Thesis - Loma Linda University
Transcription
Thesis - Loma Linda University
LOMA LINDA UNIVERSITY School of Science and Technology in conjunction with the Faculty of Graduate Studies ____________________ Characterization of Resting Holes and Use by the Antillean Manatee (Trichechus manatus manatus) by Marie-Lys Chantal Bacchus ____________________ A Thesis submitted in partial satisfaction of the requirements for the degree of Master of Science in Biology ____________________ June 2007 © 2007 Marie-Lys Chantal Bacchus All Rights Reserved Each person whose signature appears below certifies that this thesis in his/her opinion is adequate, in scope and quality, as a thesis for the degree of Master of Science. , Chairperson Stephen G. Dunbar, Assistant Professor of Biology Ronald L. Carter, Professor of Biology Robert E. Ford, Professor of International Sustainable Development and Social Policy Caryn Self-Sullivan, Researcher, President of Sirenian International iii ACKNOWLEDGMENTS I would like to first of all thank my advisor, Dr. Stephen Dunbar, for all the help, support and guidance he has given me through this process. Without his constant feedback, financial support, and equipment loans, this project would have never happened. I also want to thank my committee members: Dr. Ronald Carter, Dr. Robert Ford and Caryn Self-Sullivan. Dr. Carter patiently listened when everything seemed to be going wrong with my research and gave me a lot of advice. Dr. Ford provided funds from his ESSE 21 (http://esse21.usra.edu/ESSE21/) grant and a lot of support. Caryn was willing to take me under her wing as a student on very short notice even though she hardly knew me, and I am extremely grateful to her for that. She took time to listen and help through the difficult parts of research and was very encouraging and supportive while teaching me how to conduct research with manatees. I want to especially thank my Belizean field assistant and boat driver, Gilroy Robinson, who taught me so much about manatees and Belize. He was my constant helper, and this project would not have been possible without him. He also kept encouraging me when things were difficult and worked over-time to help me work out problems. Dorian Alvarez also worked over-time as my assistant and boat-driver; thank you, Dorian, for working so hard and being so helpful. Teresa Parkey has allowed me to stay on her island and worked very hard to give me access to a boat as much as possible during my time there. She also helped with inkind donation for my food and accommodations. iv The fieldwork would have been impossible without the help of the field assistants who helped coordinate volunteers, meals, equipment and many other details: thank you to Leigh Bird, Shauna King, Haydée Domínguez, Karen Petkau, and Ryan Roland. Volunteers from Earthwatch teams 2, 3, and 4 of the summers of 2005 and 2006 contributed through the Earthwatch Institute’s volunteer model. Thank you for your financial and physical contributions to the project; your hard work saved me a lot of time and I enjoyed getting to know all of you. There are many more people to thank from the Hugh Parkey Foundation, without whom this project would not have run smoothly, such as the boat drivers who took me back and forth from the island, the cooks who cooked wonderful Belizean food for us, and the maintenance and cleaning crews who helped keep the island running so I could focus on my research. I also want to thank many people from Loma Linda University who helped make this project possible. Thank you to Daniel Gonzalez, my manatee partner, who was willing to come to Belize and help with data collection. He also taught me how to use the sonar and stayed long enough to make sure it was working properly. He was a source of encouragement and someone with whom I could brainstorm throughout this whole process. Dr. William Hayes provided invaluable assistance with my statistics many times, and helped me make sense of my data, as did Zia Nisani. I also want to thank the Marine Research Group (LLU) for funds and assistance in reviewing drafts and Rafael Canizales for helping with thesis formatting. In addition, I would like to thank Seth Wiafe and Jesse Bliss for their help in making and analyzing my maps correctly. v I also would like to thank Katie LaCommare, Dr. Solange Brault, Dr. Christopher Marshall, Dr. Bob Bonde, Dr. James Powell, Nicole Auil, and Dr. Leszek Karczmarski for allowing me to bounce project ideas off of them and for encouraging me in my research. Thanks to Rick Ware for providing the Belize City tide tables and to Joe Breman for helping with my GIS maps. I would like to thank my sister, Stephanie Bacchus, for all her encouragement over the past three years; my mom, Coralie Lallemand, who has been very supportive, and even came to help me in the field for a couple of weeks; and my husband, Austin Bacchus, for standing behind me every step of the way, believing in me and encouraging me when things got tough. He endured my absences for 10 weeks of two summers without a complaint, and spent his vacation time helping me to collect data. He also patiently read through all my drafts and gave me great feedback. Thank you! vi TABLE OF CONTENTS Approval Page.................................................................................................................... iii Acknowledgments.............................................................................................................. iv Table of Contents.............................................................................................................. vii List of Tables ..................................................................................................................... ix List of Figures ..................................................................................................................... x Abstract ............................................................................................................................. xii Chapter 1. Introduction.........................................................................................................1 Distribution and Status.................................................................................1 Habitat Preferences ......................................................................................3 Activity Patterns and Resting Behavior .......................................................4 Objectives and Significance.........................................................................7 2. Materials and Methods........................................................................................9 Study Site .....................................................................................................9 Characterization of Resting Holes ...............................................................9 Diurnal and Nocturnal Scans .....................................................................15 Behaviors ...................................................................................................17 Data Analysis .............................................................................................18 Maps...........................................................................................................18 3. Results...............................................................................................................20 Characterization of Resting Holes .............................................................20 Diurnal and Nocturnal Scans .....................................................................40 Behaviors ...................................................................................................54 4. Discussion ..........................................................................................................61 Characterization of Resting Holes .............................................................61 Diurnal and Nocturnal Scans .....................................................................68 Behaviors ...................................................................................................72 Ecological Implications..............................................................................73 vii Conclusions................................................................................................76 Suggestions for Further Study....................................................................77 References..........................................................................................................................80 viii TABLES Table Page 3.1. Depth and surface water velocity of areas with and without resting holes ............21 3.2. Descriptive statistics of environmental characteristics of areas with and without resting holes. .............................................................................................23 3.3. Summary of all sightings and scans .......................................................................44 3.4. Sightings were divided by number of scans for each location and category to give a percentage of sightings per scans ............................................................46 3.5. Mean, minimum, maximum sea surface temperature, salinity and air temperature per year for all four scan points..........................................................48 3.6. Descriptive statistics for habitat factors when manatees were sighted compared with times when no manatees were sighted ...........................................................49 3.7. Sea surface temperature and salinity when a manatee was sighted and when no manatees were sighted ...........................................................................................50 3.8. Descriptive statistics for day and night scans of four locations .............................53 3.9. Forward stepwise logistic regression results..........................................................55 3.10. Behaviors of manatees at each location and during different tides........................57 3.11. Behaviors of manatees and associated sea surface temperatures and salinities.....57 3.12. Intervals (in seconds) between breaths for each manatee while resting.................58 ix FIGURES Figure Page 2.1. Map of Belize and the Drowned Cayes in relation to Belize City .........................10 2.2. Sonar side-imaging representation ........................................................................14 3.1. The Drowned Cayes of Belize with characterized scan points ..............................22 3.2. Depth of resting holes and non-resting holes.........................................................24 3.3. Water velocity of resting holes and non-resting holes ...........................................25 3.4. Raster interpolation of depth measurements at 43 BaCo.......................................27 3.5. Resting holes at 30 CCo and 31 CLa .....................................................................28 3.6. Raster interpolation of depth measurements at 30 CCo.........................................29 3.7. Sonar image of the resting hole at Bogue C Big Lagoon (31 CLa) .......................31 3.8. Raster interpolation of depth measurements at 31 CLa .........................................32 3.9. Raster interpolation of depth measurements at 13 Su3..........................................33 3.10. Raster interpolation of depth measurements at 49 GiGr........................................34 3.11. Sonar image of the resting hole at Gilroy’s Grassbed East Side (49 GiGr) where seagrass can be seen ....................................................................................35 3.12. Close-up of 49 GiGr and 50 GiCr..........................................................................36 3.13. Sonar image of the resting hole at Gilroy’s Grassbed East Side (49 GiGr) where coral can be seen .........................................................................................37 3.14. Raster interpolation of depth measurements at 12 StLa ........................................38 3.15. Raster interpolation of depth measurements at 33 HeCo.......................................39 3.16. Raster interpolation of depth measurements at 54 SLCW.....................................41 3.17. Raster interpolation of depth measurements at 08 G2En.......................................42 x 3.18. Total number of scans and sightings per year for each location ............................43 3.19. Sightings at each location and for each category were divided by total number of scans (168) for comparison purposes ...................................................47 3.20. Sonar image of a manatee taken in the channel of Bogue C..................................51 3.21. Day and night scans and sightings for each location .............................................52 3.22. Close-up of Bogue B..............................................................................................59 xi ABSTRACT OF THE THESIS Characterization of Resting Holes and Use by the Antillean Manatee (Trichechus manatus manatus) by Marie-Lys Chantal Bacchus Master of Science Graduate Program of Biology Loma Linda University, June 2007 Dr. Stephen G. Dunbar, Chairperson Manatees in the Drowned Cayes of Belize are known to rest primarily in depressions in the benthic substrate, called “resting holes”. These are two to over five meters deep and are usually found in quiet coves, lagoons, and dead-end bogues where there is minimal disturbance from humans, wind, waves and currents. The objectives of this study were to characterize resting holes and investigate diurnal and nocturnal use of these by manatees. Characterization of 12 resting holes in the Drowned Cayes was accomplished by collecting data on depth, type of substrate, benthic vegetation, water velocity, salinity, and temperature through the water column above the resting hole. Resting holes were significantly deeper and had slower surface water velocity than areas without resting holes. Four resting holes in the Drowned Cayes were chosen for diurnal and nocturnal scans. The four sites were visited by boat daily between 10:00 a.m. and 3:00 p.m. and again at night between 7:00 p.m. and 12:00 a.m. After each site scan, sea and weather conditions were recorded. Data were also recorded for each manatee sighting, including xii all specific activities and general behavioral states. Nocturnal scans were performed in the same manner as daytime scans, except that spotlights were used. A total of 168 scans were conducted resulting in 39 manatee sightings. There was a significant difference in the number of sightings between years and between day and night. Since there are a large number of resting holes in the Drowned Cayes, and many of these seem to be in sheltered areas with low currents, it is quite possible that manatees choose these spots precisely because of the tranquility of the area. The combination of low currents, protection from waves, low numbers of boats, and nearby seagrass beds would make these ideal areas in which to rest. Resting holes in the Drowned Cayes seem to be important to manatees. Several resting holes are used regularly by a number of animals. In light of this, one of the primary conservation goals should be to preserve this important manatee habitat, including restricting boat speed and access. xiii CHAPTER ONE INTRODUCTION Distribution and Status The order Sirenia is represented by two families: Dugongidae and Trichechidae. Dugongidae contains one living species, the dugong (Dugong dugon), which lives in the Indian and Pacific Oceans (Husar, 1978). Trichechidae contains three species: the West African manatee (Trichechus senegalensis), the Amazonian manatee (T. inunguis), found in the Amazon basin, and the West Indian manatee (T. manatus), found in many of the rivers and coastal areas of the Southeastern United States, the Greater Antilles, eastern Mexico, Central America, and parts of northern South America (Lefebvre et al., 2001). Trichechus manatus is comprised of two subspecies: the Florida manatee (T. m. latirostris) and the Antillean manatee (T. m. manatus). These subspecies are differentiated primarily by cranial characteristics (Domning and Hayek, 1986), although this two-fold division has been challenged in genetics studies (Garcia-Rodriguez et al., 1998; Vianna et al., 2006). Often, life history information from Florida manatees has been used to fill in the gaps for Antillean manatees. However, these subspecies are behaviorally distinct and are affected by different environmental factors. Therefore, it has been suggested that conservation efforts should be geared towards each subspecies separately (Reynolds, 1999). Both subspecies of the West Indian manatee are categorized as “Endangered” on the IUCN Red List (Deutsch et al., 2007). Over the years, several factors have played a 1 role in reducing populations of this marine mammal. In Florida, collisions with boats are responsible for about one-quarter of manatee deaths annually (Aipanjiguly et al., 2003). Throughout Central and South America, manatees have been hunted commercially, and for subsistence, for food, for medicinal remedies, and other purposes; this has taken a significant toll on the Antillean subspecies (Lefebvre et al., 2001). Accidental entanglement in fishing and seine nets also causes a number of deaths. Habitat degradation due to pollution, plantation run-off, and siltation is also a growing threat to the manatee’s environment in many of these countries, and may be the highest threat to these animals’ survival due to growing human population (Lefebvre et al., 2001; Reynolds, 1999). According to Reynolds (1999), conservation efforts should be focused on manatee habitat, rather than just the manatees, as habitat loss is one of the biggest threats to the survival of manatees worldwide. Additionally, conservation of an ecosystem, rather than trying to preserve one particular species or sub-species allows protection to extend to many other species, whether or not they are endangered or threatened (Simberloff, 1998). This view takes into consideration many parts of an ecosystem, rather than focusing on trying to save one particular species, and also aids in avoiding conservation conflicts between species that may utilize the same geographical area. Over the years, several aerial surveys of Belize have indicated one of the highest counts of manatees outside of Florida (Bengtson and Magor, 1979; O’Shea and Salisbury, 1991; Morales-Vela et al., 2000). According to O’Shea and Salisbury (1991), “Belize remains one of the last strongholds for this species in this part of the world,” mostly due to the high quality of habitat and low level of killing compared to other Central and South 2 American countries. During several aerial surveys in 1994 and 1995, Morales-Vela et al. (2000) found manatees to be most numerous in Chetumal Bay, Mexico, in Placentia and Indian Hill lagoons along the Belizean coast, and around the Drowned Cayes (meaning island or islet; Ford, 1991) east of Belize City. In Belize, the recent influx of tourism has led to an increase in boat traffic. Cruise ships anchor east of Belize City and ferry rising numbers of tourists to the mainland and to different cayes. This increase in boat traffic has naturally multiplied the possibility of boat collisions with manatees, and researchers have noted a rise in the number of manatees found with boat propeller scars in the last few years (Powell, pers. comm.). The growth of tourism and boat traffic also causes a loss of habitat through development, as well as degradation of food resources through loss of seagrass beds. In order to aid in the protection of habitat for the Antillean manatee, more study needs to be done to understand the factors which influence their habitat preferences. Habitat Preferences In subtropical areas such as Florida, manatees are restricted by water temperatures and tend to avoid water temperatures below approximately 20oC (Irvine, 1983; Deutsch et al., 2003). This temperature restriction is minimized in tropical areas, since water temperatures tend to remain above this threshold all year round, although manatees do seem to prefer warmer waters above 24oC (Lefebvre et al., 2001; Jiménez, 2005), and may be more influenced by the rainy and dry seasons (Reynolds, 1999). Although West Indian manatees can move freely between freshwater and saltwater, their preferred habitats have historically been reported as freshwater rivers and estuaries in Florida 3 (Hartman, 1979). When manatees have been found in marine or coastal environments, they tended to be located near freshwater sources (Powell et al., 1981; Powell and Rathbun, 1984; Rathbun et al., 1990; Olivera-Gomez and Mellink, 2005). Another important factor in manatee habitat preference is the distribution and abundance of comestible vegetation (Hartman, 1979; Kinnaird, 1985; Deutsch et al., 2003; Jiménez, 2005; Olivera-Gomez and Mellink, 2005). The West Indian manatee (Trichechus manatus) feeds on a variety of aquatic and terrestrial plants, including vascular and nonvascular plants, but tends to prefer a variety of seagrasses in waters less than 4 meters (Best, 1981; Powell et al., 1981; Reynolds et al., 1995). Manatees tend to avoid very shallow waters less than 1.5 meters and seem to prefer waters around 3 meters deep (Hartman, 1979; Lefebvre et al., 2000). Tides may also have an effect on manatee movements. Observations in Florida indicated that tides may bar access to certain shallow areas or allow manatees to feed on the margin of shoals (Hartman, 1979). In Nicaragua, manatees have been reported to be more active when tides are changing (Jiménez, 2002). Activity Patterns and Resting Behavior In Florida, manatees appear to be essentially arrhythmic, spending most of their time feeding, resting, idling, cruising, and socializing with no correlation of activity with time of day (Hartman, 1979; Bengtson, 1981). They tend to spend from six to eight hours a day feeding in sessions usually lasting up to 90 minutes, although longer feeding sessions have been observed (Hartman, 1979; Reynolds, 1981). There are no apparent predictable patterns of activities and thus, the Floridian manatee appears to lack a 4 circadian cycle with nocturnal behaviors paralleling diurnal behaviors (Hartman, 1971). It is interesting to note that Trichechus manatus latirostris may not possess a pineal gland. One female manatee from Georgia was autopsied, and did not appear to have a pineal body (Ralph et al.1985). This absence might help explain the lack of a circadian rhythm, since the pineal gland is at least partly responsible for production of melatonin in the brain, which helps regulate circadian rhythm. However, other organs, such as the retina, are known to carry out the same functions as the pineal gland in other species, and lack of a pineal organ does not automatically imply absence of a circadian rhythm (Ralph et al., 1985). In contrast, observations of a solitary Amazonian manatee demonstrated that this animal likely had a circadian rhythm, sleeping for the first half of the night (Mukhametov et al., 1992). In some countries, anecdotal information indicates that over-hunting may have caused the Antillean manatee to shift to a more nocturnal activity pattern (Rathbun et al., 1983; Reynolds et al., 1995). Still, we know little about manatee behavior on a diurnal basis since no free-ranging manatee has been observed for a full 24 hours. Likewise, there have been no consistent studies on behaviors and habitat use of the Antillean manatee at night. For the Florida manatee, resting takes up about six to ten hours of their daily regimes in periods of two to four hours, and usually occurs in water between two and four meters deep (Hartman, 1979). There are two positions that a manatee will take while resting. One is to hang suspended near the surface. This is usually a prelude to the second posture, which is to lie prostrate on the bottom in areas where the water is calmer. In both positions, the animal’s eyes are closed until it begins to surface for a breath. Resting 5 animals tend to have a predictable breathing rhythm, surfacing and taking two to four breaths in quick succession before sinking back to the bottom, usually in intervals of four to five minutes (Hartman, 1971). Hartman (1971) reported a record submergence time of 11 min 46 sec by a resting Florida manatee, while Reynolds (1981) observed one resting manatee in Blue Lagoon, Florida submerged for 24 minutes. In some Central and South American countries where Antillean manatees are found in fresh or brackish waters, they tend to take refuge and rest in deep, quiet waters. In Venezuela, during the dry season, manatees take refuge in deep holes or channels in the rivers (O’Shea et al., 1988). In Costa Rica, manatees are thought to congregate in deep holes, which are locally called “blowing holes” (Smethurst and Nietschmann, 1999). According to local former hunters, manatees in these holes engage in play and wait for high tide to swim up shallow creeks to feed. These holes tend to range from six to over 12 meters in depth (Smethurst and Nietschmann, 1999). In the brackish waters of Nicaragua, manatees tend to rest in quiet, sheltered, deep water during most of the day, only leaving to feed during the night, early morning and late afternoon (Jiménez, 2002). In all these countries, holes used by manatees are usually in very deep, fresh or brackish water, and are used primarily during the dry season or during low tide. Most of this information is anecdotal, collected from local fishermen. Aside from this, little else is known about the use of these holes, or if manatees use similar areas in other Central or South American countries. One of the difficulties in conducting surface-based nocturnal observations of manatees, or any marine life, is low ambient light. Some methods that may assist in this research would be the use of spotlights, hydrophones, and sonar. Hydrophones have been 6 used to study feeding by manatees (Bengtson, 1983) and the role of vocalization (O’Shea and Poché, 2006). In Florida, Bengtson and Fitzgerald (1985) found that manatees vocalize most during social activities, and about once every five minutes while resting. Sonar has been used mostly in the development of manatee avoidance technology for boaters in Florida (Frisch and Haubold, 2003) and to detect deep holes used by manatees in Costa Rica (Smethurst and Nietschmann, 1999). There has also been some evidence that manatees in low-visibility water can be detected using sonar (Gonzalez-Socoloske, 2007). Manatees in Florida appear to have no strict preference for locations in which to rest; they will bottom rest on all types of substrate without preference, such as limestone shelves, oyster bars, beds of aquatic plants, and even deep, man-made canals (Hartman, 1979; Reid, pers. comm.). In contrast, manatees in the Drowned Cayes of Belize are known to rest primarily in depressions in the benthic substrate called “resting holes” (Self-Sullivan, pers. comm.). Objectives and Significance Although a number of these resting holes have been found and mapped throughout the Drowned Cayes, no one has yet quantitatively described these resting holes to determine what factors might influence manatees to choose a particular area in which to create and maintain a resting hole. The objectives of this study were to: 1) characterize resting holes and visually represent them using Geographical Information System (GIS) mapping software, 2) correlate diurnal and nocturnal resting hole use by manatees with environmental parameters, 3) investigate various methods for nocturnal 7 detection of manatees and, 4) describe behaviors of manatees around resting holes. The hypotheses being tested were as follows: 1) since manatees require calm waters in which to either rest suspended or lie prostrate on the bottom, then resting holes must have lower currents than other areas of the cayes; 2) since manatees seem to lack a circadian rhythm, their use of resting holes should be identical during day and night; 3) since manatees are affected by many environmental factors, various environmental parameters should influence their use of resting holes; and 4) since manatees have been observed to be more active during changing tides, tides should have an effect on manatee use of and behaviors in resting holes. The current study will give us a better understanding of manatee habitat selection. It is especially an important first-step in understanding diurnal and nocturnal habitat use by manatees, particularly resting hole use, and in delineating which environmental characteristics play a role in these choices. Any coordinated program for managing manatee habitat will rely on this type of information when determining where to focus conservation efforts. 8 CHAPTER TWO MATERIALS AND METHODS Study Site The study site for this investigation was the Drowned Cayes (N17° 28.0', W88° 04.5'; Figure 2.1), which is a string of mangrove islands approximately 13 km long with very little land-mass. They are located approximately 8 km east of Belize City. There are numerous small channels, lagoons, coves and dead-end bogues (channels separating cayes that are maintained by tidal action; Ford, 1991) located throughout these islands. The environment is marine, although there are one or two low salinity seeps amongst the cayes (Self-Sullivan, pers. comm.). There are also extensive seagrass beds surrounding these islands, which make it ideal manatee habitat, providing excellent opportunities for research into habitat preferences of the Antillean manatee. Characterization of Resting Holes Characterization of 12 resting holes in the Drowned Cayes was accomplished by collecting data on depth, distance to shore from resting hole center, type of substrate and benthic vegetation, water velocity, salinity, and temperature through the water column above the resting hole. Initially, two holes were surveyed by two team members using a marked rope to measure the length and width of the holes along their north/south and east/west axes. However, due to poor visibility, the holes were extremely difficult to see and measure, and thus rope measurements were subsequently stopped. 9 Figure 2.1. Map of A) Belize and B) the Drowned Cayes in relation to Belize City. (Background image: Google Earth™ mapping service, 2006).1 1 Belize City: DigitalGlobe®. Acquisition date: Nov 28, 2006. Lat/Long: 17.49953°/-88.1824°. Catalog ID: 558F801. Drowned Cayes: DigitalGlobe®. Acquisition date: Dec 13, 2002. Lat/Long: 17.47585°/-88.03487°. Catalog ID: 17B3802. 10 A Hondex Digital Depth Sounder (Forestry Suppliers, Inc.; accuracy: ±1%) was used to measure depth around the resting holes. Before measuring depth, a short scan (ten minutes) of the area was performed to ensure no manatees were disturbed by boat movement over the resting hole. At the deepest of the resting hole, latitude and longitude were recorded with a Garmin Global Positioning System (GPS) 12 Personal Navigator along with the depth measurement. This deepest point was designated as the center of the resting hole, even if not the geometric center. Sea water was collected with a Student Water Sampler (Forestry Suppliers, Inc.) that allowed water to be sampled at various depths without disturbing the water column. The apparatus was slowly lowered into the water to the appropriate depth as determined by measurement of the sampler’s rope, and then triggered to trap water into the tube and slowly brought back to the surface. When obtaining a sample from the bottom of the resting hole, the sampler was lowered until the rope went slack, then raised slightly to collect water just above the substrate. Since the sampler was lowered from a boat, movement of the boat and current would sometimes require additional rope to be paid out until the sampler reached the bottom. This inherent variability resulted in a few discrepancies between actual depth, as measured by the depth sounder, and the amount of rope required to reach the substrate. Water temperature was measured using a Taylor Classic Instant Read Pocket Digital Thermometer. Salinity was measured using a WP-84 Conductivity-TDSTemperature Meter (TPS Pty Ltd.; accuracy: ±0.05%). Water samples were collected at the water surface and bottom of the resting hole, and at the mid point between the surface and substrate. Similarly, water velocity was measured down the water column at the same 11 points over the resting hole using a General Oceanics Mechanical Flowmeter (Model 2030R, Forestry Suppliers, Inc.; speed range 10 cm/s – 7.9 m/s). A dive weight was attached at one end to make sure the flowmeter stayed in one position while suspended in the water. The flowmeter was taken into the water by a snorkeler to insure that the rotor was not spinning until it was in position and the timer was ready to be started. Once the flowmeter was at the correct depth, the snorkeler released the rotor, and the timer was started. The snorkeler made sure not to kick the water around the flowmeter and matched the overall direction of the flowmeter with a compass to align overall current direction during the five minutes the meter was running. When the five minutes were finished, the snorkeler stopped the rotor and read the rotation counts. Water velocity in cm/s was calculated from rotation counts for five minutes with the following formula provided by General Oceanics Inc., Counts × 26,873 × 100 999,999 5 × 60 (1) where 26,873 is the rotor constant, 999,999 is the maximum number of rotations, all divided by five minutes multiplied by 60 seconds to obtain cm/s. Water velocity was ranked to adjust for unequal variances. Vegetation and soil in the resting holes were identified in broad categories, such as mud or sand, and seagrass or algae. Also, the distance from the resting hole to the nearest mangrove was measured using a Bushnell Yardage Pro 1000 Range Finder (accuracy: ±1 m). Water salinity and temperature were measured at three points parallel to the closest mangroves to detect any low salinity water seeping into the area. 12 Water velocity, temperature and salinity were also measured at various points within and around the Drowned Cayes following the same protocol used for resting holes. These points were chosen to include a variety of habitats (open water, reef, seagrass, lagoons), and were chosen from 54 existing points where other scans have previously been performed, and where manatees have been observed using the habitat (Self-Sullivan, pers. comm.). A 987c SI Combo Humminbird Sonar (Techsonic Industries, Inc.) was used to take several pictures of the resting holes to provide a better idea of hole depth, shape and size of the resting holes. This device transmits a sound wave and determines distance to an object by measuring the time difference between when the wave is transmitted and when it is reflected off an object. The sonar computer can then interpret location, size, and composition of an object using the reflected signal. This produces an image with detailed bottom topography, with a target separation of 6.35 cm. There are three beams emitted by the transducer: one facing down and two pointing to the side at right angles to the boat. Each time the transducer receives the sonar echoes, these are put together on the display to form an image. The rows that are at the top, closest to the boat icon, are the most recent sonar data. These then scroll down the screen as new sonar echoes become available (Figure 2.2). There are two frequencies at which the side beams can be operated. One is 262 kHz, which provides maximum bottom and side coverage (180°), while the other is 455 kHz, which provides the sharpest image with the highest resolution, but covers less area (160°). To take pictures of the resting holes, the 455 kHz frequency was used and the boat was kept moving in a straight line at a constant speed of between one and four km/h to obtain the sharpest image possible. 13 14 Figure 2.2. Side imaging representation from the Humminbird sonar 987c SI Combo (Anonymous, 2005). Diurnal and Nocturnal Scans Four locations were chosen in the Drowned Cayes for diurnal and nocturnal scans. These locations were thought to have at least one resting hole frequently used by manatees based on previous observations (Self-Sullivan, pers. comm.). The names assigned to these scans points are Bogue B Arm 1 End (Bogue B, 17° 27.7044'N, 88° 04.7070'W), Bogue C Cove at End (30 CCo, 17° 28.182285'N, 88° 04.676712'W), Bogue C Big Lagoon (31 CLa, 17° 28.339999'N, 88° 04.859209'W), and Gilroy’s Creek (50 GiCr, 17° 24.653678'N, 88° 04.231572'W). The four sites were visited by boat daily between 10:00 a.m. and 3:00 p.m. and again at night between 7:00 p.m. and 12:00 a.m. The order of the visits depended on factors such as tide and weather, but were randomized as much as possible. As the scan area was approached, the boat engine was turned off, a long pole used to pull the boat to the scan point, and the boat tied to the grounded pole. Scanning time lasted for thirty minutes, starting from the time the engine was turned off. During each scan, the research team continuously scanned for manatees, performing visual 180-degree horizontal “sweeps” from different points on the boat. The team included a boat driver, myself, and from two to eight volunteers. After each site scan, sea and weather conditions were recorded, noting tidal state, sea surface temperature (SST), salinity, sea state, air temperature, precipitation, glare and cloud cover using the same instruments as were used for the resting hole characterizations. Tidal state is defined as ebb (high tide to low tide), low (within one hour of low tide), flood (low tide to high tide), and high (within one hour of high tide). Sea state is taken from the Beaufort Scale (Huler, 2004) where 0 is sea like a mirror, 1 is 15 ripples, 2 is small wavelets that do not break, and 3 is large wavelets with crests beginning to break. Precipitation is categorized as dry, light rain or heavy rain. Glare describes the direction in which glare from the sun or moon can be seen on the water from the viewpoint of the observer. Cloud cover is categorized as clear, scattered clouds, partly cloudy, mostly cloudy and overcast. Data were also recorded for each manatee sighting, including time of sighting, distance to the manatee (as determined with a range scope when possible, otherwise estimated), all specific activities (such as breaths and paddle-dives), general behaviors (such as milling, traveling, and resting), and whether the manatee was inside the resting hole when it was spotted. To see how often these four areas might be used by manatees during the day, each of three scan locations (Bogue B, 30 CCo and 50 GiCr) were observed for a full eighthour day, with all manatee sightings and behaviors recorded. Nocturnal scans were performed in the same manner as daytime scans, except that spotlights were used most nights during the scans to compensate for lack of ambient light. In summer 2005, three lights were used, all of them under 1.5 million candlepower. In summer 2006, four Thor Platinum 15-million Candlepower Rechargeable Spotlights (Cyclops Solutions, LLC) were utilized. At first, in summer 2006, a red filter was placed over the spotlights to minimize disturbance to manatees. This greatly impeded the ability to sight manatees at night, and thus these filters were removed after two nights. As soon as the engine was turned off, one of the spotlights was turned on to help sight any manatees leaving the area. However, in contrast to daytime scans, the thirty minute scan period was not started until the boat was tied to the grounded pole and all 16 four spotlights were turned on. All of these were trained on the mangrove roots from a low angle of incidence, resulting in minimal penetration of the spotlight beam through the water, thus minimizing potential disturbance to any manatees present. The Humminbird side-scan sonar was also used to locate manatees while traveling between scan point locations, both during the day and at night. The frequency used to detect manatees was 262 kHz, which provided the widest beam that would usually reach to the mangrove line in most channels. The side-beams could reach up to 73 m, and were usually extended to the mangrove line unless the channels were wider than that distance or we were at sea traveling from place to place. In the narrow channels, the side beam was as wide as the channels, thus any manatees entering or exiting the channels would be seen. Another element of diurnal and nocturnal scans was the use of a cylindrical, omni directional, broad band hydrophone without low-pass filter (Cetacean Research Technology) to test whether manatees could be detected by their vocalizations. Once the boat was anchored, the acoustic receiver was lowered into the water, and was used for the remainder of the scan time. Behaviors Behaviors were characterized into three main activities, modified from SelfSullivan (2003). “Resting” was defined as an animal staying in one place in the water column with minimal horizontal movement and with no evidence of feeding behavior, but occasionally rising to the surface to breathe. “Traveling” was characterized by the animal swimming in one constant direction. “Milling” was classified as an animal moving in 17 random directions. The time at which breaths were taken were also recorded for as many animals as possible. Data Analysis All statistics were run in Statistical Package for the Social Sciences, version 13.0, with alpha = 0.05. Summary statistics were obtained for resting hole characterizations, diurnal and nocturnal scans of the four locations, and behaviors. Depths for each location had to be adjusted to account for tides. A chart of Belize tides at each quarter of an hour was obtained, and the tide difference was subtracted from the measured depth to obtain the actual depth during low tide for all areas. T-tests were used to test for differences between areas with resting holes and areas without for depth and ranked water velocity. Chi-squares were run to test for significant differences of sightings between years and between day and night. T-tests were run to test for an affect of SST and surface salinity on sightings, as well as test for significance of SST and surface salinity between years, and day and night. A step-wise, logistic regression was used to detect any associations between specific habitat factors and manatee sightings. Factors included location, tide, salinity, SST, day/night, and year. Maps Polygons of the Drowned Cayes were drawn in Google Earth Pro. These were then brought into ArcMap (ArcGIS 9.1) as .kmz files. The KMLer tool was then used to convert these into shapefiles and the merge tool was used to create one shapefile of the outline of the Drowned Cayes. A Google Earth Pro jpeg raster of the Drowned Cayes was 18 then brought into ArcMap and georeferenced using the outline shapefile. The geographic coordinate system for these files was WGS 1984. GPS points associated with depth and water velocity recordings of several scan points were brought into ArcMap over the outline of the Drowned Cayes and symbolized appropriately. GPS points associated with depth recordings of several resting holes were drawn in ArcMap. The Kriging tool was then used to create continuous raster surfaces of each hole. 19 CHAPTER THREE RESULTS Characterization of Resting Holes A total of 12 resting holes were characterized, along with another 20 scan points that did not include resting holes (Table 3.1, Figure 3.1). The mean depths for resting holes was 3.5 ± 0.30 m (n = 12), with a range of 2.0-5.2 m. Non-resting hole areas had a mean depth of 2.0 ± 0.12 m (n = 20), with a minimum of 1.4 m and a maximum of 3.3 m (Table 3.2). There was a significant difference in depth between resting holes and areas without resting holes (t = 4.541, df = 14.624, p < 0.001) with a mean difference of 1.5 ± 0.32 m (Figure 3.2). Mean surface water velocity for resting holes was 0.89 cm/s (n = 10, SD = 1.61), with a minimum of 0 cm/s and a maximum of 5.20 cm/s. Non-resting hole areas had a mean surface water velocity of 4.26 cm/s (n = 20, SD = 5.11), with a minimum of 0.01 cm/s and a maximum of 17.12 cm/s (Table 3.2). There was a significant difference in surface water velocity between resting holes and areas without resting holes (t = -2.880, df = 28, p = 0.008) (Figure 3.3). Caution should be taken when interpreting the water velocity readings, as the threshold for the mechanical flowmeter was 10 cm/s. Many of the readings obtained were below this point. However, the data obtained still allows us to compare water velocity between areas with resting holes and areas without. Salinity by the mangroves near the resting holes ranged from 29.1 ppt to 35.6 ppt (n = 29, x = 33.9 ± 0.32 ppt). Temperature ranged from 29.1°C to 35.5°C (n = 30, x = 20 Table 3.1. Depth and surface water velocity of areas with and without resting holes. Bold denotes locations with resting holes. Location 14 Su2 16 SuMo 25 HuCh 27 HuMo 32 MuGr 34 HeMo 35 BLa 36 BMo 37 AMi 38 AGr 39 NGA 41 ShW 44 BaE 45 BaW 46 FaE 47 FaW 48 GoE 51 GiLa Bogue B Arm 1 End Channel (B1) Bogue B Arm 1 End Cove (B2) 08 G2En 12 StLa 13 Su3 17 F2En 26 HuCo 30 CCo 31 CLa 33 HeCo 43 BaCo 49 GiGr 50 GiCr 54 SLCW Distance to Mangroves (m) 10 38 28 N/A N/A 22 40 52 16 72 28 140 Latitude Longitude Depth (m) 17° 30.2496'N 17° 29.3670'N 17° 28.6344'N 17° 28.3572'N 17° 27.7692'N 17° 27.5328'N 17° 27.5892'N 17° 27.3726'N 17° 27.1074'N 17° 27.1116'N 17° 30.5988'N 17° 26.8020'N 17° 26.5518'N 17° 26.2728'N 17° 26.3154'N 17° 25.9506'N 17° 25.5786'N 17° 24.4782'N 88° 05.1630'W 88° 05.5632'W 88° 04.8624'W 88° 05.3028'W 88° 05.3754'W 88° 04.3788'W 88° 04.6848'W 88° 04.9608'W 88° 04.0650'W 88° 03.9744'W 88° 03.1086'W 88° 04.0644'W 88° 04.0410'W 88° 04.2414'W 88° 04.0734'W 88° 04.3902'W 88° 03.9750'W 88° 04.4922'W 3.3 2.0 2.0 1.5 1.9 1.5 2.0 1.6 1.7 1.4 2.9 2.3 1.8 2.4 2.5 1.5 1.5 2.7 17° 27.6900'N 88° 04.7112'W 2.3 0.60 17° 27.7062'N 88° 04.7484'W 1.5 1.52 17°° 30.8712'N 17°° 30.3888'N 17°° 29.7558'N 17°° 30.0810'N 17°° 28.8966'N 17°° 28.2060'N 17°° 28.3158'N 17°° 27.7548'N 17°° 26.6556'N 17°° 24.6150'N 17°° 24.6612'N 17°° 25.0350'N 88°° 05.7330'W 88°° 05.5254'W 88°° 05.7246'W 88°° 04.8480'W 88°° 04.1766'W 88°° 04.6848'W 88°° 04.8558'W 88°° 04.5252'W 88°° 04.0452'W 88°° 04.1172'W 88°° 04.2726'W 88°° 04.4064'W 2.7 3.1 3.9 2.2 3.3 3.5 3.1 5.2 2.0 5.1 3.0 4.4 0.00 N/A 0.44 0.01 N/A 1.81 0.67 0.09 0.08 5.20 0.00 0.58 21 Surface Water Velocity 0.01 0.38 3.19 1.66 0.06 1.32 1.07 1.77 2.13 4.07 5.41 17.12 0.55 12.50 13.19 9.51 8.47 0.68 Figure 3.1. The Drowned Cayes, Belize, with characterized scan points (Background image: Google Earth™ mapping service, 2006).2 2 Drowned Cayes: DigitalGlobe®. Acquisition date: Dec 13, 2002. Lat/Long: 17.47585°/-88.03487°. Catalog ID: 17B3802. 22 Table 3.2. Descriptive statistics of environmental characteristics of areas with and without resting holes. 23 Depth (m) Sea Surface Temperature (°C) Surface Salinity (ppt) Surface Water Velocity (cm/s) Depth of Middle Sample (m) Middle Sample Sea Temperature (°C) Middle Sample Salinity (ppt) Middle Sample Water Velocity (cm/s) Depth of Bottom Sample (m) Bottom Sample Sea Temperature (°C) Bottom Sample Salinity (ppt) Bottom Water Velocity (cm/s) n 20 20 19 20 11 11 11 11 20 20 20 20 No resting hole present Minimum Maximum Mean ± SE 1.4 3.3 2.0 ± 0.12 30.1 33.8 31.1 ± 0.18 34.0 35.4 34.5 ± 0.08 0.01 17.12 4.26 1.0 1.7 1.2 ± 0.07 30.2 31.7 30.9 ± 0.13 33.8 35.0 34.4 ± 0.10 0 24.66 5.02 1.3 3.4 2.0 ± 0.12 30.0 32.8 30.8 ± 0.13 33.8 35.2 34.5 ± 0.07 0.01 10.47 2.00 Resting hole present n 12 12 12 10 12 12 12 9 12 12 12 10 Minimum 2.0 28.7 29.7 0 1.0 28.6 33.7 0 1.9 28.6 33.8 0 Maximum 5.2 35.6 35.5 5.20 2.7 32.4 35.5 0.63 5.7 32.4 36.4 1.24 Mean ± SE 3.5 ± 0.30 31.4 ± 0.50 34.3 ± 0.46 0.89 1.7 ± 0.15 30.7 ± 0.30 34.8 ± 0.18 0.29 3.5 ± 0.32 30.5 ± 0.31 35.0 ± 0.24 0.26 Significance <0.001 NS NS 0.008 NS NS NS NS NS NS NS NS Figure 3.2. Depth of resting holes and non-resting holes superimposed on the outline of the Drowned Cayes. 24 Figure 3.3. Water velocity of resting holes and non-resting holes. 25 31.5 ± 0.27°C). Mean distance to shore from the middle of resting holes was 45 ± 12.0 m (n = 10), with a minimum of 10 m and a maximum of 140 m. Gilroy’s Creek (50 GiCr) had a very large hole (19.3 m × 30 m) with gently slopping sides that contained three deeper holes with steeper sides (4 m × 5 m). The substrate in and around the main hole was a very silty mixture of mud and some sand, with shell debris that was easily compressed and re-suspended. There was no vegetation inside the resting hole, but instead seemed to have bacterial mats covering the bottom. Outside of the hole, there was very sparse shoal grass (Halodule sp.) which became slightly more abundant near the mangrove roots. The resting hole at Bannister Bogue Jerry's Cove (43 BaCo) was longer (10.4 m East-West) than its width (7.2 m North-South). This was the shallowest hole measured, with maximum depth of only 2.0 m. The substrate in and around the hole was a mixture of very fine and silty sand mixed with mud. There was no vegetation inside or proximal to the hole. Raster interpolation of the resting hole can be seen in Figure 3.4. The substrate in and around the hole at Bogue C Cove at End (30 CCo) was a mixture of very fine and silty sand with mud mixed in. There was no vegetation inside or proximal to the hole. There was a visible “manatee highway” to and from the hole where manatees “slide” along the bottom with their forelimbs and body in contact with the seafloor. This deeper, narrow channel started at the entrance to the cove near the mangroves and followed a straight path to the deeper resting hole (Figures 3.5 and 3.6). The resting hole and surrounding substrate at Bogue C Big Lagoon (31 CLa) was comprised of very fine and silty sand with mud mixed in. There was no vegetation inside the hole and only sparse patches of shoal grass around the hole. The shape of the hole was 26 Figure 3.4. Raster interpolation of depth measurements at 43 BaCo. 27 28 Figure 3.5. Resting holes at 30 CCo and 31 CLa. Red line denotes approximate location of manatee highway (Background image: Google Earth™ mapping service, 2006).3 3 Drowned Cayes: DigitalGlobe®. Acquisition date: Dec 13, 2002. Lat/Long: 17.47585°/-88.03487°. Catalog ID: 17B3802. Figure 3.6. Raster interpolation of depth measurements at 30 CCo. 29 almost circular with steeply sloping sides (Figures 3.7 and 3.8). The substrate in and around the hole at Sugar Bogue Arm 3 End (13 Su3) was a mixture of very fine and silty sand and mud. There was no vegetation inside or around the hole, and it was long and narrow with gentle slopes (Figure 3.9). The water was warmer and fresher on the surface with a clear layer of water, and colder and more saline on the bottom where the water was a milky white color. The resting hole at Gilroy’s Grassbed East Side (49 GiGr) was a very large hole with very gentle slopes (Figure 3.10). The substrate in and around the hole was a mix of mud and silty sand. There was no vegetation inside the hole, although the slopes were sparsely covered with different types of macro algae including Halimeda sp. and Penicillus sp. There was abundant turtle grass (Thalassia testudinum) all around the resting hole (Figures 3.11 and 3.12) and a patch of coral near the south end of the hole (Figure 3.13). The substrate in the hole at Stimpy’s Lagoon (12 StLa) was very coarse shell sand with a very thin layer of silty mud covering the bottom. Mud was the main feature surrounding the hole. There was almost no vegetation inside the hole, except for a few, very small patches of turtle grass. Around the hole was a mixture of turtle and shoal grasses. This hole also had an “arm” (about the size of a full grown manatee) jutting out of the main hole towards the mangroves (Figure 3.14). The deepest hole was at Heraclitus Cove (33 HeCo) and measured 5.2 m at the deepest part. The substrate in and around the hole was very fine silty mud with no vegetation. This was a well defined hole with steep slopes (Figure 3.15). The water was warm at the surface (30.7°C), cooler in the middle (30.3°C) and slightly warmer again at 30 31 Figure 3.7. Sonar image of the resting hole at Bogue C Big Lagoon (31 CLa). The hole can be seen to the right of the image as a large dark area. Figure 3.8. Raster interpolation of depth measurements at 31 CLa. 32 Figure 3.9. Raster interpolation of depth measurements at 13 Su3. 33 Figure 3.10. Raster interpolation of depth measurements at 49 GiGr. 34 35 Figure 3.11. Sonar image of the resting hole at Gilroy’s Grassbed East Side (49 GiGr). The boat is moving forward towards the top of the image and just leaving the resting hole. There is a sharp division between bare mud in the hole and grass as one is leaving the hole. The edge of the grass is at the top of the image. 36 Figure 3.12. Close-up of 49 GiGr and 50 GiCr. The resting hole is clearly seen at 49 GiGr as the bluer round circle. The darker areas surrounding it are seagrass beds (Background image: Google Earth™ mapping service, 2006).4 4 Drowned Cayes: DigitalGlobe®. Acquisition date: Dec 13, 2002. Lat/Long: 17.47585°/-88.03487°. Catalog ID: 17B3802. 37 Figure 3.13. Sonar image of the resting hole at Gilroy’s Grassbed East Side (49 GiGr). At the south edge of the resting hole is a small patch of coral which can be seen in this picture as protrusions on either side of the boat transect. Figure 3.14. Raster interpolation of depth measurements at 12 StLa. 38 Figure 3.15. Raster interpolation of depth measurements at 33 HeCo. 39 the bottom (30.5°C). The substrate in and around the hole at Spanish Lookout Caye West (54 SLCW) was mud. This hole was much bigger than the other resting holes with very gentle slopes (Figure 3.16). Patches of turtle grass and shoal grass occurred separately but were interspersed with mud patches inside the hole. Around the hole were patches of turtle grass and patches of shoal grass. The resting hole and surrounding substrate at Bogue G Arm 2 End (08 G2En) was comprised of a mixture of fine sandy mud. This hole had a gentle slope (Figure 3.17) with patches of mostly shoal grass and one patch of turtle grass on the slopes. There was no vegetation along the bottom of the hole. Around the hole were patches of algae, shoal grass and mud. Caution should be taken when interpreting the raster interpolation images, especially where very few depth readings were taken, such as 31 CLa, where only nine readings were collected. These images give us a general idea of what the resting holes look like by mathematically interpolating the area around the depth readings. Some margin of error is expected with these images and since the substrate is so soft and easily stirred, this is an approximation of what the hole looked like during the summer of 2006. Diurnal and Nocturnal Scans I completed 168 scans. Of these, 81 were during 2005 and 87 were during 2006. Manatees were sighted during 39 scans, 26 during 2005 and 13 during 2006 (Figure 3.18). Numbers of sightings and scans are summarized in Table 3.3. The number of sightings was divided by the number of scans for each location to calculate the percent of 40 Figure 3.16. Raster interpolation of depth measurements at 54 SLCW. 41 Figure 3.17. Raster interpolation of depth measurements at 08 G2En. 42 35 Frequency 30 25 20 15 43 10 5 0 2005 2006 Year Figure 3.18. Total number of scans and sightings per year for each location where = Bogue B, = 50 GiCr. Bars with patterns represent scans while plain colored bars represent sightings. = 30 CCo, = 31 CLa, and Table 3.3. Summary of all sightings and scans. All Sightings All Scans 2005 Sightings 2005 Scans 2006 Sightings 2006 Scans All Day Sightings All Day Scans All Night Sightings All Night Scans 2005 Day Sightings 2005 Day Scans 2005 Night Sightings 2005 Night Scans 2006 Day Sightings 2006 Day Scans 2006 Night Sightings 2006 Night Scans Bogue B 5 39 3 19 2 20 4 25 1 14 2 13 1 6 2 12 0 8 30 CCo 10 38 6 16 4 22 10 25 0 13 6 12 0 4 4 13 0 9 44 31 CLa 10 37 8 17 2 20 10 26 0 11 8 13 0 4 2 13 0 7 50 GiCr 14 54 9 29 5 25 10 31 4 23 5 18 4 11 5 13 0 12 Total 39 168 26 81 13 87 34 107 5 61 21 56 5 25 13 51 0 36 sightings for each category (Table 3.4). The number of sightings was also divided by the total number of scans for both years (168) for comparison (Figure 3.19). A chi-square indicated that there was a significant difference in the number of sightings between years, (χ2 = 6.926, p = 0.008). Usually, a sighting was of just one manatee (32 times), although dyads were sighted six times, and three manatees were sighted together one time at 31 CLa. Mean SST for all four locations was 31.0 ± 0.10°C for 2005 and 30.5 ± 0.13°C for 2006 (Table 3.5). Mean sea surface salinity was 36 ± 0.1 ppt for 2005 and 32 ± 0.3 ppt for 2006. SST and salinity were significantly different between years (SST: t = 3.292, df = 158.460, p = 0.001; salinity: t = 20.629, df = 166, p < 0.001). Other descriptive statistics for scans are found in Table 3.6. Mean SST when manatees were sighted was 30.9 ± 0.19°C and 30.7 ± 0.09°C when no manatees were sighted. Mean surface salinity when manatees were sighted was 35 ± 0.31 ppt and 34 ± 0.27 ppt when no manatees were sighted (Table 3.7). A t-test indicated a significant difference in surface salinity for sighting of manatees (t = 2.312, df = 166, p = 0.022). One manatee was first observed visually during the day, and then detected with the sonar near Bogue C (Figure 3.20). There were 107 day scans and 61 night scans, with 34 day sightings and 5 night sightings (Figure 3.21). There was a significant difference in the number of sightings between day and night (χ2 = 12.118, p < 0.001). Mean SST was significantly different for day versus night (t = 3.236, df = 162.668, p = 0.001) with SST being 30.9 ± 0.12°C during the day and 30.4 ± 0.10°C during the night (Table 3.8). There was also a significant difference in air temperature between day and night (t = 10.649, df = 68.173, p < 0.001) with air temperature being 31.5 ± 0.12°C during the day and 29.4 ± 0.08°C 45 Table 3.4. Sightings were divided by number of scans for each location and category to give a percentage of sightings per scans. All Sightings per Number of Scans 2005 & 2006 All Sightings per Number of Scans 2005 All Sightings per Number of Scans 2006 All Day Sightings per Number of Scans 2005 & 2006 All Night Sightings per Number of Scans 2005 & 2006 All Day Sightings per Number of Scans 2005 All Night Sightings per Number of Scans 2005 All Day Sightings per Number of Scans 2006 All Night Sightings per Number of Scans 2006 46 Bogue B 30 CCo 31 CLa 50 GiCr 13% 26% 27% 26% 16% 38% 47% 31% 10% 18% 10% 20% 16% 40% 38% 32% 7% 0% 0% 17% 15% 50% 62% 28% 7% 0% 0% 17% 17% 31% 15% 39% 0% 0% 0% 0% 9 8 7 Percent 6 5 4 47 3 2 1 0 All Sightings 2005 Sightings 2006 Sightings All Day Sightings All Night Sightings 2005 Day Sightings 2005 Night Sightings 2006 Day Sightings 2006 Night Sightings Figure 3.19. Sightings at each location and for each category were divided by total number of scans (168) for comparison purposes where = Bogue B, = 30 CCo, = 31 CLa, and = 50 GiCr. Table 3.5. Mean, minimum, maximum sea surface temperature, salinity and air temperature per year for all four scan points. Sea Surface Temperature (°C) Surface Salinity (ppt) Air Temperature (°C) n 81 81 81 Minimum 28.7 33 27.0 2005 Maximum 34.3 39 32.9 2006 Mean ± SE 31.0 ± 0.10 36 ± 0.1 31.1 ± 0.14 n 86 87 87 Minimum 27.2 13 26.6 Maximum 32.9 35 33.4 Mean ± SE 30.5 ± 0.13 32 ± 0.3 30.4 ± 0.16 Significance 0.001 <0.001 NS 48 Table 3.6. Descriptive statistics for habitat factors when manatees were sighted compared with times when no manatees were sighted. Sea State Cloud Cover Tide State Glare Precipitation Smooth Ripples Small Wavelets Large Wavelets Clear Scattered Partly Mostly Overcast Ebb Low Flood High North East South West Omni None Dry Light Rain Heavy Rain All Data 1 130 36 1 13 70 42 32 6 53 25 61 29 1 4 1 20 3 130 160 6 1 49 Manatee Sighted 1 33 5 0 1 13 10 12 2 13 4 12 10 0 2 1 4 1 31 35 4 0 No Sighting 0 97 31 1 12 57 32 20 4 41 21 49 19 1 2 0 16 2 99 125 2 1 Table 3.7. Sea surface temperature and salinity when a manatee was sighted and when no manatees were sighted. Sea Surface Temperature (°C) Surface Salinity (ppt) Air Temperature (°C) n 39 39 39 Manatee Sighting Minimum Maximum Mean ± SE 28.6 34.2 30.9 ± 0.19 31 39 35 ± 0.3 26.6 33.4 31.2 ± 0.25 No Manatee Sighting n 128 129 129 Minimum 27.2 13 27.4 Maximum 34.3 37 33.0 Mean ± SE 30.7 ± 0.09 34 ± 0.3 30.6 ± 0.12 Significance NS 0.022 NS 50 51 Figure 3.20. Sonar image taken in the channel of Bogue C. A manatee can be seen as a dark form to the upper right of the boat transect. 35 30 52 Frequency 25 20 15 10 5 0 Day Figure 3.21. Day and night scans and sightings for each location where = Bogue B, Bars with patterns represent scans while plain colored bars represent sightings. Night = 30 CCo, = 31 CLa, and = 50 GiCr. Table 3.8. Descriptive statistics for day and night scans of four locations. Sea Surface Temperature (°C) Surface Salinity (ppt) Air Temperature (°C) n 106 107 107 Minimum 27.2 13 26.6 Day Maximum 34.3 39 33.4 Night Mean ± SE 30.9 ± 0.12 34 ± 0.3 31.5 ± 0.12 n 61 61 61 Minimum 28.6 28 27.4 Maximum 31.8 37 32.0 Mean ± SE 30.4 ± 0.10 34 ± 0.3 29.4 ± 0.08 Significance 0.001 NS <0.001 53 during the night. Mean surface salinity for all four locations was 34 ± 0.3 ppt during the day and 34 ± 0.3 ppt during the night. Even though the hydrophone was used multiple times during both day and night, no manatee sounds were ever heard with the aid of the apparatus. Even when two manatees were seen resting together in one resting hole, no manatee vocalizations were detected. A forward, step-wise logistic regression was conducted to determine which independent variables (location, tide, salinity, SST, day/night, and year) were predictors of manatee sightings. The overall model indicated that two predictors (day/night and year) were statistically reliable in distinguishing between sighting and non-sighting of manatees (-2 Log Likelihood = 162.326; χ2(2) = 19.206, p < 0.001). The model correctly classified 76.6% of the cases, although it did this by predicting all cases as “no sighting.” Regression coefficients are presented in Table 3.9. Wald statistics indicated that day/night and year significantly predicted manatee sighting, where manatees were more likely to be sighted during the day and in 2005. Behaviors The mean initial distance to an animal when sighted was 57 ± 5.7 m, with a minimum of five meters and a maximum of 150 m (n = 44). Presence of the boat appeared to disturb manatees during a scan (such that the animal would change behavior) during six occasions out of 47 (eight instances were undetermined). A manatee was seen resting 21 times, traveling 11 times, milling six times, and manatee behavior was undetermined nine times. A manatee was seen resting in Bogue B only once compared 54 Table 3.9. Forward stepwise logistic regression results. Variables B SE Wald df Significance Year Day/Night Constant -0.894 -1.605 2.864 0.397 0.516 0.526 5.075 9.676 29.658 1 1 1 0.024 0.002 <0.001 55 Odds Ratio 0.409 0.201 17.537 with multiple times in the remaining three sites (Table 3.10). Manatees were seen resting more often during ebb and flood tides than during low and high tides, (Table 3.10), although this was not significant. SST and surface salinity for each behavior are summarized in Table 3.11. Average time between breaths while resting for five manatees was 130 ± 23.9 seconds (n = 53). The shortest time between surface intervals was 8 seconds and the longest was 528 seconds. While resting, manatees seemed to take two to four quick breaths (less than one minute between each), followed by longer intervals (four to eight minutes) (Table 3.12). Nine hours were spent at 30 CCo scanning for manatees from 9:50 to 18:50. Two manatees were sighted during the day, one at 10:39 and one at 14:31. The first manatee came into the cove then left immediately. The second manatee stayed in the cove for approximately 30 minutes where it was resting most of the time. This manatee was never seen leaving the cove. Therefore, it is unknown whether this individual swam out of the area entirely or went to another resting hole located further up a nearby creek. Another eight hours were spent at Bogue B Arm 1 End from 8:31 to 16:31. A manatee was sighted at 11:27 traveling through to another creek on the other side of the cove leading to a deadend lagoon (Figure 3.22). Presumably the same manatee was seen at 12:33 exiting the creek to leave through the entrance of the cove. Finally, eight and a half hours were spent at 50 GiCr. No manatees were sighted during this observation. After reviewing behaviors in Bogue B Arm 1 End and snorkeling through the area, it was decided that this area is likely not a resting hole. There is no deeper depression that would constitute a resting hole; instead, the whole area is a uniform 1.5 m 56 Table 3.10. Behaviors of manatees at each location and during different tides. Resting Traveling Milling Undetermined Bogue B 1 3 1 0 Locations 30 CCo 31 CLa 7 7 3 3 3 2 1 4 50 GiCr 6 2 6 4 Ebb 9 3 2 1 Low 1 2 1 1 Tides Flood 10 4 1 0 High 1 2 2 7 57 Table 3.11. Behaviors of manatees and associated sea surface temperatures and salinities. Resting Traveling Milling Undetermined n 21 11 6 9 Surface Sea Temperature (°C) Minimum Maximum Mean ± SE 29.1 34.2 31.1 ± 0.23 28.6 32.9 30.9 ± 0.44 29.5 31.6 30.8 ± 0.32 28.7 31.9 30.7 ± 0.38 Surface Salinity (ppt) n 21 11 6 9 Minimum Maximum Mean ± SE 32 39 35 ± 0.5 32 37 35 ± 0.5 31 37 35 ± 0.9 33 37 35 ± 0.5 Table 3.12. Intervals (in seconds) between breaths for each manatee while resting. Interval 1 Interval 2 Interval 3 Interval 4 Interval 5 Interval 6 Interval 7 Interval 8 Interval 9 Interval 10 Interval 11 Interval 12 Interval 13 Interval 14 Interval 15 Interval 16 Mean ± SE Manatee 1 60 420 24 10 326 60 36 381 22 21 342 34 Manatee 2 388 146 116 145 ± 48.1 217 ± 86.1 Manatee 3 13 12 15 12 528 11 16 17 474 12 20 27 527 9 15 22 108 ± 49.1 58 Manatee 4 480 8 20 135 407 11 28 58 405 10 8 12 Manatee 5 14 64 12 277 390 15 22 19 354 10 132 ± 53.3 118 ± 49.6 59 Figure 3.22. Close-up of Bogue B. The red arrow indicates the manatee’s path as it came into the cove the first time and exited up the creek to the north (Background image: Google Earth™ mapping service, 2006).5 5 Drowned Cayes: DigitalGlobe®. Acquisition date: Dec 13, 2002. Lat/Long: 17.47585°/-88.03487°. Catalog ID: 17B3802. depth and is completely flat with patches of shoal grass throughout. There were many feeding scars where manatees had dug into the mud with their noses to eat the rhizomes, which were exposed. There were also fore-limb marks from manatees “walking” on the substrate. However, no manatees were seen resting in Bogue B during the summer of 2006 and only one manatee was seen resting in the cove during the summer of 2005. My data suggest that this area is not a regular resting area, but likely serves as a feeding area and path for traveling to the other cove. 60 CHAPTER FOUR DISCUSSION Characterization of Resting Holes Areas with resting holes were deeper than areas without resting holes, with a mean depth of 3.5 m and a maximum of 5.2 m. This is deeper than the reported preferred depth of 3.0 m for manatees found in Florida and Puerto Rico (Hartman, 1979; Lefebvre et al., 2000), but concurs with the findings of Olivera-Gomez and Mellink (2005) from Bahia de Chetumal in Mexico where manatees used habitats of deeper depth (3.1-4.5 m) more often. In Crystal River and Homosassa Spring, Florida, manatees will usually rest in less than 2 m of water, although they have been seen resting in deeper areas (Hartman, 1979). However, it is not known if manatees in the Drowned Cayes choose these resting areas because there is already a depression present or if they create the deeper areas themselves. Surface water velocity was also significantly slower for areas with resting holes. Since most of the resting holes are located in dead end bogues, lagoons and coves close to mangroves (about 45 m away), they are well protected, and there is little current and wave action. This would allow manatees to rest in these areas without having to fight against a strong current. This is consistent with other studies where manatees are found in areas sheltered from high currents, surf and wind (Hartman, 1979; Lefebvre et al., 2001; Olivera-Gomez and Melink, 2005). Manatees in Puerto Rico, which are most often found in a marine environment, were also found in calm waters by the coast (Powell et al., 61 1981) and manatees in Nicaragua tended to select watercourses with slow currents (Jiménez, 2005). Compared to other resting holes, the surface water velocity at 49 GiGr was relatively fast, since this area is not located in a dead-end bogue or creek, but rather in an open area outside of the mangrove islands. However, it is still used regularly by manatees as a resting hole. This was one of the deeper holes at 5.1 meters, and there was a difference between the surface water velocity of 5.2 cm/s and the bottom water velocity of 0.39 cm/s. This hole might be deep enough with little current on the bottom to be attractive to manatees as a resting area, especially considering the large beds of varied seagrass all around this resting hole. A similar situation exists west of Swallow Caye, which represents the largest resting hole in the study area (Self-Sullivan, pers. comm.). Most of the resting holes did not have any vegetation growing in them. If there was any seagrass present prior to a resting hole being maintained, the absence of seagrass now was likely due to a combination of manatees eating any vegetation present and manatees resting in the same area, thus preventing seagrass from re-establishing. Additionally, the deepest holes may be too deep to allow enough light to penetrate the water for healthy growth of seagrass, since re-suspension of the fine sediments occurs easily and could impede sufficient, consistent light intensity for good growth. Although there was no seagrass growing in the resting holes themselves, there were many seagrass beds found close to the resting holes, which is an important characteristic in determining manatee habitat preference (Hartman, 1979; Deutsch et al., 2003). This would allow a manatee to travel a short distance between resting sites and feeding sites in calm waters. In Puerto Rico, Powell et al. (1981) noted the association of 62 manatees and seagrass beds. It is likely that the most important manatee food items there are turtle grass (Thalassia testudinum), manatee grass (Syringodium filiforme), and shoal grass (Halodule wrightii), which are also the most common seagrass beds found in and around the Drowned Cayes. In Mexico, manatee movements in Chetumal Bay were most strongly associated with food distribution (Axis-Arroyo et al., 1998). In Florida, aerial surveys have indicated a strong association between location of manatees and distribution of seagrass beds. In fact, manatee density was positively correlated with seagrass abundance (Halodule wrightii and Syringodium filiforme) in the northern Banana River (Kinnaird, 1985; Provancha and Hall, 1991). Since manatees are so large and dependent on a low-energy and low-protein diet, they must spend a large proportion of their time feeding to meet metabolic requirements. Manatees tend to feed from 5 to 8 hours per day, consuming about 7% of their body weight in wet vegetation daily (Hartman, 1979; Bengtson, 1983; Etheridge et al., 1985). Manatees should, therefore, choose foraging strategies that allow them to maximize food intake while minimizing energy output. The more familiar an animal is with locations of essential resources, the more efficient it will be in energy acquisition since this reduces time spent searching for new resources (Deutsch et al., 2003). Also, manatees returning repeatedly to graze on the same seagrass beds may promote growth of early successional species (such as Halodule wrightii) preferred by manatees (Lefebvre et al., 2000). The abundance of dugongs tends to also be correlated with areas of available seagrass (Bayliss and Freeland, 1989) where they will practice cultivation grazing, which maximizes the abundance of preferred seagrass species (Preen, 1995). In the Drowned Cayes, proximity of seagrass beds to resting holes would allow manatees to expend little energy in finding 63 food, as well as possibly causing seagrass beds to be populated with favored food through repeated grazing. In the Drowned Cayes, manatees use these resting sites so frequently that the holes have “paths” leading to them, likely created and maintained by the manatee’s body as it approaches the resting hole from shallower water. The combination of “walking” with their forelimbs and dragging their bodies and tails along these paths maintains these “highways” that are used on a regular basis to move to and from the holes. Recently-used holes often have marks in the substrate left by a resting manatee’s forelimbs and paddle. Although manatees choose these areas to rest in regularly, it is unknown if they purposefully excavate these holes, or if the hole itself is created as an artifact of resting in one area repeatedly. Since the substrate in these holes tends to be very silty and muddy, it is easily stirred up, and by resting regularly in the same place, the weight of a manatee could easily compress the mud underneath its body or stir it up and out of the area as the animal descends from taking a breath. This repeated activity may be enough to maintain the depression. Although the sonar techniques I used were not especially helpful for detecting manatees, the Humminbird sonar did provide an overall idea of the depth changes around a resting hole as the boat was driven over the area, as well as confirm the presence/absence of seagrass on the sea floor, where seagrass bed boundaries begin and end, and other features (such as corals) located on the substrate. Since seagrass beds are an important source of food for manatees (Powell et al., 1981; Kinnaird, 1985; Provancha and Hall, 1991; Axis-Arroyo et al., 1998), this tool would be useful to increase our understanding of where major seagrass beds and boundaries are located. It would also 64 allow possible discoveries of other resting holes. The sonar would enable fine characterization of manatee habitat and areas used by other animals (such as the American saltwater crocodile, Crocodylus acutus) that are found in deeper waters or murky rivers and lagoons. No low salinity seeps were detected near any of the resting holes sampled. Thus, my data suggest that manatees likely do not rest in the holes studied due to availability of nearby low salinity water, at least not during the warmer rainy season. There are a variety of behavioral, ecological, and physiological lines of evidence to suggest that drinking fresh or low salinity water may be required for osmoregulation when feeding on hypertonic vegetation (Hartman, 1979; Powell and Rathbun, 1984; Ortiz et al., 1998, 1999). A lack of freshwater for extended periods could lead to dehydration, although it seems that, at least in the short term, manatees may be able to obtain water produced from the oxidation of fat in their bodies (Ortiz et al., 1999). Even though manatees in Puerto Rico are primarily observed in marine habitats, Powell et al. (1981) observed that 85.8% of manatee sightings were within 5 km of natural or artificial freshwater sources. Also, interviews with local Puerto Rican fisherman indicated that manatees visit nearby rivers often to obtain freshwater (Powell et al., 1981). According to Lefebvre et al. (2001), this seems to be a common occurrence with manatees that tend to be found most often in a marine environment, such as in the Dominican Republic and Cuba. In fact, degradation of rivers in Cuba due to siltation, plantation run-off, and pollution may have caused manatees to shift to a more marine environment. Many areas where manatees are considered mostly marine, such as in Puerto Rico, Cuba, and the Dominican Republic lack the occurrence of ideal, broad, freshwater rivers that are deep enough for manatee 65 use (Lefebvre et al., 2001). Thus manatees may prefer freshwater areas to live in, such as rivers, but due to anthropogenic factors, may shift to a more marine environment while still requiring freshwater sources nearby from which to drink. Therefore, it is likely that manatees in the study area would also require freshwater sources. Potentially, manatees may access sufficient freshwater from low salinity seeps, or freshwater lenses following a heavy rain (Self-Sullivan, pers. comm.), or they may need to travel to the nearby Belize River (approximately 8 km away) to meet their osmoregulation needs, or to other nearby coastal rivers. The resting hole at 13 Su3 had fresher water at the surface, probably due to heavy rainfall prior to sampling. Thus, there was a freshwater lens at the surface of the resting hole area. This may allow a manatee to obtain freshwater at the surface during or right after a heavy downpour in more saline areas, but more study is needed to support this idea (Hartman, 1979). These holes seem to be used differently than the deep holes found in other Central American countries. Holes in Venezuela, Costa Rica and Nicaragua are used mainly during low tides or the dry season, are located in fresh or brackish water, and tend to be much deeper (6 - 12 m) than the holes in the Drowned Cayes (O’Shea et al., 1988; Smethurst and Nietschmann, 1999; Jiménez, 2002). They also seem to be more of a refuge from low waters or hunting pressure, rather than a main area in which to rest. Since the holes investigated in the Drowned Cayes were located in areas with minimal tidal range, there is no need to seek refuge in one of these areas during low water levels. In fact, tidal state did not have a significant effect on behavioral state of manatees. Instead, manatees appear to regularly seek these areas in which to rest. 66 Resting habitat choices in the study site are also markedly distinct from that of manatees in Florida. As opposed to resting in particular sites which are deeper than the surrounding areas, Florida manatees will rest on any substrate, from limestone shelves and oyster bars to beds of aquatic plants and sandy areas. Some animals have been seen buried in clumps of dense vegetation while resting. Also, they tend to rest in approximately 2 m of water, although manatees have been observed resting in deeper areas. If there is any current, Florida manatees will usually face the direction of the current when bottom resting (Hartman, 1979). Since manatees were frequently encountered at previously detected resting holes within the Drowned Cayes, and these holes are usually found in quiet coves and dead-end bogues with minimal current, it seems likely that manatees choose some of these areas because of the quiet conditions and create, or at least maintain, resting holes in these places. However, it is not known if an individual manatee will return to one preferred hole or will use any of the resting holes in the Drowned Cayes indiscriminately. Also, it is unknown how manatees learn about the presence of these specific resting holes. In Florida, manatees show strong fidelity to warm-season and winter ranges. During a particular season, individual manatees used a core area that encompassed about 90% of daily locations. Most manatees would return to the same seasonal ranges year after year. In the same study, tracking of several calves with their mothers and then as independent subadults, seemed to indicate strong natal philopatry to specific warm-season and winter ranges, as well as to migratory patterns (Deutsch et al., 2003). Manatees might learn of specific resting holes when resting there with their mothers as calves, and return to these same areas as adults. 67 Diurnal and Nocturnal Scans There was a significant difference in manatee sightings between years. Twice as many manatees were seen in 2005 than in 2006. The reasons for this are unclear, but longer term data for all 54 scan points detected no significant difference in the probability of encountering manatees over the period of 2001-2004 (Self-Sullivan, 2007). There was a significant difference in temperature and salinity for the four scan locations between these two years, with 2005 being warmer and more saline. However, the differences in temperature (0.5°C) and salinity (4 ppt), although significantly different between years, are relatively small and would probably not be significant to manatees given what we know about their range. There was also a significant difference in salinity when manatees were sighted as opposed to when no manatees were sighted, but similarly, the difference was 1 ppt, which is relatively small. Although manatees can sense water temperature and salinity differences (Deutsch et al., 2003), the differences of the readings in this study seem small enough to insignificantly affect manatee habitat choices. Thus the disparity in number of manatee sightings at resting holes between years may be due to random variability or other, as yet, unknown factors. Even though manatees were not seen as often in 2006, this does not necessarily mean that their use of the Drowned Cayes habitat had decreased, but rather that they were not spotted as often at the particular resting holes being studied. Manatees may have been resting at other resting holes more often or they may have been located in other parts of the Drowned Cayes, such as the seagrass beds where they feed. The number of day sightings was also significantly different than the number of night sightings. While no manatees were sighted at night in 2006, five manatees were 68 sighted at night in 2005, although this may be an overestimation. One of the night sightings was hard to identify as a manatee, and may have been a dolphin breathing at the surface and making splashing noises. Another night sighting was of a “footprint” (a flat spot on the surface of the water created by the tail of a manatee when moving up and down). While this footprint was seen by all observers, the sighting was never confirmed by seeing the actual manatee. Another “footprint” was thought to be seen at 30 CCo in 2006 by the researcher and an intern when the spotlights were first turned on, but nothing else was seen or heard, and this footprint was not seen by anyone else. It was difficult to confirm whether this was created by a manatee or another animal, such as a fish, and thus this was not included as a sighting. The hydrophone was used to listen for manatee vocalizations while the animals were in or around resting holes during multiple 30 minute scans. However, no animal was ever heard vocalizing, not even when two manatees were observed using the same resting hole at the same time. This is in contrast to Florida manatees, which have been heard vocalizing about once every five minutes while resting (Bengtson and Fitzgerald, 1985). My results suggest that the hydrophone is likely not a useful tool for detection of resting manatees, but may be more useful in detecting feeding manatees. One concern about using spotlights for night scans was that bright lights might be a disturbance to the manatees. In 2005, a manatee passed under the boat during a night scan, and seemed to avoid the light when it was shining directly on the animal right from the surface. However, disturbance by the lights should have been minimal since the boat was always located at a distance from the resting holes and the spotlights were pointed at the mangrove roots rather than directly into the water. This provided better lighting as it 69 reflected from the mangrove roots rather than being absorbed by the water. It was felt that these lights would cause no more disturbance to the manatees than the presence of our research vessel, which occasionally disturbed a manatee and caused it to leave the area even during the daytime. It is true that there were more lights used in 2006 than in 2005 and that these were also brighter. These may have been bright enough to disturb any manatees in the resting holes and keep others wanting to approach the area away from where the lights were shining. This might explain the disparity in nocturnal manatee sightings between 2005 and 2006. Manatees can often be heard taking a breath as well as sighted visually, although no breaths were heard at night. Also, several scans were conducted by moonlight to investigate presence of manatees when no spotlights were present to disturb them. Therefore, it was felt that with the combination of being able to hear manatee breaths, sonar use between holes, and using the lights at night, some manatees should have been spotted if manatees were using the holes at night as regularly as during the day. Even if the manatees were disturbed by the lights enough to leave or avoid coming to the resting hole, a manatee breath and some water movement would have been heard from manatees exiting or entering the area. Therefore, my data suggest that manatees may be using resting holes differently between day and night. If manatees are not using resting holes as often during the night, then they must be using other habitats during that time. In Florida, researchers have observed manatees leaving the warm-refuge of Blue Spring in the late afternoon to feed and remain all night feeding, returning to the refuge in the morning. They abandoned this nocturnal/diurnal cycle once winter was over and water temperatures warmed (Bengtson, 1981; Rathbun et 70 al., 1990). Interviews in Nicaragua indicated a similar nocturnal/diurnal pattern where manatees would leave deep, quiet waters where they were resting during the day, and go feed during late afternoon, night and early morning (Jiménez, 2002). Therefore, one possibility for lack of manatee sightings in resting holes at night is that they leave these areas during the night to feed among the seagrass beds. Another possible reason to use habitats in the Drowned Cayes differently during the day and night is need of freshwater. Manatees in the Drowned Cayes may return to the Belize River and its environs at night, which is the closest source of substantial freshwater. Since boat traffic in and around the city would be much lower than during the day (when it can be very heavy), this may be an opportune time to acquire needed freshwater with minimal disturbance and without having to avoid numerous boats. They may also return to other coastal rivers, such as the Sibun River, which is further south. In Florida and Costa Rica, manatees have been known to avoid areas of high boat traffic (Buckingham et al., 1999; Smethurst and Nietschmann, 1999). In fact, Florida manatees have been observed changing their behavioral state and moving out of a geographical area in response to the noise of approaching vessels (Miksis-Olds, 2006). In Costa Rica, manatees have been observed developing nocturnal activity with high human presence (Jiménez, 1999 in Jiménez, 2005). A third possibility that compliments seeking substantial freshwater sources is that of removal of ectoparasites, as suggested by Olivera-Gomez and Mellink (2005). Hartman (1979) also suggested that manatees may seek to rid themselves of barnacles and marine ectoparasites in freshwater. Aquaculture industries often use freshwater to rid their marine fish of ectoparasites (Pironet and Jones, 2000). Thus, returning to the Belize 71 River, or other nearby coastal rivers, may allow manatees to acquire needed freshwater to drink as well as remove external parasites. However, more study is needed to determine if manatees are really absent in most of the Drowned Cayes at night, and if so, where else they may be during the night. Behaviors Not surprisingly when manatees were sighted in the resting holes, they were usually resting. There were several instances when a manatee would come into the area while the boat was already there, mill around for a short while, then leave. Often in this situation, the manatee would come near the boat to investigate it. Presence of the boat might have discouraged these animals from staying to rest in the hole. Also, manatees that were categorized as milling when the boat first came into the area might have been resting at first, but arrival of the boat may have disturbed them. Since it is not know what the manatees were doing before the boat came into the area, it is difficult to discern whether or not these manatees were resting. When a whole day was spent at the 30 CCo site, two manatees were observed. One came a little distance into the cove, but not as far as the resting hole, then left rapidly. This animal might have been trying to evade a group of boats that were in the vicinity and that had driven to the resting hole a short while earlier. These were four small zodiac-like boats with keels, and were driven at high-speed by tourists led by a guide. The manatee was seen coming in and going out shortly afterwards. This is consistent with studies in Florida where approaching boats elicited a flight response from manatees (Nowacek et al., 2004). A second manatee was observed resting for about 30 minutes 72 later in the day. Even though presence of the boat did not disturb the manatee, we lost sight of this animal after 30 minutes, and did not see it exiting the area. It might have gone up a nearby creek where another resting hole was located, and continued resting there. However, if the manatee was only resting for 30 minutes, this may indicate that manatees in the Drowned Cayes rest for shorter periods of time than Florida manatees, which tend to rest in sessions of two to four hours (Hartman, 1979), although more study is needed to verify this. Since no manatees were seen all day at 50 GiCr, it is possible that manatees do not use all of the resting holes every day, but may instead use particular holes depending on where they are located and where they have been feeding. Ecological Implications Most of the resting holes in the Drowned Cayes are located in quiet lagoons, coves, and dead-end bogues where there is little current or boat traffic. Also, there are numerous healthy seagrass beds nearby on which manatees can feed. This concurs with findings in Puerto Rico, Jamaica, Dominican Republic, and Cuba where manatees were most commonly sighted along the coastline where there were numerous quiet bays, calm seas, extensive seagrass beds, and freshwater sources (Powell et al., 1981; Hurst, 1987; Lefebvre et al., 2001). In addition to these habitat characteristics, manatees in Costa Rica used areas where boat traffic was limited and where there was the least amount of habitat degradation and pollution (Smethurst and Nietschmann, 1999). Bonde et al. (2004) suggested that manatees may be sentinels of marine ecosystem health. According to these 73 researchers, the concept of ecosystem sentinels is valuable, since it provides environmental scientists with specific indicators of environmental damage which may in turn reflect the quality of overall health in marine ecosystems. In particular, manatees can serve as excellent sentinels of harmful algal blooms, as they tend to succumb relatively quickly due to their high sensitivity (O’Shea et al., 1991; Bossart et al., 1998). Also, as manatees come in direct contact with the benthic substrate, they may be among the first animals to encounter contaminants such as copper-based herbicides used to manage aquatic vegetation as well as run-off from agricultural fertilizers and pesticides (Bonde et al., 2004). Prior studies have analyzed manatee tissues and established baseline contaminant levels for a variety of pollutants (Ames and Van Vleet, 1996). Since pollution can also indirectly have an effect on manatees by reducing their food supply and thereby reducing reproduction, changes in manatee population distribution could serve as an early warning of environmental damage. These “red flags” could make it possible for scientists and politicians to enact earlier and more effective responses to potentially harmful conditions, allowing for more effective management of natural resources (Bonde et al., 2004). Whether manatees deliberately or incidentally create resting holes, it is apparent that their presence changes the environment they live in, not only in creating or maintaining these depressions, but also when grazing in the abundant seagrass beds around the Drowned Cayes. Ecosystem engineers are species that either morphologically or behaviorally create more complex habitats (Coleman and Williams, 2002). Removal of these species might affect changes in seagrass composition and might have other unknown consequences. 74 This study, while by no means a comprehensive review of resting holes and resting behavior, hints at the possibility of many other differences, both behavioral and ecological, between local populations of subspecies of the West Indian manatee. Populations have to deal with different environmental factors in different locations, such as large tide changes, more extreme dry and wet seasons and accompanying changes in water levels, and variations in abundance of fresh water sources. These differences in behavior and habitat use, and the accompanying variations in anthropogenic factors, must be acknowledged and examined by conservation managers as they develop local conservation strategies for manatees and other endangered species. Site-specific studies allow researchers to detect the diversity in behavior among populations within the West Indian species and within each sub-species. Site-specific research is better suited to highlighting the unique local socio-economic factors which will affect the institution of any recommendations from that research. While the field of manatee research has benefited greatly from the well-studied Florida manatee populations, it appears naïve to call for world-wide implementation of manatee habitat conservation protocols which rely solely on studies of Florida manatees. Furthermore, the socio-economic milieu of the United States is vastly different from the financial and social factors which affect the people of Central and South American countries, and these unique differences greatly impact the amount of money and importance applied to conservation efforts around every local population of West Indian manatees. Therefore conservation issues, and strategies for solving them, must become local in nature, based on differences in both human and manatee behaviors. Since Belize, and the Drowned Cayes in particular, are such an important area for 75 the Antillean manatee (Morales-Vela et al., 2000), preservation of this key habitat should be a priority. Increase in cruise ship tourism will continue to be a problem in this area as many of these tourists travel through the Drowned Cayes to reach the barrier reef. For the safety of the manatees and the tourists, boat speed for these excursions should be restricted and closely regulated. In addition, some of the more tranquil lagoons and coves should be deemed off-limits to tourist boats, to provide the manatees with quiet places where they can rest undisturbed. Development of the mangrove islands should also be closely monitored to assure that minimal damage is caused to the environment, especially some of the important seagrass beds near these crucial resting holes. Conclusions Manatees in the Drowned Cayes of Belize are known to rest primarily in depressions in the benthic substrate called resting holes, which are usually found in quiet coves, lagoons, and dead-end bogues where there is minimal disturbance from humans, wind, waves and currents. Since there are a large number of resting holes in the Drowned Cayes, and many of these seem to be in sheltered areas with low currents, it is quite possible that the location of these holes are chosen by manatees precisely because of the tranquility of the area. Thus, when resting either in the water column or on the substrate, a manatee could avoid the loss of energy required to maintain its position against a current. Most of these areas are also protected from wave actions and are tucked in small deadend bogues where disturbance from boats is minimal. Also, extensive seagrass beds tend to be located nearby. The combination of low currents, protection from waves, low numbers of passing boats, and nearby grass beds would make these areas ideal to rest in. 76 More manatees were sighted during scans in 2005 than in 2006, although factors that attribute to this are not yet understood. Also, more manatees were sighted during the day than at night for both years. Despite all the challenges posed by night scans, the probability is high that if manatees were using the resting holes as often at night as during the day, several manatees would have been detected over the course of both summers. Since this was not the case, this may indicate that manatees use resting holes primarily during the day and may go elsewhere at night. Manatees may migrate to seagrass beds to feed or may travel to the Belize River or other coastal rivers to obtain freshwater. Since several resting holes are used regularly by a number of animals in the Drowned Cayes, they seem to be an important part of the local manatee habitat. In light of these findings, primary conservation goals should include preservation of areas with resting holes including restriction of boat speed and access. Suggestions for Further Study As more human development occurs in the Drowned Cayes, it is becoming vital to detail the impact of construction efforts on manatee habitats. In particular, observations need to focus on the effects of disturbance of the substrate and water quality around resting holes. For example, a trial could be conducted in which frequency of resting hole use is recorded both before and after another hole is artificially created nearby an established resting hole location. This may reveal that manatees begin to use this new hole or that their use of the original hole changes after the disturbance, which would provide a good idea of whether construction efforts which change the underwater terrain impact manatee behavior. 77 While freshwater sources are of importance to manatees, the reasons for this are not well understood. Since manatees in the Drowned Cayes are living in a marine environment, they would also need to obtain freshwater from some source to survive. If sufficient freshwater cannot be found in the Drowned Cayes, manatees may be traveling to the Belize River to obtain freshwater, or to other nearby coastal rivers, such as the Sibun River. More study in trying to understand how often manatees need freshwater, where they are obtaining this water from in a marine environment, and whether or not they are returning to coastal rivers during the night would further our understanding of freshwater needs and usage. One possible way to test if manatees frequent coastal rivers to obtain freshwater would be to set up monitoring areas around the mouths of local rivers, from which day and night observations could be made, involving spotlights or other night instruments. Another method, which would require more resources for appropriate technology and maintenance, would be an acoustic tagging system which would activate acoustic gates located at the mouth of these rivers, thereby detailing the frequency and timing with which uniquely identified manatees enter and exit the rivers. Alternatively, tagging manatees with radio tags could theoretically provide detailed “route maps” of pathways used to access these rivers. These tagging methods would help to determine whether particular manatees use preferred rivers, or if all manatees return to one or two rivers that may be more pristine and have less disturbance from humans and boats. In addition, manatees may seek out large bodies of freshwater intentionally to rid themselves of ectoparasites. A study encompassing research into the change in numbers of ectoparasites on manatees after freshwater usage might help elucidate this. The current study hints that manatees in the Drowned Cayes rest in shorter periods 78 of about 30 minutes, compared to Florida manatees which rest in periods of two to four hours (Hartman, 1979). In order to clarify whether manatees may be going to different holes to rest in short periods, or if they really do rest for shorter periods before transitioning to different behavioral states, two general methods could be utilized. One approach would be to perform longer observations of resting manatees, conducting more eight-hour day observations at different resting holes. Another method would be to perform focal follows, either under visual sighting from a boat, or by tagging them with radio transmitters which would allow a boat to follow with minimal disturbance. Finally, being able to track individual manatees around the clock would provide very useful information, especially about their nocturnal activities. Reliable radio or GPS tracking systems could help identify whether manatees regularly use resting holes at night, and if not, where else they might be going at night. Manatee-specific tracking systems could best detail whether an individual returns to the same resting hole on a regular basis or if it indiscriminately uses all of the resting holes, and also whether there is any natal philopatry to resting holes. Alternatively, long-term studies using underwater video to identify uniquely marked animals could help determine site fidelity patterns, if they exist. Boat-mounted side-scan sonar could also be used to monitor for nocturnal behaviors, by performing nocturnal transects around large seagrass beds. Most recently, a few manatees have been tagged in the Drowned Cayes, but signals from these animals tend to be quickly lost (Auil, pers. comm.). It is not known if the tags break down or detach from the animal and are unable to be found again for various reasons. Therefore, tagging of individual manatees in this area would first require improved technology and methods than what has been used to date. 79 REFERENCES Aipanjiguly, S., Jacobson, S.K., Flamm, R., 2003. Conserving manatees: knowledge, attitudes, and intentions of boaters in Tampa Bay, Florida. Conservation Biology 17:1098-1105. Ames, A.M., Van Vleet, E.S., 1996. Organochlorine residues in the Florida manatee. Trichechus manatus latirostris. Marine Pollution Bulletin 32:374-377. Anonymous, 2005. Humminbird 987c SI Combo Installation and Operations Manual. Humminbird, Eufaula, AL, 88 pp. Axis-Arroyo, J., Morales-Vela, B., Torruco-Gomez, D., Vega-Cendejas, M.E., 1998. Variables asociadas con el use de habitat del manatí del Caribe (Trichechus manatus), en Quintana Roo, México (Mammalia). Revista de Biologia Tropical 46:791-803. Bayliss, P., Freeland, W.J., 1989. Seasonal distribution and abundance of dugongs in the western Gulf of Carpentaria. Australian Wildlife Research 16:141-149. Bengtson, J.L., 1981. Ecology of manatees (Trichechus manatus) in the St. Johns river, Florida. Ph.D. Thesis, University of Minnesota, Minneapolis, 126 pp. Bengtson, J.L., 1983. Estimating food consumption of free-ranging manatees in Florida. Journal of Wildlife Management 47:1186-1192. Bengtson, J.L., Fitzgerald, S.M., 1985. Potential role of vocalizations in West Indian manatees. Journal of Mammalogy 66:816-819. Bengtson, J.L., Magor, D., 1979. A survey of manatees in Belize. Journal of Mammalogy 60:230-232. Best, R.C., 1981. Foods and feeding habits of wild and captive Sirenia. Mammal Review 11:3-29. Bonde, R.K., Aguirre, A.A., Powell, J.A., 2004. Manatees as sentinels of marine ecosystem health: are they the 2000-pound canaries? EcoHealth 1:255-262. Bossart, G.D., Baden, D.G., Ewing, R.Y., Roberts, B., Wight, S.D., 1998. Brevetoxicosis in manatees (Trichechus manatus latirostris) from the 1996 epizootic: gross, histologic and immunohistochemical features. Toxicologic Pathology 26:276-282. Buckingham, C.A., Lefebvre, L.W., Shaerfer, J.M., Kochman, H.I., 1999. Manatee response to boating activity in thermal refuge. Wildlife Society Bulletin 27:514-522. 80 Coleman, C., Williams, S.L., 2002. Overexploiting marine ecosystem engineers: potential consequences for biodiversity. Trends in Ecology and Evolution 17:40-44. Deutsch, C.J., Reid, J.P., Bonde, R.K., Easton, D.E., Kochman, H.I., O’Shea, T.J., 2003. Seasonal movements, migratory behavior, and site fidelity of West Indian manatees along the Atlantic coast of the United States. Wildlife Monographs 151:1-77. Deutsch, C.J., Self-Sullivan, C., Mignucci-Giannoni, A. Trichechus manatus. In: IUCN 2007. 2007 IUCN Red List of Threatened Species. <www.iucnredlist.org>. Domning, D.P., Hayek, L.C., 1986. Interspecific and intraspecific morphological variation in manatees (Sirenia: Trichechus). Marine Mammal Science 2:87-144. Etheridge, K., Rathbun, G.B., Powell, J.A., Kochman, H.I., 1985. Consumption of aquatic plants by the West Indian manatee. Journal of Aquatic Plant Management 23:21-25. Frisch, K., Haubold, E., 2003. Florida manatee avoidance technology: a pilot program by the Florida Fish and Wildlife Conservation Commission. The Journal of the Acoustical Society of America 144:2415. Ford, R.E., 1991. Toponymic generics, environment, and culture history in preindependence Belize. Journal of the American Name Society 39:1-26. Garcia-Rodriguez, A.I., Bowen, B.W., Doming, D.P., Mignucci-Giannoni, A.A., Marmontel, M., Montoya-Ospina, R.A., Morales-Vela, B., Rudin, M., Bonde, R.K., McGuire, P.M., 1998. Phylogeography of the West Indian manatee (Trichechus manatus): how many populations and how many taxa? Molecular Ecology 7:11371149. Gonzalez-Socoloske, D., 2007. Status and distribution of manatees in Honduras and the use of side-scan sonar. M.S. Thesis, Loma Linda University, Loma Linda, CA, 91 pp. Google Earth™ mapping service, 2006. http://earth.google.com/. Hartman, D.S., 1971. Behavior and ecology of the Florida manatee, Trichechus manatus latirostris (Harlan), at Crystal River, Citrus County. Ph.D. Thesis, Cornell Univsersity, Ithaca, NY, 285 pp. Hartman, D.S., 1979. Ecology and behavior of the manatee (Trichechus manatus) in Florida. American Society of Mammalogists, Special Publication 5. Huler, S., 2004. Defining the Wind: the Beaufort Scale, and How a 19th-Century Admiral Turned Science in to Poetry. Crown Publishers, New York, 304 pp. 81 Hurst, L.A., 1987. Review of the status and distribution of the West Indian manatee (Trichechus manatus) in Jamaica, with an evaluation of the aquatic vegetation of the Alligator Hole river. M.S. Thesis, University of Florida, Gainesville, 170 pp. Husar, S.L., 1978. Dugong dugon. Mammalian Species 88:1-7. Irvine, A.B., 1983. Manatee metabolism and its influence on distribution in Florida. Biological Conservation 25:315-334. Jiménez, I., 1999. Estado de conservación, ecología y conocimiento popular del manatí (Trichechus manatus, L.) en Costa Rica. Vida Silvestre Neotropical 8:18-30. Jiménez, I., 2002. Heavy poaching in prime habitat: the conservation status of the West Indian manatee in Nicaragua. Oryx 36:272-278. Jiménez, I., 2005. Development of predictive models to explain the distribution of the West Indian manatee Trichechus manatus in tropical watercourses. Biological Conservation 125:491-503. Kinnaird, M.F., 1985. Aerial census of manatees in northeastern Florida. Biological Conservation 32:59-79. Lefebvre, L.W., Marmontel, M., Reid, J.P., Rathbun, G.B., Domning, D.P., 2001. Status and biogeography of the West Indian manatee. In: Wood, C.A., Sergile, F.E. (Eds.), Biogeography of the West Indies: Patterns and Perspectives, 2nd ed. CRC Press, Boca Raton, FL, pp. 425-474. Lefebvre, L.W., Reid, J.P., Kenworthy, W.J. Powell. J.A., 2000. Characterizing manatee habitat use and seagrass grazing in Florida and Puerto Rico: implications for conservation and management. Pacific Conservation Biology 5:289-298. Miksis-Olds, J.L., 2006. Manatee response to environmental noise. Ph.D. Thesis, University of Rhode Island, Narragansett, 228 pp. Morales-Vela, B., Olivera-Gomez, L.D., Reynolds, J.E., Rathbun, G.B., 2000. Distribution and habitat use by manatees (Trichechus manatus manatus) in Belize and Chetumal Bay, Mexico. Biological Conservation 95:67-75. Mukhametov, L.M., Lyamin, I.S., Chetyrbok, A.A., Vassilyev, A.A., Diaz, R.P., 1992. Sleep in an Amazonian manatee, Trichechus inunguis. Experientia 48:417-419. Nowacek, S.M., Wells, R.S., Owen, C.G., Speakman, T.R., Flamm, R.O., Nowacek, D.P., 2004. Florida manatees, Trichechus manatus latirostris, respond to approaching vessels. Biological Conservation 119:517-523. 82 Olivera-Gomez, L.D., Mellink, E., 2005. Distribution of the Antillean manatee (Trichechus manatus manatus) as a function of habitat characteristics in Bahia de Chetumal, Mexico. Biological Conservation 121:127-133. O’Shea, T.J., Correaviana, M., Ludlow, M.E., Robinson, J.G., 1988. Distribution, status, and traditional significance of the West Indian manatee Trichechus manatus in Venezuela. Biological Conservation 46:281-301. O'Shea, T.J., Poché, L.B., 2006. Aspects of underwater sound communication in Florida manatees (Trichechus manatus latirostris). Journal of Mammalogy 87:1061-1071. O’Shea, T.J., Rathbun, G.B., Bonde, R.K., Buergelt, C.D., Odell, D.K., 1991. An epizootic of Florida manatees associated with a dinoflagellate bloom. Marine Mammal Science 7:165-179. O’Shea, T.J., Salisbury, C.A., 1991. Belize - a last stronghold for manatees in the Caribbean. Oryx 25:156-164. Ortiz, R.M., Worthy, G.A.J., Byers, F.M., 1999. Estimation of water turnover rates of captive West Indian manatees (Trichechus manatus) held in fresh and salt water. Journal of Experimental Biology 202:33-38. Ortiz, R.M., Worthy, G.A.J., Mackenzie, D.S., 1998. Osmoregulation in wild and captive West Indian manatees (Trichechus manatus). Physiological Zoology 71:449-457. Pironet, F.N., Jones, J.B., 2000. Treatments for ectoparasites and diseases in captive Western Australian dhufish. Aquaculture International 8:349-361. Powell, J.A., Belitsky, D.W., Rathbun, G.B., 1981. Status of the West Indian manatee (Trichechus manatus) in Puerto Rico. Journal of Mammalogy 62:642-646. Powell, J.A., Rathbun, G.B., 1984. Distribution and abundance of manatees along the northern coast of the gulf of Mexico. Northeast Gulf Science 7:1-28. Preen, A., 1995. Impacts of dugong foraging on seagrass habitats: observational and experimental evidence for cultivation grazing. Marine Ecology Progress Series 124:201-213. Provancha, J.A., Hall, C.R., 1991. Observations of associations between seagrass beds and manatees in east central Florida. Florida Scientist 54:87-98. Ralph, C.L., Young, S., Gettinger, R., O’Shea, T.J., 1985. Does the manatee have a pineal body? Acta Zoologica 66:55-60. Rathbun, G.B., Powell, J.A., Cruz, G., 1983. Status of the West Indian manatee in 83 Honduras. Biological Conservation 26:301-308. Rathbun, G.B., Reid, J.P., Carowan, G., 1990. Distribution and movement patterns of manatees (Trichechus manatus) in Northwestern peninsular Florida. Florida Marine Research Publications 48:1-33. Reynolds, J.E., 1981. Behavior patterns in the West Indian manatee, with emphasis on feeding and diving. Biological Sciences 44:233-241. Reynolds, J.E., 1999. Efforts to conserve the manatee. In: Twiss, J.R., Reeves, R.R. (Eds.), Conservation and Management of Marine Mammals. Smithsonian Press, Washington D.C., pp. 267-295. Reynolds, J.E., Szelistowski, W.A., Leon, M.A., 1995. Status and conservation of manatees Trichechus manatus manatus in Costa Rica. Biological Conservation 71:193-196. Self-Sullivan, C., 2007. Conservation of Antillean manatees in the Drowned Cayes of Belize. Ph.D. Thesis, Department of Wildlife & Fisheries Sciences, Texas A&M University, College Station, 149 pp. Self-Sullivan, C., Smith, G.W., Packard, J.M., LaCommare, K.S., 2003. Seasonal occurrence of male Antillean manatees (Trichechus manatus manatus) on the Belize Barrier Reef. Aquatic Mammals 29:342-354. Simberloff, D., 1998. Flagships, umbrellas, and keystones: is single-species management passé in the landscape era? Biological Conservation 83:247-257. Smethurst, D., Nietschmann, B., 1999. The distribution of manatees (Trichechus manatus) in the coastal waterways of Tortuguero, Costa Rica. Biological Conservation 89:267-274. Vianna, J.A., Bonde, R.K., Caballero, S., Giraldo, J.P., Lima, R.P., Clark, A., Marmontel, M., Morales-Vela, B., de Souza, M.J., Parr, L., Rodrigues-Lopez, M.A., MignucciGiannoni, A.A., Powell, J.A., Santos, F.R., 2006. Phylogeography, phylogeny and hybridization in trichechid sirenians: implications for manatee conservation. Molecular Ecology 15: 433-447. 84