. In mid-day periods, if an aggregation is large, lights are not powerful enough to provide significant illumination compared to ambient light. During low light periods, when they would be more useful, the disturbance factor increases greatly.

Garth (1991: 48) described what happened when video lights were used to try and film Nassau grouper spawning. He writes "they were all swollen with milk (sic) and eggs, so spawning was still to come. However, we noted that the lights from our cameras spooked them. Some even changed back to their normal color as fast as the light hit them. As it happens, the fishermen are still complaining that the catch has not been as good since Cousteau came several years ago. In an attempt to capture the spawning on film, the Cousteau team had set up lights that burned all night. Nothing of substance was obtained, presumably because of the lights".

If using still cameras, attempts to document spawning behavior can be done using high- speed film, either color or black-and-white, or flash photography. While fishes are usually disturbed by video lights, they are less likely to be disturbed by electronic flash. Consequently, some good photographs showing color patterns and such can be obtained by judicious use of electronic strobe photography. The flash may interrupt behavior, but this is usually just temporary. It can be argued that an electronic flash is something which is not totally alien to reef fishes, as they would be exposed to light from lightening flashes both day and night, whereas a continuous light emanating from a video or movie light is something they would not normally encounter, hence would view it as more disturbing.

Using Video Equipment

How the video equipment is to be used is important. It is advisable to tape a short section at the beginning of each recorded segment showing the time, date and location where the tape is to be recorded, for reference purposes. We believe it is extremely useful to adopt the strategy of continuously recording activity at the spawning site. In other words, the video camera is started at some point and run continuously until the end of a dive or observation period. This goes somewhat contrary to impulse, in that it would appear to "waste" video-tape, however, if the intention of the project is to extract maximum information from observations, it is a reasonable thing to do. First, by knowing the time the camera was started, this allows the time of any event occurring to be determined. Some of the more advanced digital video cameras have a time clock built into their recording system, that would allow this information to be obtained without continuously running the tape, but not all cameras have this feature. If the tape is constantly recording, then when something happens suddenly, the observer can tape it simply by pointing the camera towards it, rather than having to activate the record system, which may take several seconds and valuable information will be lost. Also, the timing of rapid spawning events can be determined later by analysis of the recording, resulting in better data. The humphead wrasse example, below, is a good one of spawning frequency information extracted from video-tapes.

One unusual advantage of using a video camera is that the activity of, say, a pair of fish in one direction, can be recorded by simply pointing the camera at them while the observer is actually looking at something else (a second pair of fish) in a second direction. Rather than twisting the head back and forth, the tape can be reviewed later to see what the first pair of fish was doing. One unusual advantage of using a video camera is that the activity of, say, a pair of fish in one direction, can be recorded by simply pointing the camera at them while the observer is actually looking at something else (a second pair of fish) in a second direction. Rather than twisting the head back and forth, the tape can be reviewed later to see what the first pair of fish was doing.

Disadvantages to this continuous taping approach are that unless the camera battery is well charged it may die before the end of the tape is reached. This is not a good thing to have happen. Also a lot more recorded video-tape is generated, with greater costs for tapes and need to store them.

It is also useful to consider mounting the video camera on a tripod, or simply setting it on the bottom in a safe place (where there is minimal surge, etc), then at some point starting the record function of the camera and swimming off to leave the camera to run for its tape duration. The diver starting the camera can go elsewhere, or totally exit the water. Fish very quickly adapt to the presence of the camera and behave as they would normally. This technique proved useful in studies of spawning behavior of Nassau grouper and especially the humphead (Napoleon) wrasse. Also because the camera does not move, changes in numbers of fish in its field of view can be quantified over time, giving perhaps insight into activity that would be impossible to obtain by direct diver observations.

We have used cheap standard camera tripods to mount video cameras underwater and these have proved satisfactory over periods of months. Try to avoid tripods with lots of easily corrodable metal parts, look for plastic legs, locks and other parts. Some lead diving weights can be taped to the legs to increase the weight and stability of the tripod underwater in currents and swell. In extreme conditions, or when you might leave the camera and tripod overnight or longer, it would be advisable to tie the legs of the tripod to structures on the bottom. Simple, non- adjustable tripods can be made from concrete reinforcing steel (rebar) welded together. Such camera mounts can be adjusted by actually bending the legs underwater once a particular site is selected.

The next step beyond a tripod mounted stationary video system would be a system which can be put in place at any time, with a self contained computer programmable timer/controller, that will turn the video camera on and off at times selected in advance. Such a unit, which we could call an "Autonomous Underwater Video System", or AUVS (Fig. 41) would prove, we think, exceptionally useful in documenting parameters of many spawning aggregations, particularly those in deep water or remote locations where regular diving observations are not feasible. Besides the underwater video system, an AUVS could include a pan and/or tilt mechanism to increase its observational view, controls for lights or other equipment. One of us developed a prototype system some years ago (Fig. 41), which was successfully used to document

Figure 41. Autonomous underwater video system (AUVS) being programmed prior to deployment (left) and in use on the reef (right).

feeding activity in garden eels, cleaning activity at a cleaning station and occurrence of fishes in a "ghost" fish trap. However, it has not yet been used to document aggregation presence and behavior, its originally intended role! A new generation of AUVS is under development, using digital video camera and lower power consumption components. Such a unit will find much use in aggregation studies and many other areas of marine biology and oceanography (Fig. 42)

There is also great potential in using stereo video in documenting fish aggregations. Harvey and Shortis (1997) have described a system useable to assess the accuracy and precision of measurements of fish length and distance from the cameras. This type of system has yet to be utilized on any spawning aggregation, but would seem a natural extension of this technology.

Figure 42. Appearance of Cheilinus undulatus at a spawning site in Palau, determined by unattended video camera on a tripod at the site. Number of females (dark circles) and males (open circles) within the field of view during a 10 period each minute (PLC)

Instrumentation on Aggregation Sites

Gathering additional information on physical parameters at aggregation sites is very useful and important, as we discuss below. The most important among these are probably temperature and current speed/direction.

Temperature measurement

Temperature is relatively simple to document since there are many data-logging instruments now available. Temperature monitoring at sites can range from making measurements only during an aggregation period, to installing instruments to monitor temperature continuously over the year at the site. The latter option is best, since it would allow you to characterize the annual temperature cycle at the site and look at the temperature trends before and

Figure 43. Annual temperature pattern at a Nassau grouper aggregation site, Long Island, Bahamas. Data are weekly means with range represented by the vertical bars. The times of aggregation and spawning are indicated by arrows. (After Colin 1996)

after the time of aggregation. Colin (1992) installed a thermograph at a Nassau grouper aggregation site in the Bahamas (Fig. 43). This study found that the aggregation occurred on two different months at temperatures of 25-25.5o C. The aggregations occurred during the general downtrend of temperatures in winter, but were not at the time of the yearly minimum. If temperature had been measured only during the aggregation, there would have been no way to place aggregation occurrence in relation to the annual thermal cycle. It is also useful to monitor temperature at some sites not used for aggregation, to see if the aggregation site might possibly have some different temperature regime, say induced by upwelling, compared to non-aggregation sites. Where multiple instruments are involved, they should be carefully cross-calibrated to provide comparable numbers.

Current measurement

Currents involve both a current speed and a direction making them more difficult to measure than temperature. Their measurement can be approached as either a simple or complex undertaking. In most cases we would be focused on "why" a fish is using a particular site, so it is important to gather information on the currents at the site both during spawning and at other times and at other sites nearby. There are, unfortunately, no inexpensive current meters comparable to the recording thermographs now available. Recording current meters typically cost a minimum of a few thousand dollars, but if the project budget can afford them, they are well worth the investment. Sancho et al. (2000) used InterOcean S-4 current meters mounted on 1.5 m tripods in a channel with an average depth of 4.5 m. Colin (1992) used the General Oceanics tilt vane current meters successfully during the Nassau grouper work to document currents both at aggregation and non-aggregation sites (Fig. 44). This meter uses a special stand-off from the mooring wire that reduces meter oscillation due to swells.

Figure 44. Currents around Long Island, Bahamas, as recorded by stationary current meters, October 1988-March 1989 (redrawn from Colin, 1996).

In most cases, the currents at and above the level in the water column where the gametes are released are of most interest. This usually means the current meter needs to be moored above the bottom using a mooring line topped by a float capable of supporting the meter and mooring above the bottom in currents (which would tend to pull the mooring down) expected at the site. For anything other than just a short deployment, we would recommend using stainless steel wire rope for the mooring, with the ends made into loops, with stainless steel thimbles inserted, using either crimped nikopress sleeves or wire rope clamps. If properly crimped or clamped, there is little chance the mooring line will fail. The mooring must be anchored to the bottom, and we recommend a heavy (20-30 kg) lead weight with a solid eyebolt to which the mooring line is shackled. It is likely the mooring will be set up by a diver and in the case of the mooring anchor, the heavy lead weight is best moved into place using a lift bag. There are benefits in having the entire mooring (anchor, line, meter and float) prepared on the surface and put into place as a unit. It would help to have the site selected for the meter carefully located and buoyed at the surface so there is little need to swim any distance with the mooring. If possible the exact depth of the meter on the mooring can be adjusted after deployment through some type of moveable clamp on the mooring.

Flowmeters can be used for measuring currents at spawning sites, but have the disadvantage of having to be read manually, usually by a diver, at certain time intervals and provide no current direction information (although a diver can determine the direction of the flow meter using an underwater compass). General Oceanics Inc. makes an excellent plastic flow

Figure 45. (Left) General Oceanics flowmeter, showing the low speed rotor and mechanical counter for revolutions. (Right) Vertical array of four flowmeters, tethered to reef by lines and held up by subsurface float.

meter (GO Model 2030 mechanical flowmeter) which counts the revolutions of a small propeller and is designed to face into the current. The version with the "low speed rotor" is probably most suitable for aggregation current measurements. A stacked array of flowmeters can be used to give a vertical picture of current distribution (Fig. 45), as in the illustrated case of the current coming off a shallow reef as it reached deeper water. The flow meter has to be manually read at intervals, and the current speed later determined by a revolutions/time determination. At a minimum, the flow meter could be read at the start of a spawning period and again at the end, with current direction measurements taken similarly (Fig. 46).

Figure 46. Currents flowing off Lighthouse Reef, Palau, after high tide, as determined by moored flow meters. Black symbols are spring tide days while open symbols are neap tide days (PLC).

Flow meters could also be something as simple as a model airplane propeller or some type of fan blade in which the revolutions are counted for some period of time, such as 1 minute.

The flow meter would need to be calibrated against some device for which relative current speeds are known, but would be an inexpensive alternative.

At the simplest extreme Shapiro et al., (1993) measured current velocity by releasing fluorescein dye from a syringe 1.5 m off the bottom and timing 150 cm of movement of the leading edge of the dye. Shapiro (pers. comm.) also used a simple device to estimate current speed. A neutrally buoyant object, like a small buoy or solid ball is attached to a string of known length. The object is released by a hand also holding the end of the string, the time it takes the object to reach the end of the string provides an estimate of current speed. For example, if it takes the object 10 s to reach the end of a 1m line, the current is approximately 10 cm sec-1.

Light Measurement

As far as we are aware, no investigators have measured light intensity quantitatively in connection with spawning aggregation studies. There is no reason why this could not be done, and might be interesting to do for species that aggregate and spawn in late afternoon, irrespective of tide, such as Acanthurus coeruleus and A. bahianus. Whether the time of spawning is affected by the amount of light on any given day, which can vary at the same time due to cloudy conditions, is unknown. Instruments to quantify light intensity could either be a self contained unit, or a unit with the sensor on a cable and the display portion on the surface, usually in a boat.