xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200397Ka 4500
controlfield tag 001 002029501
007 cr mnu|||uuuuu
008 090917s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002920
Powell, Jessica R.
Depredation and angler interactions involving bottlenose dolphins (Tursiops truncatus) in Sarasota Bay, Florida
h [electronic resource] /
by Jessica R. Powell.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 129 pages.
Thesis (M.S.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
ABSTRACT: Typical depredation behavior by cetaceans involves stealing or damaging prey items already captured by recreational or commercial fishing gear. Depredation among cetaceans has been reported to be increasing in both severity and frequency globally. This behavior is of particular concern for small stocks of cetaceans since any interaction with fishing gear has the potential to injure or kill animals leading to unsustainable losses. In Florida, depredation became evident in 2006 when the number of bottlenose dolphin (Tursiops truncatus) strandings resulting from fishing gear ingestion or entanglement sharply increased. For the resident dolphin community in Sarasota Bay, modeling showed continued mortalities from recreational fishing gear interactions were not sustainable.The major goals of this study were to 1.) characterize depredation and recreational angler interactions involving dolphins in Sarasota Bay, 2.) reduce dolphin-angler interactions through outreach, 3.) examine a case study to investigate the link between dolphin hearing loss and angler interaction behavior, 4.) test the effectiveness of passive acoustics in monitoring dolphin depredation at a fishing pier. Findings from this study provided a better understanding of depredation and angler interactions. Results indicated that dolphin-angler interactions in Sarasota Bay are increasing in frequency and are affecting an increasing number of dolphins, specifically adult males. Some dolphins in Sarasota Bay appear to utilize depredation as a foraging method (not just an opportunistic behavior) and were significantly more likely to be within 50 m of an active fishing line.Depredation and angler interaction behavior appear to increase in times of prey depletion (such as during a red tide) and heightened angler fishing activity. Educational outreach using an informational card proved successful in a case study showing about a 30% reduction in dolphin provisioning rates. The case study of F201 offers preliminary evidence that hearing loss is linked to depredation behavior and death for wild dolphins. Also, by detecting echolocation clicks as a proxy for dolphin presence, passive acoustics showed potential as an inexpensive method for monitoring depredation in problematic areas. Conclusions from this study can be utilized by scientists and managers when assessing depredation rates for a cetacean community and implementing an action plan.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Co-advisor: David A. Mann, Ph.D.
Co-advisor: Randall S. Wells, Ph.D.
Auditory evoked potentials (AEP)
x Marine Science
t USF Electronic Theses and Dissertations.
Depredation and Angler Interacti ons involving Bottlenose Dolphins ( Tursiops truncatus ) in Sarasota Bay, Florida by Jessica R. Powell A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Co-Major Professor: David A. Mann, Ph.D. Co-Major Professor: Randall S. Wells, Ph.D. Andrew J. Read, Ph.D. Date of Approval: March 30, 2009 Keywords: behavior, fishing, passive acous tics, auditory evoked potentials (AEP), outreach, habitat, activity budgets, social, red tide Copyright 2009, Jessica R. Powell
Dedication This thesis is dedicated to my family, Joel Sarah, Kristin, and Amanda Powell. For my father, who never accepts anything less than my best and always pushes me to not just follow my dreams, but catch them as well. For my mother, whose caring, supportive words always carry me through. For Kristin, whose strength and ambition never cease to inspire me. And for Amanda, whose spirit I admire most. Thank you.
Note to Reader The original of the document contains color that is necessary for understanding the data. The original thesis is on file with the USF library in Tampa, Florida.
i Table of Contents List of Tables iii List of Figures iv Abstract viii Preface x 1. Introduction 1 1.1 Goal of this Thesis 5 2. Behavior: Characterization of Dolphins engagi ng in Angler Interactions 7 2.1 Introduction 7 2.1.1 Anthropogenic Factors of Depredation a nd Implications 8 2.1.2 Environmental Factors of Depredation and Implications 9 2.1.3 Behavioral Factors of De predation and Implications 10 2.2 Methods 12 2.2.1 Study Area and Population 12 2.2.2 Behaviors of Interest 13 2.2.3 Distance Estimation 14 2.2.4 Pier Observations 14 2.2.5 Focal Dolphin Selection and Focal Anim al Behavioral Follows 17 2.2.6 Methods of Analyses 18 2.2.7 Education Outreach: Card Design and Distribution 25 2.3 Results 27 2.3.1 Depredation History and Dem ographics 27 2.3.2 Depredation related to Red Tide ( K. brevis ) Blooms and Boating Activity 28 2.3.3 Habitat Selection 29 2.3.4 Proximity to Fishing Boats and Lines 29 2.3.5 Home Range 29 2.3.6 Pier 29 2.3.7 Activity Budgets 30 2.3.8 Depredation 31 2.3.9 Social Behavior 32 2.3.10 Case Studies: Beggar and F106 33 2.4 Discussion 39
ii 3. F201 Case Study: Exploring the Pote ntial Link between hearing impairment and Depredation 88 3.1 Introduction 88 3.2 Methods 92 3.2.1 Auditory Evoked Potentials (AEP) 92 3.2.2 Post-Release Monitoring 93 3.3 Results 94 3.4 Discussion 95 4. Depredation Monitoring using Passive Acoustics 102 4.1 Introduction 102 4.2 Methods 104 4.2.1 Hydrophone Deployment and Setup 104 4.2.2 DSGLab Automatic Echolocation Detection 105 4.2.3 Visual Survey vs. Passive Acoustic Recordings 106 4.3 Results 107 4.3.1 DSGLab Automatic Echolocation Detection 107 4.3.2 Visual Survey vs. Passive Acoustic Recordings 108 4.4 Discussion 109 5. Literature Cited 119
iii List of Tables Table 2.1 Working definitions of activ ity categories as defined for the Sarasota Bay dolphin community. 49 Table 2.2 Depredation and related worki ng definitions create d to quantify dolphin behavior when interacting with anglers, fishing vessels, or fishing piers. 50 Table 2.3 Location information on the six piers and jetties w ithin the home range of the Sarasota dolphin co mmunity. 51 Table 2.4 Depredator a nd control focal dolphins se lected for behavioral follows in summers of 2007 and 2008. 52 Table 2.5 Dolphin-angler interaction rates (including all behaviors such as patrol, beg, scavenge, attempted depredation, line depredation, and provision) in Sarasota Bay, Florida for 2000-2007. 53 Table 2.6 Depredator dol phins in Sarasota Bay a nd percent of total population from 2000-2007 calcu lated for monthly photoidentification surveys and all sightin g efforts. 54 Table 2.7 Age class and sex distribu tion of documented confirmed and possible depredating dolphins in Sara sota Bay (2000-2007). 55 Table 4.1 Summary of vi sual dolphin sightings a nd corresponding acoustic recordings over six days at Anna Maria City Pier. 113
iv List of Figures Figure 2.1 Area surveyed monthly for Sarasota dolphin community members. 56 Figure 2.2 Mean error for estimating distance (n=2389) for the primary investigator (J.R. Powell). 57 Figure 2.3 Aerial view of Anna Maria City Pier fr om 2008 Google satellite maps. 58 Figure 2.4 The 95% kernel home range a nd 50% kernel core area for focal animals from January 2004 through August 2008. 59 Figure 2.5 The first edition of the Dolphin-Friendly Fishing & Viewing Tips educational card. 62 Figure 2.6 The second edition of the Dolphin-Friendly Fishing & Viewing Tips educational card. 63 Figure 2.7 Dolphin-Friendly Fishing and Viewing Tips card distribution between January and October 2008. 64 Figure 2.8 Dolphin-angler interaction rates for th e Sarasota Bay dolphin community from 2000-2007 based on monthly photo-identification survey data. 65 Figure 2.9 Dolphin-angler interaction rates for th e Sarasota Bay dolphin community from 2000-2007 based on all sighting effort. 66 Figure 2.10 Data collected from monthl y photo-identification surveys showing the percent of animals in the Sa rasota Bay population engaging in angler interaction behavior fr om 2000-2007. 67 Figure 2.11 Data collected from all sighting e ffort showing the percent of animals in the Sarasota Bay popul ation engaging in angler interaction behavior from 2000-2007. 68
v Figure 2.12 Data collected from monthly photoidentification surveys showing the cumulative percentage of dolphins in the Sarasota Bay population from 2000 through the specified year that have been observed engaging in angler interac tion behavior. 69 Figure 2.13 Data collected from all sighting effort showing the cumulative percentage of dolphins in the Sarasota Bay population from 2000 through the specified year that have been observed engaging in angler interaction behavior. 70 Figure 2.14 Monthly angler interaction rates for the Sarasota Bay dolphin community from 2000-2007 based on monthly photo-identification surveys (Avg surveys) and all sighting effo rt (Avg all effort). 71 Figure 2.15 Comparison of dol phin-angler interaction ra tes (based on monthly photo-identification surveys and al l sighting effort) and average monthly Sarasota County hotel/m otel occupancy rates from 20002007. 72 Figure 2.16 Comparison of angler intera ction rates (based on monthly photoidentification surveys and all sighting effort) and boat fuel sales from Cannons Marina 2003-2007. 73 Figure 2.17 Average of dolphin-angler in teraction rates in Sarasota Bay, Florida over three environmental periods based on Florida red tides (non-bloom, K. brevis bloom, and lag time following a bloom) from 2000-2007. 74 Figure 2.18 Habitat use by focal dolphins during behavioral focal animal follows from summer 2007 and 2008. 75 Figure 2.19 The mean presence of dolphins and the mean number of fishing lines (rigged with either bait or lures) by month during pier surveys from May-July 2007, and October 2007-April 2008. 76 Figure 2.20 The mean presence of dolphins and mean number of fishing lines (rigged with either bait or lures) by hour during pier surveys from May-July 2007, and October 2007-April 2008. 77 Figure 2.21 Overall mean activity budgets for control (n=8) and depredator (n=8) focal dolphins. Data compiled from summer 2007 and 2008 focal animal behavioral follows. 78 Figure 2.22 Overall mean activity budgets for control (n=8) and depredator (n=8) focal dolphins. 79
vi Figure 2.23 Individual mean activity budgets for control (left) and depredator (right) focal dolphins. 80 Figure 2.24 Comparison of 58 depredation events documented during pier surveys and behavioral focal animal follows (2007-2008). 83 Figure 2.25 Comparison of 70 total depr edation events documented during pier surveys, behavioral focal animal follows, monthly population and photo-identification surveys, and various other research initiatives (2007-2008). 84 Figure 2.26 Angler and boater reactions following a dolphin depredation event documented over 58 incidents during pier surveys and behavioral focal animal follows (2007-2008). 85 Figure 2.27 Overall depredator focal dolphin group size compared to mean group size when focal was depredatin g or engaging in depredation related behavior. 86 Figure 2.28 Sequence of photos (in order from top to bottom) illustrating the order of behaviors that BEGR engages in when begging from boaters. 87 Figure 3.1 F201 photographed by the Sarasota Dolphin Research Program on January 19, 2007. 99 Figure 3.2 Comparison of F201 auditory evoked potential audiogram to the mean ( SD) audiograms measured for 32 male and 29 female freeranging bottlenose dolphins in Sarasota Bay, Florida (Cook 2006). 100 Figure 3.3 Spectrogram of F201 echolocation click train followed by a whistle recorded on April 17, 2007 (post-release). 101 Figure 4.1 The percentage of 10 s segm ents containing echolocation click trains per month at Anna Maria C ity Pier compared to the monthly depredation rate (determined by pi er surveys) for Sarasota Bay, FL (R2=0.65, p =0.053). 114 Figure 4.2 The percentage of 10 s segments containing echolocation click trains per hour at Anna Maria C ity Pier compared to the hourly depredation rate (determined by pi er surveys) for Sarasota Bay, FL (R2=0.53, p =0.027). 115
vii Figure 4.3 The percentage of 10 s segm ents containing echolocation click trains compared to the mean nu mber of dolphins sighted per hour (determined by pier surveys) at Anna Maria City Pier. 116 Figure 4.4 The percentage of 10 s segments with echolocation click trains compared to the mean number of total fishing lines per hour (determined by pier surveys) at Anna Maria City Pier. 117 Figure 4.4 Comparison of F106 signature whistles from the Sarasota Dolphin Whistle Catalog (top) (courtesy of L. Sayigh and V. Janik) and from pier recordings (bottom). 118
viii Depredation and Angler Interactions involving Bottlenose Dolphins ( Tursiops truncatus ) in Sarasota Bay, Florida Jessica R. Powell ABSTRACT Typical depredation behavior by cetaceans involves stealing or damaging prey items already captured by recreational or co mmercial fishing gear. Depredation among cetaceans has been reported to be increasing in both severity a nd frequency globally. This behavior is of particular concern for small stocks of cetacean s since any interaction with fishing gear has the potential to injure or kill animals leading to unsustainable losses. In Florida, depredation became evident in 2006 when the number of bottlenose dolphin ( Tursiops truncatus ) strandings resulting from fishing gear ingestion or entanglement sharply increased. For the re sident dolphin community in Sarasota Bay, modeling showed continued mortalities from r ecreational fishing gear interactions were not sustainable. The major goals of this study were to 1.) characterize depredation and recreational angler interactions involvi ng dolphins in Sarasota Ba y, 2.) reduce dolphin-angler interactions through outreach, 3.) examine a case study to investigate the link between dolphin hearing loss and angler in teraction behavior, 4.) test the effectiveness of passive acoustics in monitoring dolphin depr edation at a fishing pier.
ix Findings from this study provided a better understanding of depredation and angler interactions. Results indicated that do lphin-angler interactions in Sarasota Bay are increasing in frequency and are affecting an increasing numbe r of dolphins, specifically adult males. Some dolphins in Sarasota Bay appear to utilize depr edation as a foraging method (not just an opportunist ic behavior) and were significantly more likely to be within 50 m of an active fishing line. Depredation and angler interaction behavior appear to increase in times of prey depletion (such as during a red tide) a nd heightened angler fishing activity. Educational outreach using an informational card proved successful in a case study showing about a 30% reduction in dolphin provisioning rates. The case study of F201 offers preliminary ev idence that hearing loss is lin ked to depredation behavior and death for wild dolphins. Also, by de tecting echolocation clicks as a proxy for dolphin presence, passive acous tics showed potential as an inexpensive method for monitoring depredation in problem atic areas. Conclusions from this study can be utilized by scientists and managers when assessing depredation rates for a cetacean community and implementing an action plan.
x Preface I thank my advisory committee for allo wing me the opportunities and freedom to explore so many ideas in my pursuit to understand depredation and dolphin-angler interactions. I thank David Mann for his guida nce and support as well as for constantly challenging me to problem-solve a nd think critically about ideas. I thank Randy Wells for his unwavering encouragement, advice, a nd continually working with me to improve my skills as a scientist. I am grateful to Andy Read for first sparking my passion for marine mammal conservation and research a nd for all his insightfu l feedback on project design and analyses. This work was made possible by the supe rb efforts of many field assistants and volunteers. I especially thank Kim Atwater, Robin Bisel, Kristen Burtch, Maurcio Cantor, Catherine Deveau, Rachel Eubank, Kerry Foltz, Ellie Glasser, Rebeccah Hazelkorn, and Carolyn Kovacs. I am grateful to the members of the Sarasota Dolphin Research Program for their assistance in teaching me boating skills, field photoidentification, and helping me w ith the details of data protoc ols. I thank both members of the Sarasota Dolphin Research Program and Mann lab for helpful discussions and support related to project design, an alyses, and card distribution. I thank Mandy Cook for taking the time to teach me AEP field techniques and theory. I am grateful to Jim Locascio for his assistance with hydrophone installation. I thank Janet Gannon for her help with GIS analyses. I thank Kim Bassos-Hull for her time spent reviewing thesis drafts and her
xi helpful comments. A special thanks to Robi n Perrtree with her assistance with card distribution. I thank Gary Kirkpatrick and the Phytoplankton Ecology Lab at Mote Marine Laboratory for providing me with the cell count s necessary for red tide analyses. Also, I thank Cannons Marina for their use of boat fuel sales data and Virginia Haley for providing me with the hotel occupancy data for Sarasota County. I appreciate the cooperation of Anna Maria City Pier Staff, especially manager Dave Sork, in their support with pier surveys and hydrophone installation. I am also grateful to Martha Wells for her time spent coaching me on public-relations and interface. This work was supported by two grants from the Disney Wildlife Conservation Fund to R.S. Wells and J.R. Powell. This re search was also funded by the University of South Florida, College of Marine Science graduate assistantship a nd a scholarship from Fish Florida.
1 1. Introduction Depredation by a predator is the act of stealing or da maging a prey item already captured by some other process (Zollett and Read 2006). Depredation of commercial and recreational fishing gear by cetaceans is a growing problem around the world and has been documented in areas such as Australia the Pacific Ocean, southern Brazil, the Mediterranean Sea, and the eastern Unite d States (Broadhurst 1998, Secchi and Vaske 1998, Donoghue et al 2002, Noke and Ode ll 2002, SPREP 2002, Cox et al. 2003, Lauriano et al. 2004, Brotons et al. 2008, and Sigler et al 2008). Long-line fisheries depredation by larger odontocetes has recen tly been recognized as increasing in frequency, geographic extent, and severity (R ead 2008). Removal of, or damage to, bait or catch by cetaceans create an economic loss, degrade a recreational experience, and increase the chance of retaliation by the angler (Read 2008). In addition, this behavior increases the animals risk of ingesting or becoming enta ngled in fishing gear which could then result in injury or death (Gorzelany 1998; Wells et al. 1998, 2008). Any activities that bring dolphins into cont act with fishing gear have the potential to seriously injure or kill the animals thr ough entanglement or ingestion (Wells and Scott 1994, Wells et al. 2008). In Sarasota Bay, Florida, ingestion of gear usually leads to mortality through suffocation or starvation as wrapped monofilament line constricts, obstructs, or damages the goosebeak or the esophagus (Wells et al. 2008). Severe constrictive monofilament entanglements (unless wrapped around distal ends of flippers,
2 flukes, and dorsal fins) are also considered to be fatal without human intervention (Wells et al. 2008). For long-term management purposes, it is important to understand the nature of the depredation behavior, such as why animal s engage in depredation and if they are targeting specific bait or catch. Depredation has been shown to cause changes to ranging patterns, habitat selection, and natural activity patterns (Chilvers and Corkeron 2001, Reeves et al. 2001, Finn et al. 2008). It can also lead to a decrease in natural foraging predation and a change in prey species (SPREP 2002, Zollett and Read 2006). These changes may disproportionately affect certai n sex and age classes. For example, in Western Australia, illegal feeding of dolphins disproportionately involved adult males and subadults (Finn et al. 2008). It is also imperative to examine the frequency of depredation, the proportion of the population that engages in the behavior, and whether or not these frequencies increase with time. In the state of Florida, there has be en a recent increase in bottlenose dolphin ( Tursiops truncatus ) entanglement and ingestion of fishing gear (NOAA 2006 a). In 2005, five dolphins were recovere d that had ingested or be come entangled in fishing gear. In 2006, thirteen animals that had inge sted or were entangled in gear stranded and there were many reports of non-stranded, entangled dolphins from around the state (NOAA 2006 a). This trend may be a result of incr eased depredation rates. Therefore in response to the greater number of deaths, NOAA, working in conjunction with Mote Marine Laboratory, Chicago Zoological Societ y, Hubbs SeaWorld Research Institute, and anglers and fishing guides, developed and disseminated a list of Best Fishing Practices for Avoiding Interactions with Wild Dolphins (NOAA 2006 a).
3 One area hit particularly hard by fi shing-related mortalities during 2006 was southwest Florida. In 2006, the Mote Marine Laboratory Stranding Investigations Program retrieved 5 dolphins, four adults and one calf, whose deaths were related to interactions with fishing gear. The fluke of the calf was nearly severed by monofilament line. The four adults were all long-term re sidents of the Sarasota Bay dolphin community that has been studied since 1970 (Wells 1991, 2003) Of the three adults believed to have died from ingestion of fishing gear, only one animal had a previous history of angler interactions (Wells et al. 2006). The other adult died with a large fishing lure caught in her mouth, but a stingray barb was considered to be the primary cause of death. In comparison, Mote recovered only one entang led dolphin in 2005 and death was from other causes. Of the total Sarasota Bay r ecovered strandings in 2006, approximately 25% were a result of fishing gear interaction, compared to th e average 2.9% rate for dolphin deaths attributed to fishing gear for the years of 2000-2005 (NOAA 2006 a). The deaths in 2006 resulted in a loss of more than 2% of the resident Sarasota dolphin community (Wells et al. 2006). Continued losses at th is rate, in addition to those from previously-existing mortality sources, we re determined to be unsustainable for the Sarasota Bay dolphin community, through pr eliminary population modeling using the program Vortex (Wells et al. 2006). In terms of conserva tion management, small stocks of cetaceans, like the Sarasota Bay bottlenos e dolphin population of about 160 dolphins, are especially susceptible to depletion by local ized impacts such as effects of depredation and angler interactions (Read 2008, NOAA 2006a). In addition to the documented mortalities increasing numbers of interventions or rescues have occurred in r ecent years along the west coast of Florida, involving dolphins
4 entangled in fishing gear. In January 2007, a resident 1.5-year-old female dolphin (F201) was rescued and admitted to Motes dolphin hospital because of a severe entanglement around its fluke. In summer of 2007, a 42-yea r-old male dolphin and long time resident of the Sarasota Bay area (FB28) was enta ngled with monofilament wrapped around the dorsal fin and flukes. With use of a cutting t ool, the animal was freed from the majority of the threatening line without further inte rvention. Also in 2 007, a 26-year-old male dolphin (F106) that was frequently seen depred ating from a local fishing pier and was the pair of an adult male that died of fishi ng gear ingestion in 2006, disappeared and is presumed dead, possibly as a result of a fishing related injury. Depredation is expected to be a pers istent and rising problem as humans and cetaceans compete for the same resources (R ead 2008). The depred ation behavior and dolphin-angler interactions will likely increase due to a combination of factors like the decline of prey populations due to overfishing from co mmercial (both current and emerging) and recreational fish eries and cultural transmission of the behavior throughout cetacean populations (Myers and Worm 2003, Sutinen and. Johnston 2003, Wells 2003, Coleman et al. 2004, Whitehead et al. 2004, Read 2008). The use of long-term data sets, like those available for the Sarasota Bay bottlenose dolphin community, offers a unique opportunity to employ, experiment, and evalua te the value of exploratory methods. Evaluating and comparing the effectiveness of the methodologies used in this study to evaluate and mitigate depredation will set th e foundation for future research design and dolphin-angler intera ction management.
5 1.1 Goal of this Thesis The overall goal of this thesis is to characterize the be havior of dolphins interacting with anglers (referred to as angler interaction dolphins or depredators) and to identify all possible factors that may be associ ated with the rise in depredation. Chapter one focuses on understanding dolphin depreda tion in terms of animal behavior. Depredation is evaluated in rela tion to fishing effort, red tide ( Karenia brevis ) blooms, dolphin activity budgets, dolphin ha bitat use, dolphin sociali zation patterns, and angler behaviors. This chapter also examines larg er trends in depreda tion, identifying potential patterns over time, age class, sex, and maternal lineages. Chapter two is a case study of an entangled dolphin (F201) and explores th e potential link between hearing impairment and depredation behavior. Chap ter three tests the effectiveness of passive acoustics as an inexpensive tool for monitoring dolphin depr edation behavior at a fishing pier. This thesis also compared the usefulne ss and power of different methodologies in understanding the depredation behavior and its effects on dolphin be havior and biology. Methods included visual and pa ssive acoustic pier surveys, auditory evoked potentials, focal animal behavioral follows, and the use of longitudinal data sets. These results can serve as a base line when examining the pot ential success of met hodologies for studying dolphin-angler interactions in other areas. Another, very central goal of this project was to reduce the human role (specifically focusing on boaters and anglers) in depredati on and other adverse dolphin interactions through public aw areness and outreach. By developing and distributing Dolphin-Friendly Fishing and Vi ewing Tips informational cards, the intention was to
6 reduce depredation by teaching people simple steps that are designed to reduce the association between people and fish that so me dolphins have learned to recognize.
7 2. Behavior: Characterization of Dol phins Engaging in Angler Interactions 2.1 Introduction The Sarasota Bay bottlenose dolphin co mmunity is exposed to a variety of anthropogenic disturbances. Previous studies in the area have focused on the effects of boat presence and noise to change s in dolphin behavior (Nowacek et al. 2001, Buckstaff 2004). Another study initiated in 1997 identified and quantified illegal feeding and boat interactions with a commonl y begging dolphin (Cunningham-Smith et al 2006). The longitudinal data set collected in Sarasota Bay provides a rare opportunity to study, in detail, the rise and effect of anthr opogenic impacts on a dolphin community. Anthropogenic, environmental, and behavioral sources likely all supported the rise of dolphin depredation in Sarasota Bay to detrimental levels. The relative contributions of sources to the increase in dol phin interactions with recreational fishing gear were investigated in or der to evaluate the problem a nd identify specific management needs. Understanding the demographics of depredation behavior, such as sex and age classes most commonly involved, was also nece ssary for the evaluation of the behavior. The frequency of occurrence and when the beha vior first entered in to the population was
8 also considered. Furthermore, Dolphin-Friendly Fishing and View ing Tips cards were distributed in an effort to increase pub lic awareness and reduce dolphin depredation. 2.1.1 Anthropogenic Factors of Depredation and Implications Dolphins have become increasingly exposed to recreational fishing and boaters as the population of resident and visiting humans rises in Florida coastal areas. The Florida population has risen by 135% from 1970 to 2000, with growth skew ed towards coastal areas (Florida Charts 2007). Within Sarasota and Manatee Counties, the home range of the Sarasota Bay resident dol phins, the total number of regi stered boats has quadrupled since 1970 to 44,839 boats in 2005. In addition, marine recreational fishing in the United States increased by 20% from 1996 to 2000 (S utinen and Johnston 2003). Broadly, the Gulf of Mexico and the Atlant ic region have the greatest numb er of saltwater recreational anglers in the nation (Van Voorhees and Pritchard 2008). More specifically, the state of Florida has the greatest number of resi dent (64%) and visiting (64%) saltwater recreational anglers (2,002,000) (U .S. Dept. of the Interior et al. 2006). These factors increase the probability for a dolphin to come within the vicinity of an angler or boater. Furthermore, in 2007, more anglers were engaging in catch and release: 58% of recreational caught fish we re released alive in 2007 (Van Voorhees and Pritchard 2008). As the number of anglers rise with more anglers releasing injured catch, dolphins may begin to associate boats or fishi ng piers with easy pre y, especially if the animals learn that these can be sources of bait, caught fish on line, or released catch. This
9 potential problem would be exacerbated thr ough direct feeding of dolphins, as has been documented in the Sarasota Bay area (Cunningham-Smith et al. 2006). 2.1.2 Environmental Factors of De predation and Implications Natural environmental factors were also examined relative to the increase in depredation events. Red tide, a form of ha rmful algal bloom (HAB) common to the Gulf of Mexico, involves blooms of a toxic dinoflagellate, Karenia brevis resulting in massive die-offs of fish. K. brevis produces a suite of brevetoxins (neurotoxins) which impact many organisms through inhalation and/or trophic transfer (Tester et al 2000, Flewelling et al 2005, Fire et al. 2008). Brevetoxins can cause mortalities for marine mammals and studies have shown that prey fish act as vect ors for toxin transfer to dolphins during a red tide outbreak (Flewelling et al 2005, Fire et al. 2008). A severe red tide, such as occurred in Sarasota Bay and surrounding areas in 2005, can deplete prey fish populations and may force dolphins to search for other means of nourishment, such as angler s bait or catch. Ongoing studies by the Sarasota Dolphin Research Program have shown that the 2005 bloom not only depleted dolphin prey but also caused a change in composition of the local fish community to more pelagic species, not typically found in the diets of resi dent dolphins (Barro s and Wells 1998, Gannon et al. 2009). These effects of red tide on prey species were correlated with declines in dolphin body condition (Wells et al. 2006) and behavior changes such as increased group size, shifts in habitat use, and increased reports of begging an d depredation (Gannon et al. 2009).
10 2.1.3 Behavioral Factors of Depredation and Implications Habitat Selection and Home Range When investigating habitat selection a nd home range, depredating dolphin habitat preference for heavily fished areas must be measured (Reeves et al. 2001). Results from previous behavioral fisheries interac tion studies (e.g. Chilvers and Corkeron 2001, Cunningham-Smith et al 2006, Finn et al. 2008) would suggest that dolphins interacting with anglers may more commonl y select for areas with hi gh numbers of passing boaters or a concentration of anglers, such as near fishing piers, and in passes or channels. Depredating dolphins may also have ove rall, smaller home ranges since dolphin movements depend largely on the lo cation of prey species (Shane et al. 1986, Ballance 1992). For example, the encounter rate s for begging dolphins in Cockburn Sound, Australia were significantly correlated with the density of recreational boats (Finn et al. 2008). In Moreton Bay, Australi a, a coastal community of Tursiops aduncas known to feed in association with a tr awl fishery, selected for dee p, offshore habitat conducive to trawling rather than shallow, coastal areas and had ranges half the size when compared to non-trawler associated dolphins (Chilvers and Corkeron 2001). Activity Budgets Activity budgets and changes in behavior al states can be valuable tools for assessing conservation impacts to cetacean popu lations and have been used to assess potential anthropogenic impacts related to boat and swimmer disturbance in multiple studies (Lusseau 2003, 2004, 2006; Constantine et al 2004, Danil et al. 2005, Bejder et al. 2006, Williams et al. 2006). Baseline data collected by Waples (1995) found that in
11 summer months resident dolphi ns in Sarasota Bay spent 67% of time traveling, 14% feeding, 13% milling, 5% socializing, and 2% re sting. It is particularly important to understand and quantify changes to the natura l activity states as a result of anglerinteraction behavior. Social Behavior The well-established ability of dolphins to learn by observation may increase the frequency of depredation behavior through social transmission (Donoghue et al 2002, Wells 2003, Whitehead et al. 2004, Cunningham-Smith et al 2006). In general, dolphin foraging and feeding behaviors in Sarasota Ba y are composed of a number of behaviors which demonstrate the behavioral plasticity of these animals (Nowacek 2002, Wells 2003). In particular, an angler-interaction dolphin may have a gr eater probability of having a depredating mother since bottlenose dolphins teach feeding behaviors to their calves (Nowacek 2002, Wells 2003). Sarasota Bay bottlenose dolphins live in a fission-fusion society where group compositions and dolphin associates cha nge by minutes, hours or days (Wells et al 1987). Typical group size for bottlenose dolphi ns ranges from two to fifteen animals (Shane et al. 1986) and for Sarasota Bay, group size is usually between five to seven animals (Wells et al. 1980, 1987). Sarasota dolphins are typically found in larger groups in open waters with large, patchy prey school s whereas smaller groups or solitary animals are typically found in seagrass beds where pr ey is more evenly distributed (Wells et al. 1980, 1999; Shane et al. 1986). The idea that group size is influenced by prey availability is further suppor ted by the large group sizes recorded in Sarasota Bay during
12 a severe harmful algal bloom in 2005 in wh ich only clupeids, a schooling fish, were known to thrive as other prey spec ies were utterly wiped out (Gannon et al. 2009). Depredation lends itself to a smaller gr oup size or solitary foraging strategy since multiple animals are not needed to cooperati vely locate or herd fish. Depredation typically involves a single prey item struggli ng on a line, a debilitated thrown-back fish, or the hand-feeding of fish to dolphins. A ll of these scenarios would be optimal for a solitary forager. For example, mother-calf pairs ( Tursiops sp.) in Australia did not form as large or as cohesive of groups and calve s had fewer associates inside provisioning areas than in non-provisioning ar eas (Mann and Smuts 1999). 2.2 Methods 2.2.1 Study Area and Population This study was conducted within an approximately 125 km2 area including Sarasota Bay, Florida and surrounding wate rs (including southern Tampa Bay, Palma Sola Bay, Anna Maria Sound, Venice Inlet and coastal waters) (Figure 2.1). This area is home to a community of about 160 resident bottlenose dolphins th at has been closely monitored by the Chicago Zoological Societys Sarasota Dolphin Research Program (SDRP) for over 38 years (Wells 1991, 2003). Monitoring is conducted through monthly photo-identification surveys a nd occasional health assessments (Wells 1991, 2003). This community of dolphins was appropriate for study because of the incidence of dolphinangler interactions and the wealth of data th at already exists on fa mily lineages, stranding
13 records, age, sex, behavioral history, distributi on, social associations, and hearing abilities (Wells 1991, 2003; Cook 2006). 2.2.2 Behaviors of Interest Activities measured for this project were defined by the Sarasota Dolphin Research Program (SDRP 2006) and were s upplemented with definitions specific to angler interaction behaviors of interest (Tables 2.1, 2.2) Working definitions for additional behavioral categories were deve loped for the terms patrol, beg, scavenge, attempted depredation, line depredation, and provision (Table 2.2). Definitions of bait and catch were also established. Bait was define d as a fish, invertebrate or part of either that was attached to an anglers line and th en lowered into the water with the intent of catching fish (e.g. shrimp or pinfish). Ca tch was defined as a fish that was free swimming in its natural habitat before it wa s hooked and caught by an angler (catch was not present when the angler initially lowered the line into the water). However, a fish initially defined as catch is not finite. In some instances, an angler may then use the catch as bait. In this case, the fish would now be redefined as bait. During the course of this study, SDRP members that witnessed a dolphin-angler interaction were asked to record if the dol phin took bait or catch from the line, the size and species of depredated item, the number of fishing lines in the water, and the distance (m) of the dolphin from the angler.
14 2.2.3 Distance Estimation Distance estimation was essential for m easuring interactions between dolphins and anglers. At the beginning of each day, th e field team practiced distance estimation at the Sarasota Sailing Squadron mooring field. Field workers individually estimated the distance of 25 randomly-selected boats or buoys 10 to 300 m away and then compared estimates to a reading from a laser rangefinder. I estimated most distances in the field and stationary distances for boa ts or anglers were always c onfirmed with the rangefinder. My mean error for distance range categories was calculated by computing the absolute value of the difference between the actual distance and the estimate (n=2391) (Figure 2.2). Mean error increased with object distance, ranging from 4 m (SD=7.92) for distances between 10-25 m to 29 m (SD=22.52) for dist ances greater than 201 m. 2.2.4 Pier Observations Four fishing piers and two jetties within the home range of the Sarasota dolphins were surveyed for dolphin presence, fishing e ffort, environmental conditions, and dolphin depredation behaviors. In May, June, and July of 2007, Anna Maria City Pier, Rod and Reel Pier, Venice Pier, and the north and sout h Venice jetties, were monitored on rotation five days a week (weather permitting), usuall y including both weekend days (Table 2.3, Figure 2.1). Surveys were conducted between 06:30 and 18:30 for an average of 6.5 hours. Efforts in later afternoon hours were le ss successful due to frequent thunderstorms forcing the suspension or end of surveys. From October 2007 through April 2007, Anna Maria City Pier, Rod and Reel Pier, and Br adenton Beach City pier (added in October 2007) were monitored on rotation four times per month (weather permitting) for seasonal
15 changes in dolphin depredation and fishing effo rt. Sustained effort on piers or jetties was based on the ease and quality of data collection, opport unities available ( e.g. ability to use a hydrophone), number of residents sighted, and apparent or possible dolphin depredation behavior. A total of 64 pier /jetty surveys were conducted. The Venice Pier and jetties were droppe d from the project due to a lack of observations of dolphin-angler interactions and the difficult circumstances presented to monitor and identify animals. The Venice jettie s were adjacent to an inlet, a corridor of heavy boat use, and half of the dolphins identified were not members of the study population, further reducing the va lue of these sites. Venice Pier is approximately 6 m above the water, making photo-identification very difficult and during surveys, only one dolphin was seen farther than 200 m. Bradent on Beach City Pier was added to the study in October 2007, after it re-opened following th ree years of closure. The pier was of much interest since dolphins had often been s een feeding in this area and I wanted to document the possible transition of the area into a site of depredation once the pier reopened. Surveys of each pier/jetty involved constant monitoring for dolphin presence within a 200 m radius (Figure 2.3). All fiel d workers rotated visu al monitoring areas on the half-hour to reduce sightability bias. Standard SDRP sighting data were collected and photo-identification was completed using a Nikon D100 digital camera with a 70-300 mm lens and standardized photo-identific ation techniques (SDRP 2006, Wrsig and Wrsig 1977). During surveys, every half hour, data was collected on the number of lines in the water, if bait or lure was in us e, the presence or absence of dolphins within 100 m, and environmental conditions (surface water temperature, Beaufort sea state,
16 cloud cover). Bait or lure had to be visual ly confirmed by a field worker and was not assumed based on comments from an angler, fishing technique, or presence of a bait bucket. If the animal stayed within 100 m of th e pier for five minutes or more, a focal animal behavioral follow was started. If more than one dolphin was present, I selected an adult or juvenile focal animal (focal) at random (calves were excluded). Using standardized techniques (Altmann 1974, Ma nn 1999), I collected instantaneous data every three minutes on the dolphins associ ates, group spread, and number of active fishing lines within 15 m of the animal. The total number and type of depredation events and the focals dominant activity (the activity in which the animal engaged in for the majority of observations over the three-mi nute interval) were documented (Tables 2.1, 2.2). The follow was terminated when the focal dolphin left the 100 m radius around the pier for nine consecutive mi nutes (three time-points). When an act of depredation was obser ved, the depredating dolphin, depredation type, size and species of catch/b ait taken, lost fishing gear, nu mber of lines within 15 m, reaction of the angler, the dol phins associates, and the beha vior of the dolphin following the event was documented. A single depreda tion event was defined as a single act or multiple consecutive depredation acts by a single dolphin involving the same individual (angler or boater) or set of individuals (e.g. three anglers on a fishing boat). Photos or video of depredation events were obtained whenever possible. In order to avoid influencing angler be havior during surveys, the field team was restricted from wearing clothi ng or bringing gear with dol phin images or organization logos. When asked by anglers about the spec ifics of the study during a survey, team
17 members provided vague responses, and did not indicate that the focus of the study was dolphin depredation and angler interaction behavior. 2.2.5 Focal Dolphin Selection and Fo cal Animal Behavioral Follows Depredating focal dolphin (n=8) selec tion was based on history of angler interactions (dolphins had inte racted with a recreational angl er or were entangled on at least two occasions prior to Ju ly 2007). A matching set of control dolphins (n=8) never observed interacting with anglers but with sim ilar ages (+/5 years) sex and home ranges were selected to match a specific depredati ng dolphin (Table 2.4, Figur e 2.4). In total, 78.9 h and 66.1 h of follow time were completed for depredator and control focal dolphins, respectively. Focal animal behavioral follows were conducted from a 19-foot center-console outboard boat in summers of 2007 and 2008. Using standardized techniques developed for work with Sarasota Bay dolphins (Altmann 1974, Mann 1999, SDRP 2006), focal dolphins were followed for up to two hours per day and instantaneous data was collected on position, associates, group spread, habitat, an d number of active fishing lines within 50 m and fishing boats within 100 m every thr ee minutes. Possible habitats in Sarasota Bay included coastal Gulf waters, open bay, sand, seagrass meadows, mangrove, channel, and pass (SDRP 2006). Activity was not instanta neous, but recorded as the behavior in which the dolphin engaged in for the majo rity of observations over the three-minute interval (Tables 2.1, 2.2). Po sition was collected with a Ga rmin GPS 12, and associates within 200 m were identified using photo-identificati on (SDRP 2006). Acts of depredation were documented in the same ma nner as with pier/jetty surveys. After
18 plotting initial activity and habitat changes for each focals first follow, a two hour follow time period was found to be the best option since activity and habitat choices varied widely in duration and the depredation behavior was difficult to capture. In addition, respiration data were recorded continuously throughout the follow: each time the dolphin took a breath, the hour, minute, a nd seconds were recorded. 2.2.6 Methods of Analyses Depredation Longevity and Demographics Data on interactions between dolphins and anglers from observations during 1972 to 2007 were considered when examining rate s of occurrence over time. However, only dolphin-angler interac tion rates for the year s of 2000 to 2007 were used for analyses. This time frame was chosen for several reason s: 1.) it is after th e 1995 implementation of a state-wide commercial net fi shing ban, 2.) it follows enactment of amendments to the U.S. Marine Mammal Protection Act and cour t upheld prohibitions making it illegal to feed wild dolphins, 3.) and it includes a period of consistent data collection and field effort in Sarasota Bay. Microsoft Excel 2003 was used to compile and calculate descriptive statistics. Upda ted data on Sarasota Bay dolphin abundance and residency were provided by Randall Wells (unpublished data). I calculated yearly and monthly dolphi n-angler interacti on rates (2000-2007) based on possible and confirmed depredati on and related behavi ors (patrol, beg, scavenge, and provision) (Table 2.2). All rates were standa rdized by effort (number of boat-days). Dolphin-angler interaction rates were c onsidered for 2000-2007 under two conditions: 1.) only for incidents that o ccurred during monthl y photo-identification
19 surveys, and 2.) for all sighting effort (inc luding survey, opportunistic and capture-release sighting effort) in Sarasota Bay. Surveys offe r a high level of consistency in that every month the SDRP field team completes 10 dail y surveys (weather-permitting) covering the entirety of the survey range twice. Dolphinangler interaction rates were further analyzed by excluding data on interactions with a long -time, regularly-begging dolphin (BEGR) in order to look at recent changes in depredat ion behavior among Sarasota dolphins. In an attempt to relate dolphin-angler interaction rates to variations in the presence of anglers and boats in the study ar ea, several proxies for human activities were considered. The mean dolphin-angler interaction rates from mont hly photo-identification surveys and all sighting effort were compared to hotel/motel occupancy rates in Sarasota County for 2000-2007 as well as to boat fu el sales for 2003-2007 from Cannons Marina, located south of Longboat Pass within the hom e range of Sarasota dolphins. A Spearman rank correlation in StatSoft Statistica 6.1 was used to compare interaction rates and sales. Based on all archived field notes, all dolphins that were part of the Sarasota Bay dolphin community from 2000 to 2007 were considered for entanglements or angler interaction behavior. Dolphins were classified as either a confirmed depredator (an animal that has clearly engaged in patrolling, begging, scavenging, attempted depredation, line depredation, or provisioning ) or a possible depredator (field notes were vague). A percentage of the number of resi dent dolphins that depredated during the specified year was then calculated. Cumulative values were also calculated in order to determine what percentage of the current dol phins in the population had ever participated in angler-interaction behavior. A dolphin was taken out of the analysis the year after its
20 death. If a dolphin went missing for at least one year, that dolphin was assumed dead and was only considered in the popul ation as of the date of its last sighting. Data from monthly photo-identification surv eys as well as all sighting effort were analyzed. Based on age-sex class composition of th e Sarasota population, expected numbers for male and female, immature and mature confirmed depredators were calculated and compared to the observed number depredat ors of known age and sex (n=50) using a goodness of fit G-test in PopTools (version 3.0, bu ild 5). Animals that had a coefficient of association (COA) of 0.50 or greater with their mother were considered calves; if less than 0.50, the dolphin was considered a juvenile Females were not considered adults until they gave birth to their first calf, and males were considered adults at age 13. Depredation related to Red Tide (Karenia brevis ) Blooms Monthly dolphin-angler interaction rates were compared across K. brevis bloom months, months following K. brevis blooms, and non-bloom months from 2000-2007. A bloom with potential for killing fi sh was considered to occur when K. brevis cell counts reached greater than or equal to 100,000 cells / L for three consecutive weeks (Spencer Fire, personal communication; Steidinger et al. 1998). If any day was considered part of a K. brevis bloom, the month was counted as a bl oom month. A bloo m was considered complete on the day when cell counts fell below 100,000 cells/L and cell counts remained below this threshold value fo r a consecutive thr ee weeks. Daily K. brevis cell counts were provided by Mote Marine Laboratory, courtesy of Gary Kirkpatrick. The three months following a bloom (or less, if a new bloom occurred) were deemed the lag period. The lag period is the time following a red tide when fish stocks are still exposed
21 to brevetoxins remaining in the environmen t and environmental recovery is beginning (Fire et al. 2007, 2008; Gannon et al. 2009). There were six red tide blooms in Sarasota Bay from 2000-2007; SeptemberDecember 2001, July-September 2002, January-October 2003, January-February 2004, January-December 2005, and August-December 2006. Statistical software SPSS 16.0 (Graduate Student version) for one-way ANO VAs and post-hoc Tukey tests was used to compare bloom periods with respect to m onthly dolphin-angler interaction rates. To further explore differences among red tide periods and make comparisons to the number of anglers on the water, I calcula ted dolphin-angler inte raction rates per unit of fuel sales, by dividing the monthly dolphin-angler intera ction rate by the corresponding monthly fuel sale. I then used an ANOVA in SPSS to compare across the three HAB categories. Habitat selection For each follow, the number of times each habitat was recorded was tallied (habitat per time-point was counted as 1, unless the habitat was split in which each habitat would then count as 0.5.) Cumulativ e habitat scores were averaged across the total number of time-points, and further aver aged across the total number of follows for an individual. Individual averages were combined to determine the overall mean for control and depredator focal animals. Habitat means for control and depredator focal animals were then compared statistically using a custom randomization program built in MATLAB (version 7.4). The program randomizes a focal follow habitat or activity matrix 10,000 times in order to estimate the probability of randomly finding differences in habitat or activity means
22 between depredator and control dolphins. The analysis was run as a one-tailed test using a test statistic calculated as the ratio be tween depredating and control dolphins. Proximity to Fishing Lines and Boats At every three-minute inte rval during behavioral follows, the number of actively fishing boats within 100 m and active fishing lines within 50 m of the dolphin were recorded. The mean number of boats or lin es per three-minute interval of a follow was calculated, then averaged across all follows for the dolphin, and further averaged over depredator and control categories. Compar isons between depredator and control focal dolphins proximity to boats or lines we re made using Mann-Whitney U tests in Statistica. Home Range Home range determination for all focal animals included all sightings from 2004 (the first year with a dolp hin-angler interaction, not in cluding begging) to August 2008. I used the Animal Extension Movement Spat ial Analyst (version 2.04 beta) in ArcView GIS 3.3 and XToolsPro (version 1.0.1, build 19) in ArcMap (version 9.0) to calculate the areas of the 95% kernel home ranges and 50% core kernels for all focals. The area of overlapping land was clipped and subtracted from the core and range areas. The 95% and 50% kernels of depredator and control focals we re then compared using a Students t-test in SPSS.
23 Pier Data from half-hourly scans were used in a logistic regression in Statistica to determine the effectiveness, if any, of the number lines with bait, number of lines with lures, and total number of lines as predictors for dolphin presence at a fishing pier. I examined fishing effort effects on dolphin presence for monthly, seasonal and hourly timescales. Using a one-way ANOVA in SPSS, monthly averaged differences in dolphin presence, number of baited lines, number of lines with lures, and total number of lines were examined. Seasonal analyses we re performed in a similar manner after categorizing months as rainy (summer/fall) or dry (spring/winter) using a Mann-Whitney U test in Statistica. Rainy season (n=42 surveys) included May through October and dry season (n=22 surveys) included November through April. In addition, using an ANOVA and a post-hoc Tukey test in SPSS, I looked at differences in dolphin presence and total fishing effort as a result of time of day. Fo r purposes of analysis, times were considered in hour blocks (e.g. 6:00-6:59). Activity Budgets I created activity budgets for each focal an imal and stratified them relative to their classification as depredators or controls. Activities recorded as dominant in a threeminute interval were mill, forage (combination of feed and probable feed behaviors), travel, rest, social, beg, patrol and provision. Activity data were prepared and analyzed using the same methods as with the habitat data.
24 For further analysis, non-natural fora ging behaviors (begging, provisioning, and patrolling) and natural foraging were collapsed in order to determine if depredators and control focals spend similar amounts of time foraging. Depredation Depredation data were recorded thro ugh concerted efforts by SDRP members during 2007-2008 while conducting monthly photo-identification surveys, research for this study, or various other res earch initiatives. Detailed not es on depredation taken from SDRP staff and students were analyzed for commonalities in dolphin-angler interaction behaviors. The aspects of depredation consider ed were the species of fish depredated (or attempted), the most common types of depredation (e.g. scavenging, provisioning, attempted depredation, or line depredation), lo st gear, and angler or boater reaction to depredation. Social Behavior: Group Size For Sarasota Bay, a group is defined as all dolphins associat ing within 100 m (Wells et al. 1980, 1987). Group size for each focal dolphin was determined using all available sightings between the years of 2000-2007 from the Sarasota Dolphin Research Program database. Group size means were fu rther calculated for de predator focals when engaging in angler interacti on behavior versus the overall mean. Mean group size was also calculated for all focals based on data from focal animal behavioral follows from summers of 2007 and 2008. A Students t-test was performed on these data sets in SPSS.
25 Social Behavior: Coefficients of Association Coefficients of associations (COAs) we re calculated for each focal animal from two data sets: focal animal behavioral follows and the Sarasota Dolphin Research Program database. For purposes of the behavi oral focal follow analysis, all identifiable dolphins within a 100 m radius of the focal dol phin recorded at the three minute interval were considered an associate (Wells et al. 1987). The total time the associate and focal were recorded together was tallied and then divided by th e total minutes spent following the focal in 2007 and 2008. The associates that spent at least 10% of total follow time with the focal were used in a further analysis to determine if depredator focals spent significantly more time with other angler intera ction associates than did control focals. I also calculated top associates for focal animals using the SDRP database. Only sightings from monthly photo-identification surveys from 2004 (first report of dolphinangler interaction behavior) until May 2008 were included in the analysis. For each focal animal, the half-weight index COA of all possible associates was calculated. Only associates with COAs of at least 0.10 were in cluded in a further analysis to determine if depredator focals spent significantly more time with dolphins that also engaged in angler interaction behavior than di d control focals. Microsoft Excel and the SPSS Students ttest were used for both COA analyses. 2.2.7 Educational Outreach: Card Design and Distribution In an effort to raise public awarene ss for depredation and other adverse humandolphin interactions, guidelines on Best Fishin g Practices for Avoiding Interactions with Wild Dolphins developed by NOAA, Mote Ma rine Laboratory, the Chicago Zoological
26 Society, Hubbs-Sea World Research Institut e and anglers and fishing guides were reproduced into an informational card, Dol phin-Friendly Fishing and Viewing Tips. The card is compact and water resistant to a llow for card storage in pockets, tackle-boxes, or boat consoles. The first edition of the ca rd featured a close-up of a dolphin face on the cover (Figure 2.5). The second edition of the card was redesigned to include a variety of images on the cover in order to elicit the atte ntion of the broad target audience including, recreational boaters and anglers (both tourists and locals) (Figure 2.6). Inside the card are message points an d tips about how to avoid or address interactions with wild dolphins The cards offer bold headings as well as explanations in laymans terms of each dolphin-friendly fishi ng and viewing tip to help the audience understand how such actions can be helpful. The information is presented in a positive manner to encourage boaters and anglers to buy in to the mitigation measures voluntarily. The second edition of the card offers slightly more condensed text with larger font (Figure 2.6). The first edition of the Dolphin-Fr iendly Fishing and Viewing Tips card (64,800 copies) was printed in January 2008. An additional 51,800 first edition cards were printed in February 2008. Beginning in January 2008, cards were distributed to a variety of facilities w ith the ability to reach the target au dience with the majority of cards going to conservation and edu cation organizations such as aquariums and universities primarily within the state of Florida (Figure 2.7). Other areas of high card distribution included Florida state and local governments as well as fishing, bait and tackle shops, boat rental facilities, an d marinas (Figure 2.7).
27 Local distribution was co nducted in January and February of 2008 and a second delivery and restock took place in May 2008. Th e focus of local card distribution within Sarasota and Manatee counties (counties coasta l to the Sarasota Bay dolphin community home waters) included Mote Marine Labor atory & Aquarium, Coast Guard Auxiliary chapters, fishing, bait and tackle shops, boat rental facilities, mari nas, fishing piers and jetties, and waterfront rest aurants. The second edition of the card was printed in November 2008. The printing supplied 228,950 English cards and 17,300 Spanish version cards. Local distribution of sec ond edition cards was conducted in November and December 2008. 2.3 Results 2.3.1 Depredation History and Demographics With the exception of two consistently-begging dolphins, MOCH and BEGR, only four dolphin-angler in teraction cases were re ported from 1972 to 2000. The dolphin-angler interaction rate (2000-2007) for monthly surveys rose continually since 2003 and for all sighting effo rt, rates began rising consistently in 2005 (Table 2.5; Figures 2.8, 2.9). The peak of angler interactions in 2001 for all sighting data is explained by a study focused on BEGR rega rding begging rates. When the dolphin, BEGR, was excluded from the analyses, ra tes continually rose from 2003 for both monthly survey and all sighting effort (Tab le 2.5; Figures 2.8, 2.9). The number of animals involved in confirmed depredation t ype behaviors has been rising since 2004
28 (Table 2.6, Figures 2.10, 2.11). Cumulatively, it was estimated that between 13 to 27% of the 2007 population had engaged or possibly e ngaged in an angler interaction one or more times from 2000 to 2007 (Table 2.6, Figures 2.12, 2.13). The expected ratio of immature and mature males and females were significantly different from expected values ( p =0.003) with adult males being the most over-represented and adult females being the most under-represented depredator age-sex class (Table 2.7). Angler interaction rates we re greatest in March with smaller peaks in May and November (Figures 2.14). Hotel/motel occ upancy was greatest in Sarasota County in March (Figure 2.15). Boat fuel sales rose in February and March and remained high throughout the summer months, and there was a small increase in sales in November (Figure 2.16). However, no significant correlations were found between angler interaction rates and hotel occupancy or boat fuel sales ( p=0.99, p=0.077, respectively). 2.3.2 Depredation related to Red Tide ( K. brevis ) Blooms and Boating Activity Dolphin-angler interaction rates were significantly different between K. brevis lag months (0.347 dolphin-angler interactions/lag month), K. brevis bloom months (0.089 dolphin-angler interac tions/bloom month), and non-bl oom periods (0.133 dolphin-angler interactions/non-bloom month) ( p=0.031) (Figure 2.17). A posthoc test revealed that dolphin-angler inte raction rates for K. brevis bloom and lag periods were significantly different ( p=0.024). Significant differences were also found between HAB periods and interaction rates with regard boat fuel sales ( xbloom=1.58x10-5, xnon-bloom=3.04x10-5, xlag=3.77x10-5; p=0.034). A post-hoc Tukey test reveal ed that the greatest differences was between red tide and lag periods ( p=0.052).
29 2.3.3 Habitat Selection Depredating and control animals were not found to spend significantly different amounts of time in different habitats (Fig. 2.18). However, depredation by three focal animals during the 2007-2008 was only seen to occu r in three habitats: open bay (around a fishing pier), channel, and pass. 2.3.4 Proximity to Fishing Boats and Lines Depredator focals were not significantly more likely to be within 100 m of an actively-fishing boat than were controls ( x depredator =0.05, SD=0.01; x control=0.015, SD=0.04; p=0.32). However, depredator focals we re significantly more likely to be within 50 m of active fishing lines ( x depredator=0.10, SD=0.21; x control=0.01, SD=0.03; p=0.02). 2.3.5 Home Range No statistically significant difference was found between control or depredator focals for 95% kernel home range ( x depredator=94.02 km2, SD=73.28; x control=135.86 km2, SD=83.15; p=0.304) or 50% core kernel areas ( x depredator=14.88 km2, SD=13.46; x control=22.33 km2, SD=15.92; p=0.330) (Figure 2.4). 2.3.6 Pier Numbers of lines (bait, lure, or total) we re not predictors of dolphin presence. For baited lines and dolphin presence, lines with lures and dolphin presence, and total number
30 of lines and dolphin presen ce, logistic regression R2 values were 0.001, 0.004, and 0.001, respectively. March 2008 had the greatest means for both total fishing and dolphin presence, however, no significant differences between pier survey months were found for any tested conditions, including dolphin presence (p=0.513), number of baited lines ( p=0.073), number of lines with lures ( p=0.262), and total number of lines ( p=0.136) (Figure 2.19). Total fishing effort was significantly greater during the rainy season than the dry season ( x rainy =5.30368, SD=5.31; x dry =5.12999, SD =5.13; p=0.02). However, there was no significant differen ce between seasons for dolphin presence ( p=0.301), baited lines ( p=0.44), or lines with lures ( p=0.20). Dolphin presence at piers was found to vary significantly with time of day ( p=0.001). Specifically, a post hoc test revealed that ther e were significantly more dolphins at the pier at 16: 00 than during 09:00:00 (Figure 2.20). Fishing effort also varied significantly with time of day ( p<0.001) increasing in the early morning and remained elevated through mu ch of the day (Figure 2.20). 2.3.7 Activity Budgets Depredating focals were found to spe nd significantly more time milling ( p=0.05) and significantly less time enga ging in activities like foraging (p=0.05) and traveling ( p=0.02) (Figure 2.21). When the non-natural and natural foraging strategies were collapsed, there were marginally to signifi cant differences with depredators spending more time milling ( p=0.056) and less time traveling ( p=0.02) (Figure 2.22).
31 Individual activity budgets reveal the differences betw een depredators and their control counterparts (F igure 2.23). In total, four depredator focal animals (BEGR, F106, F109, and F232) were recorded engaging in angler-interaction behavior as a dominant activity during a follow. 2.3.8 Depredation During pier surveys and behavioral focal follows, 58 acts of depredation (including attempted depredation, line depr edation, provisioning, and scavenging) by a total of four animals were documented. Of these four animals, one was a known animal (BEAN) from the Eckerd College Dolphin Pr ogram in St. Petersburg, Florida, and the other three dolphins were focal Sarasota Ba y residents (BEGR, F106, F109). Overall, the most common type of depredation was pr ovisioning, followed by scavenging, attempted depredation, and line depredation (Figure 2.24). However, each dolphin appeared to exhibit a different depredation strategy. BE GR depredates solely by provisioning. F109 was only seen scavenging and F106 engaged in all four types of de predation behavior. BEAN depredated mainly by taking or attemp ting to take fish off fishing lines. After a dolphin depredated the first fis h, in 95% of documented cases the animals continued to engage in a depredation or beha viors like begging or pa trolling. At least 11 different species of fish were depredated (or attempted). No particular fish species was favored by dolphins. However, the numbers are hi gh for bait fish and sardines as this is what was commonly used by anglers and thus, fed to dolphins. The 2007-2008 reports from other SDRP fiel d initiatives showed that of the 12 confirmed depredation events witnessed (not including begging or patrolling), 42% were
32 scavenging, 25% were provisioni ng, 17% were attempted depr edation, and 17% were line depredation. When added to the total of depredation ev ents collected for this study, provisioning was the most common type of depredation, followed by scavenging, attempted depredation, and line depredation (Figure 2.25). Angler or Boater Reaction to Depredation Few people engaged in the recommended tips of moving locations or reeling in their line to avoid further interactions with the dolphin following a dolphin depredation event (Figure 2.26). Most people encouraged another interaction by either continuing to feed the dolphin or summoning it closer to the pier or boat (Figur e 2.27). Other anglers or boaters engaged in negative behavior towards the dolphin such as taunting or intentionally casting at the animal (Figure 2.26). Nineteen percent of anglers or boaters showed no change in behavior following a dolphin depredatio n event (Figure 2.26). 2.3.9 Social Behavior Group Size The mean depredator focal follow group size of 2.3 (SD=0.7) was not significantly different from the contro l focal follow group size of 3.3 (SD=1.4; p=0.087). From the SDRP database sightings, the mean depredator focal group size ( x= 5.0, SD=1.2) was significantly less than the control focal group size ( x= 6.5, SD=1.6; p=0.052). Group size during angler -interaction behavior ( x= 3.0, SD=1.1) was significantly less than the overall gro up mean for the depredator focal ( x= 5.0, SD=1.2; p=0.010) (Figure 2.27).
33 Association Patterns Confirmed Sarasota depredator dolphins that spent greater than or equal to 10% of follow time with a focal accounted for 5.5% of control focal and 32.1% of depredator focal associates. However, neither control nor depredator focals spent significantly more time with depredator associates than non-depredator associates ( p=0.759, p=0.737, respectively). None of the control focal an imals had a depredator as their most common associate, but 25% of depredator focals had a fellow depredator as their most common associate. From the SDRP database, 20% of contro l focal and 24.1% of de predator focal top associates (COA 0.10) were confirmed depredators. However, neither control nor depredator focal dolphins had significantly grea ter COAs with depredator associates than with non-depredator associates (p=0.659, p=0.095, respectively). Tw enty-five percent of focal controls and 50% of focal depredator s most common associates were confirmed depredators. Control focals had significantly higher levels of association with their top associate than did depredator focals (control x =0.198, depredator x =0.178, p=0.0502). 2.3.10 Case Studies: Beggar and F106 Beggar Beggar (BEGR) is a confirmed male resident of the Sa rasota Bay dolphin population, first observed as a subadult in 1990. This animal has never been temporarily captured for a health assessment, because he is rarely outside the deep water of the ICW. Beggar was first seen on 10 August 1990 al ready exhibiting begging behavior.
34 A typical BEGR begging sequence begins in the slow speed zone with the dolphin surfacing approximately 25 m in front of and pe rpendicular to the bow of an approaching boat (Figure 2.28). Beggar then parallels the ve ssel (less than a 1 m in distance) on either the starboard or port stern and rolls with his ventral side facing the boat. Beggars heading remains in line with boat (Figure 2.28). If the animal is further enticed either by a boater putting hands over the boats gunwale or offering fi sh, BEGR will spyhop out of the water with his mouth open (Figure 2.28). Beggar further pursues the boat by swimming just centimeters below the propeller, a nd surfacing on the right or left side of the engine (Figure 2.28). Beggar has been observed interacting w ith a boat in 91% of 317 sightings from January 1990 through August 2008. From the 14 hours of behavioral focal follows that were conducted in summers 2007 and 2008, BEGR spent nearly 69% of time begging from watercraft. This dolphin has never b een seen outside of approximately a 6.8 km extent of the ICW in which 2.2 km are slow speed for vessels (Cunningham-Smith et al. 2006). BEGR depredates purely through provisi oning and has also been suspected of teaching other dolphins how to beg (Cunningham-Smith et al. 2006). In at least one instance, the calf of a begging female associated with BEGR died with numerous indications of human inte ractions (Cunningham-Smith et al. 2006). Beggar is a well-known attraction and bait to feed the dolphin has been blatantly sold to boaters from a nearby marina. In 2006, NMFS sent a number of businesses in this area a letter and educa tion packet reminding them of the MMPA, specifically concerning the ille gality of feeding wild dolph ins (Stacey Horstman, NMFS, personal communication). Includ ed in the letter were responsible viewing guidelines for
35 businesses such as avoiding advertisements th at depict people feedi ng or touching a wild dolphin (NOAA 2006 b). A study initiated in 1997 compared pre a nd post educational efforts, and showed little success in reducing BEGR-human interac tions in the absence of marked vessels dedicated to reducing interactions (Cunningham-Smith et al. 2006). The 1997 Beggar study did not distribute educational inform ation to the numerous businesses around the area, but instead focused on educational efforts within the waterway through direct interactions with boa ters and increased si gnage (Cunningham-Smith et al. 2006). By distributing Dolphin-Friendly Fishing and Viewing Tips car ds to local businesses and organizations frequented by boa ters and anglers, the exp ectation was that a broader audience could be reached a nd education would become effective before boaters or anglers were approached by BEGR (Figures 2.5, 2.6). Cards were distributed throughout the Sarasota Dolphin home range, however only businesses from south of the Sarasota Bay Ringling Bridge to the Englewood area were included in this analysis. This enabled me to look at more fine-scale effects of educational efforts through card distribution. The area just south of the Sarasota Bay Ringling Bridge was included in a special e ffort to incorporate a large yacht marina, Marina Jacks. Vessels with greater than a 5 foot draft docked at this marina only have two options for reaching the Gu lf of Mexico: travel 21 km north to reach Longboat Pass or 25 km south (through BEGRs range) to pass through Venice Inlet. Between January 2008 and May 2008, 4,680 cards were distribu ted to 28 different businesses or organizations within this area, including Ma rina Jacks. Major points of distribution included bait and tackle shops, boat-rental agencies, and waterfront restaurants.
36 The months of May-August 2007 were chos en as pre-card distribution months and for post-card distribution, the months of May-August 2008 were selected. The same four summer months were selected for pre and post distribution comparison in order to control for ecological effects of prey availability and seasonal effects of tourist seasons in the area. Although distribution began in January 2008, the decision was made to not to analyze for educational effects until Ma y 2008 in order to allow time for card dissemination from their initial distribution point to the public. During the selected pre and post time fram es, the number of sightings, the total length of time across sightings, and if BEGR was begging and provisioned were recorded. A provisioning rate was calc ulated by dividing th e total number of provisioning events by the total observation hours. In May-August 2007, BEGR was sighted 14 times (276 min of observation), wa s observed begging in all sightings, and was provisioned seven times by five different boaters. For 2007, Beggars provisioning rate was 1.522 provisioning events per hour In May-August 2008, BEGR was sighted 16 times (696 min of observation). Beggar wa s again observed begging in all sightings and provisioned on 12 incidents by seve n different boaters. For 2008, Beggars provisioning rate was 1.034 provisioning ev ents per hour, a 32% decrease from 2007. More specific data were also collected during summer 2007 and 2008 focal follows. In 2007 and 2008, 240 min and 600 min, respectively, of behavioral data were collected. Beggar was seen begging 67% of the time in 2007 and 69% of the time in 2008. The dolphin was also recorded provision ing 1% of the time in 2007 and 4% of the time in 2008. In 2007, during each follow, BEGR was provisioned three times by two separate boats; a provisioning rate of 1.5 pr ovisioning events/hour. In 2008, BEGR was
37 provisioned three times in the first follow, tw ice in the second follow, once in the third and fourth follow, and four times by two boats in the fifth follow. The provisioning rate declined by 27% in 2008 to 1.1 provisioning events/hour. The Dolphin-Friendly Fishing and View ing Tips cards were well received and were readily accessible to boaters and anglers at numerous locations within the immediate area. It seems from preliminary analyses showing an approximately 30% decline in human interactions with Beggar that the cards may have had an important educational impact and were successful in re ducing dolphin-angler interactions in this particular case study. F106 F106, calf of F191, was a male resident of the Sarasota Bay dolphin community seen 773 times since his birth in 1981. F106 met male pair bond criteria with FB06 since 2005 (Owen et. al 2002). Pair bonds are a comm on strategy among male bottlenose dolphins in Sarasota Bay (Owen et. al 2002; Wells 1991, 2003). FB06 stranded on 12 June 2006 with fishing hooks, line, and a lure in his mouth and throat (Wells et al 2008). The circle hook and lure in the dolphins throat were consider ed to be the cause of death (Wells et al. 2008). FB06 was only seen engaging in an angler interaction on 19 May 2006 when FB06 was noted as stalking a fishing boat. FB06, a Sarasota Bay resident born in 1984 to FB71, was seen a total of 568 times before his death. F106 was first recorded engaging in an angler interaction about two months following FB06s death. On 24 July 2006, F106 was reported as being taunted by anglers at Anna Maria City Pier who were dipping fish hooked on their lines in and out of the
38 water. Since that date, F106 has been r ecorded as engaging or possibly engaging in angler interaction behavior in 33 sightings. Specifically, during pier surv eys and focal follows of 2007, F106 engaged in 35 act s of depredation (including scavenging, attempted depredation, line depredation, a nd most commonly, provisioning) at Anna Maria City Pier. F106 was never observed to entangle in or ingest fishing gear. From visual and acoustic pier obser vations, it became obvious that F106 was using the pier as a foraging ground. Data from a continuously recording HTI-96 hydrophone installed underneath the pier show ed that F106 echolocated during at least four depredation acts. Also, on 19 May 2007, F106 spent 5.58 continuous hours within 200 m of Anna Maria City Pier engaged in at least five depredation acts, and spent 89.9% of this time milling or patrolling. On this day, F106 spent at least 2.85 hours within 15 m of at least one fishing line. Four other groups of dol phins traveled by the pier during the day and F106 never joined or so cialized with any of the passing dolphins. A typical behavioral pattern involved F106 repeatedly circling the pier, about 1020 m from the edge. Before completing a fu ll circuit around the pier, F106 would swim down and along the boardwalk where cast fishermen often fished. If F106 was being provisioned or successfully scavenged in an area, the dolphin would spend time tightly circling within 10-20 m of the depredated area. F106 was last observed on 27 August 2007 patrolling Anna Maria City Pier. At this time, the dolphin was not seen to be entang led or to have lost a significant amount of weight, which can be indicative of fish ing-gear obstructing feeding (Wells et al. 2008). However, due to the sighting frequency of F106 (seen on average of about 29 times per year), and the increasing freque ncy of F106s involvement in angler interaction behavior,
39 I expect that F106 died due to either inges tion or entanglement of fishing gear. Because male pairs often work together coop eratively for foraging purposes (Owen et al. 2002; Wells 1991, 2003), I assume that F106 engaged in similar depredation type foraging strategies as FB06, therefore increasing the probability that F106 met a similar outcome to FB06s. Moreover, in Sarasota Bay, inge stion of fishing gear, in all 12 documented cases was found to be fatal, and severe constrictive entanglement s around the flipper or fluke insertions can cautiously be cons idered to lead to mortality (Wells et al. 2008). 2.4 Discussion Overall, the most concerning result from this project was the recent and continuing increase in depredation behavior in the Sarasota Bay dolphin community. Dolphin-angler interactions rose both in frequency and in number of dolphins. The 10year projection for the number of dolphin-recreational angler interactions in Sarasota Bay (based on the angler-interaction rate from 2003-2007 monthly photo-identification surveys) indicates that if unchecked by 2017, 84% of sightings will involve depredation or a related behavior. Furthermore based on the average rise in the number of resident dolphins depredating from 2004-2007, the conservative estimate suggests that in 2017, 29% of animals in the Sarasota Bay popul ation will be seen engaging in anglerinteraction behavior.
40 Behavioral transmission of angl er interaction behaviors Dolphins sometimes cooperatively forage with associates that share similar feeding strategies (Nowacek 1999, Mann and Sargeant 2003), so associates could serve as vectors for transfer of the depredation behavior (Whitehead et al. 2004). Behavioral transmission of long-line depredation has b een hypothesized for a number of species in the South Pacific including popul ations of killer whales ( Orcinus orca ) and sperm whales ( Physeter macrocephalus ) (Donoghue et al. 2002). One major avenue for behavioral transmission and social lear ning that has been well establ ished in the Sarasota Bay bottlenose dolphin community is between mothers and calves (Nowacek 2002, Wells 2003). Therefore, if depredation is a fo raging strategy of the female, then it is conceivable that this behavior would be passed down to offs pring especially as the calf learns to forage (Nowacek 2002, Wells 2003) For example, calves in Shark Bay, Australia hunted with foraging techniques exclusive to their mothers repertoire including begging from boats (Mann and Sargeant 2003). Anecdotally, maternal transmission of angler-interaction behaviors has been documented in three separate lineages in Sara sota Bay. The best example of maternal transmission of depredation involves the FB79 maternal lineage. FB79, a depredator and focal animal, was the first dolphin to ever be documented as patr olling a recreational fishing boat and has been seen on four separate occasions engaging in this behavior. FB79s daughter, F109, has also been docum ented patrolling fishing boats and scavenging on anglers thrown-back fish. Furt hermore, F109s two-year-old female calf (1091) was confirmed scavenging a released fish when patrolling a fishing boat in a local pass with her mother in February 2008. Th e depredation event by 1091 marked the third
41 generation of depredators from the FB79 lineag e. Another juvenile male focal animal, F232, that has been confirmed engaging in angler interaction behavior on multiple occasions was the calf of confirmed female depredator, FB75. FB75 died in 2006 from a lure lodged in her throat. Cunningham-Smith et al (2006) reported on the case of 4-yrold male BRD2 who was observed begging w ithin the home range of BEGR several weeks before he died. His mother, BRDO, who frequented BEGRs range, was observed begging on occasion, including when she was 11 months pregnant with BRD2. The 3.5-year-old female (Ginger), daught er of control focal F127, stranded in December 2008 and was brought to Mote Ma rine Laboratorys Dolphin and Whale Hospital for rehabilitation. Neither focal F 127 nor calf Ginger had ever been observed engaging in angler interacti on behavior. During her time in rehabilitation, Ginger refused to eat dead or struggling fish and would only eat live, swimming pinfish, mullet, ladyfish, and other local prey. This is furt her evidence that Ginger was not acclimated to dead or incapacitated prey and did not le arn depredation as a foraging strategy from her mother, F127. Overall, the combination of evidence for the transmission of depredation through maternal lineages is st rongly supported by anecdotal observations. Because adult males were found to engage in depredation at a disproportionably high rate, this age-sex group may be another source for teaching the depredation behavior or more likely to incorporate the behavior in to their foraging repertoire once learned. Most depredation behaviors appear to be favor solitary foragers, therefore making this non-natural foraging strategy ideal for unpaired males (Owen et al 2002). In fact, all focal adult males seen depredating during 2007-2008 (BEGR, C354, FB78, F106) were
42 unpaired at the time. Also, in Western Au stralia, males were the most common group conditioned to human interacti ons, especially feeding (Finn et al. 2008). However, results from this study do not strongly support the id ea that depredator dolphins associate significantly more often with other angler interaction animals. Results from COAs and statistical analyses found no significant differences between control and depredator focals associati on patterns with othe r known angler inte raction dolphins. Yet, because the nature of depredation foraging lends itself to a solitary strategy, it may be difficult to document transmission or soci al learning of the behavior. For example, overall group size was significantly smaller for depredator focals when compared to controls and group size was signi ficantly reduced when focals were depredating. It is possible that learning of angler interaction behavior take s place on only a few occasions prior to execution of the be havior by the animal. A soci al network analysis during a dolphins different life stages would prove useful in understanding and pinpointing the transmission of angler interaction behaviors among the Sarasota Bay dolphin community. Dolphin depredation as a result of prey resource depletion and increased anglers Results indicate that the gr eatest rates of dolphin-angler interactions occur when there is prey depletion and increased numbers of anglers on the water. Other areas experiencing similar issues regard the depredation problem as a consequence of direct competition for resources between anglers a nd dolphins (Peddemors 2001, Read 2008). The greatest rates of dolphin-an gler interactions in Sarasota Bay were in March, which corresponds with the height of the tourist and seasonal resi dent period in the Sarasota county area. Boat fuel sales data from Ca nnons Marina show a substantial increase in
43 sales between February and Ma rch. Although the overall yearly rates of hotel/motel and boat fuel did not correlate with yearly dolphinangler interaction rate s, it may be possible that in this particular month when depredat ion rates, hotel/motel occupancy, and boat fuel sales all rise, there is an increase in the num ber of boaters and anglers on Sarasota waters therefore increasing the probability of a dolphin-angler interaction. March is also part of Sarasotas winter season when dolphins spend more time foraging in passes and coastal Gulf habita ts than in seagra ss meadows (Wells 1991, 2003). Prey in Gulf waters are not evenly dispersed, making fish potentially more difficult to locate possibly forci ng dolphins to spend more time searching for prey (Wells et al 1980, Wells 1991, 2003). Passes, corridors to Gulf waters, concentrate anglers to the ideal fishing conditions; the combinati on of deep water and a strong current. Anecdotal reports from local angl ers suggest that March is a time when fish is difficult to find and catch, but March 2008 was found to have the greatest mean fishing effort and dolphin presence at surveyed fishing piers. The combination of these factors increases the probability for dolphin-angler interactions. The need for energy due to lack of prey and diminished energy stores can also explain the significant peak in angler interactions and angl er-interactions per unit fuel sales following a red tide bloom. During a bl oom, dolphin prey can be severely depleted as the result of fish kills from brevetoxin (Gannon et al. 2009). Furthermore, dolphins poor body condition from lack of prey may be exacerbated from the toxic effects of brevetoxin (Bossart et al. 1998, Bossart 2006, Wells et al. 2006, Fire et al. 2008). As a result, this study suggests that dolphins may turn to anglers and boaters as a potential source of food. Significantly greater rates of dolphin-angler inte ractions take place
44 during the lag months following a K. brevis bloom. During a la g period, anglers are expected to begin to return to the water for recreation; no longer avoiding aerosolized brevetoxins that cause respir atory irritation (Kirkpatrick et al. 2004). Although anglers are again fishing in the bay and coastal waters, fish populations are still low and will take some months to recover (Gannon et al. 2009). Therefore, it is reasonable to assume that dolphins might depredate bait or catch to sustain them during this period of environmental recovery. A continua tion of this study conducted during a K. brevis bloom and the following three month lag peri od would be useful in determining finescale depredation effects as a direct result of red tide. Depredation as part of the foraging repertoire If depredation behaviors are based exclus ively on prey resource availability, then it would expected that once resources returned to sustainable leve ls, dolphins would no longer engage in angler interaction behavior and again forage naturally. However, data from this study show that this is not necessa rily the case. Conservative estimates show that dolphin-angler interactions have b een on the rise since 2003. Year 2007 was the highest on record for dolphin-a ngler interactions despite th e lack of a major red tide bloom. Thus prey depletion may trigger spik es in the behavior, but is not the single factor explaining dolphin depredation. Activity budgets compared between contro l and depredator focal dolphins showed that control animals spent significantly more time foraging naturally while depredator dolphins allotted more time to other activities such as milling and depredation behaviors. A previous study in Sarasota Bay by Waples (1995) showed summer dolphin activity
45 budgets and control dolphins had the same order of activities (traveling, milling, foraging, socializing, and resti ng) and similar values for behaviors. When Waples (1995) data are compared to de predator focal dolphins, the or der of activities varies due to the inclusion of angler interaction behaviors and the differences between activity budget categories were much greater. These data allow for strong conclusions about the behavior of depredating dolphi ns since these dolphin were found to differ not only from control counterparts but also from a larger study dedicate d to understanding the natural activity patterns of bottlenose dolphins in Sa rasota Bay. Depredating focal dolphins are modifying their daily activities to incorporat e depredation behaviors. Natural activities such as natural foraging and traveling are occurring at a lower frequency while other behaviors such as milling and depredation are occurring at a much greater rate. Depredator and control focal activity budgets were still ma rginally to significantly different in traveling and milling activity when non-natural, depredation foraging strategies were collapsed togeth er into a single foraging category. This is evidence that depredation for some dolphins has become part of the foraging behavior repertoire. Depredator animals have rescheduled their na tural activity budgets to adjust for angler interaction behaviors, now investing more time into behaviors that are energetically conducive to foraging via depredation. Ho wever, depredation behaviors are not necessarily energetically less costly than natu ral foraging behaviors. If depredation was in fact a low cost foraging strategy, more dol phins would be expected to frequent fishing piers on days and times when more anglers were present. However, statistical analysis showed that none of the measures including th e total number of anglers, number of baited lines, and number of lines with lures were significant predictors of dolphin presence.
46 This is further supported by F106 echolocating during depredation acts. Echolocation is a potentially costly sensory activity, in term s of both energetics and ecology, and it is thought that dolphins use echoloc ation sparingly (Nowacek 1999, Gannon et al. 2005). The effects of depredation behav ior on habitat and home range Although initial analysis proved that depr edators did not spen d significantly more or less time in different habitats or have sma ller home ranges and core areas, I think it is necessary to consider more habitat features before concluding that dolphin habitat selection or home range in not affected by the incorporation of depredation. From this study, depredation by focal animals was only witn essed in three habita ts: channel, pass, and open bay (around a fishing pier). A measure of distance to various habitats from focal follow waypoints would allow me to determine habitat selection for focals (Allen et al. 2001). For instance, I would expect that depr edator focals are more likely to be in habitats near passes, channels, and fishing piers and incorporate them into their home range since there would be ample opportunitie s to depredate. Sm aller scale habitat selection is already supporte d by the results that depredat or focals were significantly more likely to be within 50 m of an active fishing line than were control focals. A further habitat analysis is especially necessary for management purposes since varying habitat usage due to depredation could potentially hold increased predation and mortality consequences outside possible injury or death caused by recreational fishing gear (Wells et al. 2008). Natural dolphin foraging behavior in Shark Bay, Australia shows that dolphins foraged less than expected in habitat with the greatest levels of prey resources, but with increased risk of tiger shark encounters (Heithaus and Dill 2002). On
47 the west coast of Florida, sharks are more co mmon in deeper habitats such as passes, an area which also concentrates anglers, than in shallow waters such as seagrass meadows (Wells et al 1980). If dolphins are spending more tim e in passes or just nearby seeking depredation foraging opportunities, probabilities for shark attacks may increase for these dolphins having further negative impacts on the Sarasota dolphin community. A complement to this analysis would to de termine and compare survivability rates for depredating and control animals. Public awareness and education: the key to reducing dolphin depredation Educating the local and visiting public, specifically anglers and boaters, is a valuable means of reducing depredation. Th ese data have shown that provisioning and scavenging combined were the most common forms of depredation accounting for 79% of all depredation within th e Sarasota Bay study area. These two behaviors could be mostly extinguished in the Sarasota dolphin community if the human component involved in these forms of depredation was c ontrolled. Yet, the majority of angler and boater reactions to dolphin depredation events were not id eal, with 73% of reactions either showing no change in behavior or encouragement of the dolphin to continue interacting. One mechanism of change for human be havior is education and outreach. The Dolphin-Friendly Fishing and Viewing Tips cards have been designed to target both anglers and boaters. The first five tips lis ted on the card are necessary for extinguishing the majority of depredation types, sp ecifically the most common provisioning and scavenging, in the dolphin population. Th ese five tips include federal laws and
48 guidelines like never feed wild dolphins, reuse or share leftover bait reel in your line if dolphins appear, change locations if dolphins show interest in bait or catch, and release catch quietly away from dolphins. If anglers and boaters heeded this information, all types of depredation behaviors would declin e since awareness of how to appropriately react to dolphin-angler interac tion would be heightened. Making depredation difficult or less of an opportunity for dolphins would reduce depredation and stop the incr easing trend of the number of animals in the population depredating, as the behavior would no longer be reinforced with a fish reward. The fewer animals in the population utilizing depreda tion as a foraging strategy or given the opportunity to learn the behavior, the lesser th e chance that depreda tion will continue to spread through the population vi a behavioral transmission. Education and outreach have the ability to reach a large portion of the target audience in a very positive manner. By satu rating target sites fre quented by visiting and local anglers and boaters of a ll experience levels, the probabi lity of the target audience becoming aware and informed about the i ssue of dolphin depredation prior to an encounter greatly increases. Overall, the Dolphin-Friendly Fishing and Viewing Tips cards have been well-received by the vast majority of businesses and organizations and there has been an overwhelming response in requests for more cards. The apparent decline in interactions with Beggar since card distribution in his home range is quite encouraging, and worthy of continued monitoring.
49 Table 2.1 Working definitions of activity ca tegories as defined for the Sarasota Bay dolphin community. Definitions taken from SDRP Manual for Field Research and Laboratory Activities (2006). Activity Definition Mill Non-directional movement, and often occurs in conjunction of other activities. Feed Recorded whenever a dolphin is observed with a fish in its mouth. Probable Feed Recorded when there are indications of feeding, but the feeding cannot be confirmed (e.g., active milling by a dolphin with frigate birds diving on it). Travel Directed movement, including zig-zag movement. Play Involves interactions with objects other than dolphins (e.g., throwing a stingray repeatedly) Rest Involves slow, quiescent activity in the absence of other identifiable activities. Leap, Tailslap, Chuff Includes individual aeria l or acrobatic behaviors of any kind. Social Includes all active interactions wi th other dolphins, including contact, chasing/following, sexua l interactions, etc. With Boat Includes all cases where the dolphins are interac ting with a boat, including bow riding, stern wake ri ding, making figure-eights ahead of the boat, etc. This could be cons idered a sub-category of play, but it should be recorded separately in addition to play. Other A catch-all category to accomm odate the dolphins behavioral flexibility. The behavior should be described in the comments section of the sighting data sheet.
50 Table 2.2 Depredation and related worki ng definitions created to quantify dolphin behavior when interacting with angler s, fishing vessels, or fishing piers. Activity Definition Patrol Dolphin is traveling in repeated directions along fish ing lines, fishing boats or pier edge or when a dolphin continues to mill after multiple surfacings near fishing boats, fishin g lines, or pier. Dolphin must be within at least 15 m of boats, lines or pier. Beg Dolphin is behaving in way to elicit food from a person such as bringing head out of the water an d/or opening mouth at surface. Scavenge When a dolphin is observed feedi ng on an anglers bait or catch that was thrown back into the water (not on an anglers line). The intent of the angler was not to feed the dolphin but rather to throw back unwanted bait or catch. Line Depredation When a dolphin successfully takes and feeds on the bait or catch from an anglers line. Even if the dolph in only takes part of the fish from the anglers line, this is still considered a successful depredation event. Attempted Depredation When dolphin attempts to take bait or catch off an anglers line but is unsuccessful or aborts the behavior before taking bait or catch (e.g., dolphin chases line with catch but line is removed from the water before dolphin takes catch). This cate gory is also used when it is not possible to determine the success of the depredation attempt. Provision Dolphin intentionally being fed bait, catch, or other items by individual(s). Person( s) may be directly dr opping item in dolphins mouth or throwing item at dolphin.
Table 2.3 Location information on the six piers and jetties within th e home range of the Sarasota dolphin community. Also included is the duration of effort and total hours of survey monitoring for each pier or jetty. Pier Latitude, Longitude Location Description Duration of effort Total hours of effort (h) Anna Maria Historic City Pier 27.5345, -82.7305 North end of Anna Maria Island inside Tampa Bay May-July 2007; October 2007April 2008 169.75 Rod and Reel Pier 27.5383, -82.7393 North end of Anna Maria Island at the mouth of Tampa Bay May-June 2007; October 2007April 2008 109.78 Bradenton Beach City Pier 27.4668, -82.6939 South east side of Anna Maria Island just south of the Cortez bridge October 2007April 2008 50.51 Venice Pier 27.0725, -82.4529 In the Gulf of Mexico south of Venice inlet July 2007 8.19 North Venice Jetty 27.1129, -82.4697 Rock jetty bordering the North edge of Venice inlet May-June 2007 18.57 South Venice Jetty 27.0725, -82.4529 Rock jetty bordering the South edge of Venice inlet May-June 2007 18.08 51
Table 2.4 Depredator and control focal dolph ins selected for behavioral follows in summers of 2007 and 2008. Depredator dolphins had been observed engaging in angler interaction behavior or were entangled in monofilament on more than one occasion. Paired control animals have never been obser ved interacting with anglers and were also chosen for being the same sex, and having a similar age (+/-5 years) and range to the focal depredator. Mothers w ith * engaged in angler in teractions behavior. (BEGRs birth year is a minimum based on date for first observation). Focal type Focal dolphins Sex Birth year Mother Summer 2007 follow time (min) Summer 2008 follow time (min) Depredator BEGR M <1990 Unknown 240 600 Control F110 M 1984 Unknown 240 360 Depredator C354 M 1992 FB35 240 360 Control C834 M 1992 FB83 240 360 Depredator FB78 M 1972 Unknown 240 360 Control FB36 M 1972 Unknown 120 246 Depredator FB79 F 1979 Unknown 240 360 Control FB65 F 1983 FB67 240 360 Depredator F106 M 1981 F191 240 X Control FB10 M 1981 FB63* 120 X Depredator F109 F 1995 FB79* 240 360 Control F127 F 1995 FB13 240 360 Depredator F222 M 1998 Unknown 234 240 Control F196 M 1998 F101 114 363 Depredator F232 M 2002 FB75* 240 363 Control F224 M 2002 FB27 240 360 52
Table 2.5 Dolphin-angler interaction rates (including all behaviors such as patrol, beg, scavenge attempted depredation, line depredation, and provision) in Sarasota Bay, Florida for 2000-2007. Rates were mo deled from two different data sets: monthly photo-identification surveys and all sighting effort. Interaction rates were also an alyzed including and excluding the male BEGR, a well-known beggar. Interaction rates were calcul ated by dividing the number of interactions by the number of boat-days for a given year. Monthly Photo-identification Estimates Year No. of boat-days No. of interactions Interaction rate No. of interactions (BEGR excluded) Interaction rate (BEGR excluded) 2000 95 7 0.074 0 0.000 2001 92 4 0.043 0 0.000 2002 107 12 0.112 0 0.000 2003 103 11 0.107 0 0.000 2004 102 13 0.127 1 0.010 2005 98 16 0.163 3 0.031 2006 107 22 0.206 10 0.093 2007 117 37 0.316 21 0.179 All Sighting Effort Year No. of boat-days No. of interactions Interaction rate No. of interactions (BEGR excluded) Interaction rate (BEGR excluded) 2000 357 7 0.020 0 0.000 2001 377 43 0.114 1 0.003 2002 250 15 0.060 1 0.004 2003 262 11 0.042 0 0.000 2004 188 14 0.074 1 0.005 2005 234 16 0.068 3 0.013 2006 231 38 0.165 16 0.069 2007 375 92 0.245 67 0.179 53
Table 2.6 Depredator dolphins in Sarasota Bay and their percent of the total population from 2000-2007 calculated from monthly photo-identification surv eys and all sighting efforts. Estimates for the number of depredators to date include cumulative totals for known depredators in the po pulation from past years until the present year. Monthly Photo-identification Surveys Year Sarasota dolphin population Depredators (confirmed) Depredators (confirmed): % of pop. Depredators (confirmed + possible) Depredators (confirmed + possible): % of pop. Depredators to date (confirmed) Depredators to date (confirmed): % of pop. Depredators to date (confirmed + possible) Depredators to date (confirmed + possible): % of pop. 2000 175 3 1.71% 3 1.71% 3 1.71% 3 1.71% 2001 177 1 0.57% 1 0.57% 2 1.13% 2 1.13% 2002 178 1 0.56% 1 0.56% 2 1.12% 2 1.12% 2003 174 1 0.58% 1 0.58% 2 1.15% 2 1.15% 2004 175 1 0.57% 8 4.57% 2 1.14% 9 5.14% 2005 171 5 2.92% 5 2.92% 5 2.92% 12 7.02% 2006 170 11 6.47% 20 11.77% 12 7.06% 26 15.29% 2007 166 16 9.64% 22 13.25% 21 12.65% 36 21.69% All Sighting Effort Year Sarasota dolphin population Depredators (confirmed) Depredators (confirmed): % of pop. Depredators (confirmed + possible) Depredators (confirmed + possible): % of pop. Depredators to date (confirmed) Depredators to date (confirmed): % of pop. Depredators to date (confirmed + possible) Depredators to date (confirmed + possible): % of pop. 2000 175 3 1.71% 3 1.71% 5 2.86% 5 2.86% 2001 177 4 2.26% 4 2.26% 6 3.39% 6 3.39% 2002 178 1 0.56% 3 1.69% 5 2.81% 6 3.37% 2003 174 3 1.72% 3 1.72% 5 2.87% 6 3.45% 2004 175 1 0.57% 9 5.14% 4 2.29% 13 7.43% 2005 171 5 2.92% 5 2.92% 7 4.09% 15 8.77% 2006 170 17 10.00% 26 15.29% 20 11.76% 34 20.00% 2007 166 18 10.84% 29 17.47% 24 14.46% 45 27.11% 54
55 Table 2.7 Age class and sex di stribution of documented confirmed depredati ng dolphins in Sarasota Bay (2000-2007). Percentages are based on the tota l number of depredating animals. The age of the animal at the time of the first angler interaction incident was used for the analysis. Age Class Total Males Females Unknown Percent Total Percent Male Percent Female Percent Unknown Calves 3 1 1 1 6.98 2.33 2.33 2.33 Juveniles 14 8 5 1 32.56 18.60 11.63 2.33 Adults 18 14 4 0 41.86 32.56 9.30 0.00 Unknown 8 4 3 1 18.60 9.30 6.98 2.33 Totals 43 27 13 3 62.79 30.23 6.98
Figure 2.1 Area surveyed monthly for Saraso ta dolphin community members. Sarasota Bay and surrounding waters are located on the central west coast of Florida. Piers monitored are marked with red stars. 56
0 5 10 15 20 25 30 35 10-2526-5051-75 76-100101-125126-150151-175 176-200201+ Distance bin (m)Mean error (m) Figure 2.2 Mean error for estimating distance (n=2389) for the primary investigator (J.R. Powell). The line graph shows the average e rror of investigator estimates for actual distances included in the distance bin. 57
Figure 2.3 Aerial view of Anna Maria City Pier from 2008 Google satellite maps. The red outline encompasses the 200 m radius of ar ea that was visually monitored during pier surveys. The black diamonds represent areas on the pier where the three field workers would monitor from. The radi us and the field worker obser vation positions are similar for all pier and jetty survey sites. 58
Focal Depredator Focal Control Figure 2.4 The 95% kernel home range (stripes ) and 50% kernel core area (dots) for focal dolphins from January 2004 through A ugust 2008. Focal depredators are plotted on the left and focal controls on the right. 59
60 Focal Depredator Focal Control Figure 2.4 (Continued).
61 Focal Depredator Focal Control Figure 2.4 (Continued).
Figure 2.5 The first edition of the D olphin-Friendly Fishing & Viewing Tips educational card. This version includes a single image on the front cover and more detailed text than the second version. 62
Figure 2.6 The second edition of the D olphin-Friendly Fishing & Viewing Tips educational card. This version includes multiple images on the front cover as well as more condensed text. This version was also translated into Spanish. 63
0 10000 20000 30000 40000 50000 60000 70000Other Fishing Events & Tournaments USCG/USCGA NOAA/NMFS Fish Florida/Sea Grant Fishing/tackle/boat rental/waterfront restaurants Florida State and Local Governments Conservation/Education (Aquariums, Univ., etc)Type of organizationNo. of cards distributed Figure 2.7 Dolphin-Friendly Fishing and Viewing Tips card distribution from January to October 2008. 64
0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 20002001200220032004200520062007 YearDolphin-angler interaction rate All BEGR excluded Figure 2.8 Dolphin-angler in teraction rates for the Sarasota Bay dolphin community from 2000-2007 based on monthly photo-identifi cation survey data. Interaction rates were analyzed including and excluding th e male BEGR, a well-know n beggar since 1990. Interaction rates were calculated by dividing the number of angler interactions by the number of boat-days for a given year. 65
0.000 0.050 0.100 0.150 0.200 0.250 0.300 20002001200220032004200520062007 YearDolphin-angler interaction rate All BEGR excluded Figure 2.9 Dolphin-angler in teraction rates for the Sarasota Bay dolphin community from 2000-2007 based on all sighting effort. In teraction rates were analyzed including and excluding the male BEGR, a well-known beggar. Interac tion rates were calculated by dividing the number of dolphinangler interactions by the number of boat-days for a given year. 66
0% 2% 4% 6% 8% 10% 12% 14% 20002001200220032004200520062007 YearPercent population Conf. Depredators Conf. + Poss. Depredators Figure 2.10 Data collected from monthl y photo-identification surveys showing the percent of dolphins in the Sarasota Bay population engaging in angler interaction behavior from 2000-2007. Data sets plotted include confirmed incidents and confirmed plus possible incidents. 67
0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 20002001200220032004200520062007 YearPercent population Conf. Depredators Conf. + Poss. Depredators Figure 2.11 Data collected from all sighting effort showing the percent of animals in the Sarasota Bay population engaging in angler interaction behavior from 2000-2007. Data sets plotted include confirmed incidents and confirmed plus possible incidents. 68
0% 5% 10% 15% 20% 25% 20002001200220032004200520062007 YearPercent population Conf. Depredators Conf. + Poss. Depredators Figure 2.12 Data collected from monthly photoidentification surveys showing the cumulative percentage of dolphins in the Sarasota Bay population from 2000 through the specified year that have been observed engaging in angler inte raction behavior. Data sets plotted include confirmed incidents and confirmed plus possible incidents. 69
0% 5% 10% 15% 20% 25% 30% 20002001200220032004200520062007 YearPercent population Conf. Depredators Conf. + Poss. Depredators Figure 2.13 Data collected from all sighting effort showing the cumula tive percentage of dolphins in the Sarasota Bay population from 2000 through the specified year that have been observed engaging in angler interaction behavior. Data se ts plotted include confirmed incidents and confir med plus possible incidents. 70
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4J a n u ary February March April M a y J un e J ul y Augu s t Se ptember Oct o be r Nov e m b e r De c e m b e rMonthDolphin-angler interaction rate Avg surveys Avg all effort Figure 2.14 Monthly angler in teraction rates for the Sara sota Bay dolphin community from 2000-2007 based on monthly photo-identifi cation surveys (Avg surveys) and all sighting effort (Avg all effort). Intera ction rates were calculated by dividing the number of dolphin-angler interactions by th e number of boat-days for a given month and then plotting the average for all months across all years. 71
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%Ja n u a ry Februa r y March April M a y June Ju l y Au g u st S e p t emb e r Oc t o b e r No v em b e r DecemberMonthPercent occupancy0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Avg. hotel occ. Avg. AI (surveys) Avg. AI (all effort)Dolphin-angler interaction rate Figure 2.15 Comparison of dol phin-angler interaction ra tes (based on monthly photoidentification surveys and all sighting effo rt) and average monthly Sarasota County hotel/motel occupancy rates from 2000-2007. 72
0 2000 4000 6000 8000 10000 12000January Feb ru ary M a rch A pril M ay June Jul y A ugust Sep te mber O ctob er November D ecem b erMonthFuel sales0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Avg. fuel sales Avg. AI (surveys) Avg. AI (all effort)Dolphin-angler interaction rate Figure 2.16 Comparison of angler intera ction rates (based on monthly photoidentification surveys and all sighting effort) and boat fuel sales from Cannons Marina 2003-2007. 73
0 0.1 0.2 0.3 0.4 0.5 0.61 Red Tide ConditionAngler interaction rate No bloom Bloom Lag Figure 2.17 Average of dolphin-a ngler interaction ra tes in Sarasota Bay, Florida over three environmental periods based on Florida red tides (non-bloom, K. brevis bloom, and lag time following a bloom) from 2000-2007. Differences between HAB periods were significant ( p=0.031). 74
0% 5% 10% 15% 20% 25% 30% 35% 40% ChannelGulfMangroveSandSeagrassOpen bayPass HabitatPercent time Control Depredator Figure 2.18 Habitat use by focal dolphins durin g focal animal behavi oral follows from summer 2007 and 2008. (n=16: 8 control focals, 8 depredator focals) 75
0 50 100 150 200 250 300 JanFeb MarAprMayJunJulAugSepOctNovDecMonthMean no. line s 0 1 2 3 4 5 6 7 Average Total lines Average No. DolphinsMean no. of dolphins Figure 2.19 The mean presence of dolphins and the mean number of fishing lines (rigged with either bait or lures) by month during pi er surveys from May-July 2007, and October 2007-April 2008. 76
0 2 4 6 8 10 12 6:007:008:009:0010:0011:0012:0013:0014:0015:0016:0017:0018:00 TimeMean no. lines0 0.2 0.4 0.6 0.8 1 1.2 Fishing effort Dolphin presenceMean no. of dolphins Figure 2.20 The mean presence of dolphins a nd mean number of fi shing lines (rigged with either bait or lures) by hour during pi er surveys from May-July 2007, and October 2007-April 2008. Times were binned into one hour increments. 77
Controls Forage 20% Mill 19% Travel 58% Rest 0% Social 3% Beg: 0% Patrol: 0% Provision: 0% Depredators Beg 9% Forage 9% Mill 33% Travel 38% Provision 1% Patrol 6% Social 4% Rest: 0% Figure 2.21 Overall mean activity budgets for control (n=8) and depr edator (n=8) focal dolphins. Data compiled from summer 2007 and 2008 focal animal behavioral follows. Depredator focals spent signi ficantly less time foraging ( p=0.053) and traveling ( p=0.02) and more time engaging in milling behavior (p =0.054). 78
Controls Forage 20% Mill 19% Travel 58% Social 3% Rest 0.3% DepredatorsForage 23% Mill 33% Social 4% Travel 40% Rest: 0% Figure 2.22 Overall mean activity budgets for control (n=8) and depr edator (n=8) focal dolphins. Forage is a combined categor y including natural and non-natural foraging strategies (feed, probable feed, beg, patrol, and provision). Data compiled from summer 2007 and 2008 focal animal behavioral follows. When the forage category is compiled, depredator focals still spent significantly less time traveling ( p=0.015) and more time engaging in milling behavior ( p=0.0557). 79
C834 Forage 11% Mill 24% Travel 64% Social 1% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% C354Forage 4% Mill 42% Travel 49% Social 5% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% F110Forage 21% Mill 12% Travel 67% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% Social: 0% BEGR Beg 69% Travel 12% Provision 3% Mill 16% Forage 0.2% Patrol: 0% Rest: 0% Social: 0% F127Forage 14% Mill 29% Travel 57% Social 0.4% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% F109 Forage 9% Mill 52% Travel 30% Patrol 4% Social 5% Beg: 0% Provision: 0% Rest: 0% Figure 2.23 Individual mean ac tivity budgets for control (left) and depredator (right) focal dolphins. Activity budgets are paired by depredator and their corresponding control counterpart. Data compiled from summe r 2007 and 2008 focal animal behavioral follows. 80
F196 Travel 60% Forage 13% Mill 7% Rest 2% Social 18% Beg: 0% Patrol: 0% Provision: 0% F222Forage 5% Mill 38% Travel 47% Social 10% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% F224Forage 11% Mill 63% Travel 26% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% Social: 0% F232 Forage 22% Mill 34% Travel 34% Patrol 1% Social 9% Beg: 0% Provision: 0% Rest: 0% FB10Forage 66% Travel 34% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% Social: 0% Mill: 0% F106 Forage 9% Mill 15% Patrol 41% Travel 31% Social 1% Provision 3% Beg: 0% Rest: 0% Figure 2.23 (Continued). 81
FB36 Travel 81% Mill 2% Forage 17% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% Social: 0% FB78 Forage 8% Mill 37% Travel 55% Social 0.4% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% FB65 Mill 16% Travel 80% Rest 0.2% Social 2% Forage 2% Beg: 0% Patrol: 0% Provision: 0% FB79Mill 31% Travel 58% Social 1% Forage 10% Beg: 0% Patrol: 0% Provision: 0% Rest: 0% Figure 2.23 (Continued). 82
Provision 64% Scavenge 17% Attempted depredation 14% Line depredation 5% Figure 2.24 Comparison of 58 depredation events documented during pier surveys and focal animal behavioral follows (2007-2008). Events are organized by depredation type. 83
Provision 58% Scavenge 21% Attempted depredation 14% Line depredation 7% Figure 2.25 Comparison of 70 total depredati on events documented during pier surveys, focal animal behavioral follows, monthly population and photo-identification surveys, and various other research initiatives (20072008). Events are organized by depredation type. 84
Continue feeding/ throwback 50% Cast at animal 2% Taunt 7% Pull line out 14% Move locations 4% No change 19% Summon dolphin 4% Figure 2.26 Angler and boater reactions following a dolphin depredation event documented over 58 incidents during pier surveys and focal animal behavioral follows (2007-2008). 85
0 1 2 3 4 5 6 7 BEGRC354F106F109F222F232FB78FB79 Focal DepredatorGroup size Overall During angler interaction Figure 2.27 Overall depredator focal dolphin group size compared to mean group size when focal was depredating or engaging in de predation related behavi or. Data analyzed from the Sarasota Dolphin Research Program 2000-2007. 86
Figure 2.28 Sequence of photos (in order from top to bottom) illustrating the order of behaviors that BEGR engages in when beggi ng from boaters. The top photo illustrates the animal surfacing perpendicular to the vesse l to first elicit a ttention. Secondly, the animal parallels the vessel, rolling to one si de. Beggar then spyhops out of the water in the classic begging pose if fed or taunted. Fi nally, the animal pursues the boat within a few cm of the propeller. P hotos are from different inci dents in summer 2008 but were chosen for best illustrating the described behavior. 87
88 3. F201 Case Study: Exploring the Potent ial Link between Hearing Impairment and Depredation 3.1 Introduction Optimal foraging theory predicts that animals will forage in a way to maximize energetic benefits so to in crease their fitness (Emlen 1966, MacArthur and Pianka 1966). Exemplifying this theory is the depredation behavior in which a predator steals or damages a prey item already captured by so me other process (Zo llett and Read 2006). For hearing-impaired dolphins which are poten tially less successful at natural foraging, depredation of injured fish struggling on a hook or released by anglers may yield more calories per energy expended than foraging for live fish. Hearing in odontocetes is considered one of their most valuable senses. Bottlenose dolphins ( Tursiops truncatus ) rely upon the auditory sense for communication with conspecifics and navigation. Hearing is essential for both passive and active foraging strategies of dolphi ns (Jones and Sayigh 2002, Gannon et al. 2005, Nowacek 2005). Studies suggest that dolphins search for soniferous prey items by listening passively (Barros and Wells 1998, Gannon et al. 2005). Utilizing hear ing abilities to
89 listen for prey may offer an energy efficient m echanism for the initial detection of prey as well as the determination of prey number, size, and location (Gannon et al. 2005). Dolphins actively use echolocation fo r foraging, especially in the pursuit and capture phases of the hunt (Au 1993, Gannon et al. 2005). In Sarasota Bay, Florida, dolphin echolocation is more often associat ed with foraging behaviors than with nonforaging behaviors (Jones and Sayigh 2002, No wacek 2005). Furthermore, echolocation is produced more often by solitary foragers in sand and seagrass-edge habitats (Nowacek 2005). Peak frequencies for dolphin echolocation clicks ar e typically between 100 kHz to130 kHz (Au 1993, 2004), within range of the dolphins hearing sensitivities (Houser and Finneran 2006a, 2006b; Johnson 1967, Cook 2006, Finneran and Houser 2006). High frequency hearing impairment would affect the dolphins ability to use echolocation optimally (Au 1993, 2004; Nowacek 2005) whil e low frequency hearing impairment would limit prey detection by passive listening (Gannon et al. 2005) possibly forcing an animal to utilize alternative stra tegies, such as depredation. Depredation of commercial and recr eational fishing gear by cetaceans is a growing problem around the world (Broa dhurst 1998, Secchi and Vaske 1998, Noke and Odell 2002, Somoa 2002, Cox et al. 2003, Lauriano et al. 2004, Brotons et al. 2008, and Sigler et al 2008). Long-line fisheries depredati on by larger odontocetes has recently been recognized as increasing in frequency, geographic extent, and se verity (Read 2008). Depredation behaviors which bring dolphins near or in contact with fishing gear have the potential to seriously injure or kill the animals through entanglement or ingestion (Gorzelany 1998, Wells and Scott 1994; Wells et al. 1998, 2008). In the state of Florida, specifically the Sarasota Bay area located on the central west coast, bottlenose dolphin
90 ( Tursiops truncatus ) deaths from entanglement and inge stion of recreati onal fishing gear are increasing (NOAA 2006 a). Of the total Sarasota Ba y recovered strandings in 2006, approximately 25% were a result of fishing gear interaction, compared to the average 2.9% rate for dolphin deaths attributed to fishing gear for the years of 2000-2005 (NOAA 2006a). Sarasota Bay, Florida and surrounding waters (an area of approximately 125 km2) are home to a community of about 160 resident bottlenose dolphins that have been closely monitored by the Chicago Zoologica l Societys Sarasota Dolphin Research Program for over 38 years (Wells 1991, 2003) (Figure 2.1). Monitoring is conducted through monthly photo-identification surveys, occasional health assessments, and other research initiatives (Wells 1991, 2003). These studies have provided a wealth of data that exists on family lineages, stranding records, age, sex, behavioral history, distribution, social associations, and hearing abili ties (Wells 1991, 2003; Cook 2006). These longterm, multi-faceted data sets allow for de tailed examination of topics, such as depredation, and their relati onship to typically unknown fact ors, like hearing abilities. For management purposes, understanding c ontributing factors to depredation is essential as this behavior is expected to be a persistent and rising problem as humans and cetaceans compete for the same resources (Read 2008). Depredation has not been explored in conjunction with hearing impairment or loss and has the potential to explain a dolphins shift away from natu ral foraging behaviors. In this study, the case of the entangl ed dolphin, F201, was examined. F201 was the calf of F193, a female documented over a number of years in Charlotte Harbor, Florida before moving to Sarasota. F201 a nd F193 inhabited mainly the southern portion
91 of the Sarasota Dolphin Research Program survey range, and were considered recent additions to the resident Sarasota Bay dolphin community. Although F193 has never been documented as engaging in depredati on or angler interac tion behavior, F201s mother is the most common associat e of BEGR, a notorious begging dolphin (Cunningham-Smith et al 2006). F201 was first observed as a young calf with her mother on 7 November 2005 and was last observed with F193 on 18 May 2006. F201 was next seen alone on 12 December 2006, with healed boat propeller wounds and trailing monofilament fishing line from its peduncle, when the survey team monitored the dolphin as it stayed very close to the port stern quarte r of the vessel for about 40 min. F201 was again observed alone on 18 January 2007 and the monofilame nt line trailing from the peduncle had accumulated algae (Figure 3.1). The dolphin was monitored over the following days and a rescue was carried out on 30 January 2007. Upon capture, it was determined that the dolphins wounds from the monofilament entanglement warranted fu rther medical attention and F201 was subsequently admitted to Mote Marine Laboratorys Dolphin Hospital. Later that day, hospital staff reported that F201 vomited some plastic and it was determined that the loops of monofilament around th e peduncle that had cut thro ugh the flesh and were now constricting the bone would have to be removed through surgery (C. Manire, veterinary case synopsis). While recove ring in captivity, auditory e voked potential (AEP) methods were used to measure the hearing abiliti es of F201. The dolphin was then further monitored post-release.
92 3.2 Methods 3.2.1 Auditory Evoked Potentials (AEP) We tested F201s hearing using audito ry evoked potential (AEP) methods which have allowed researchers to determine the range of hearing capabilities for a number of odontocete species through a relatively brief, non-invasive method (Ridgway et al. 1981, Supin et al. 1993, Supin and Popov 1995, Nachtigall et al. 2004, Cook et al. 2006, Nachtigall et al. 2007, Houser et al. 2008) that has been standardized against behavioral audiograms (Szymanski et al. 1999, Yuen et al. 2005, Cook 2006, Finneran and Houser 2006, Houser and Finneran 2006a). For both cap tive and wild bottlenose dolphins, recent studies have focused on measuring hearing thresholds within th e range of 5-150 kHz (Houser and Finneran 2006a, 2006b; Cook 2006, Finneran and Houser 2006). In-air auditory evoked potentials were used to m easure the hearing of F201 on two test days: 12 and 19 March, 2007. The same equipment and methods as us ed by Cook (2006) were used for testing F201. Data were collected using a Tucker-D avis Technologies (TDT) AEP workstation with SigGen and BioSig software on a lapt op computer. AEP reco rding electrodes were 8 mm Ag-AgCl electrodes embe dded in suction cups which we re composed of silicone. Sound was delivered with a jawphone composed of an ITC-1042 transducer embedded in an RTV silicone suction cup. Hearing thresh old measurements collected on F201 used an Envelope Following Response (EFR) procedure in which a 600 Hz amplitude-modulated rate was used to present 14-15 ms tone bursts at the same time as the auditory evoked potential was recorded. Calibrations were conducted underwater using a Reson
93 hydrophone placed 10 cm from the jawphone at 0.5 m from the surface. All signals were digitized at 260 kHz. Detailed methods a nd procedures are desc ribed by Cook (2006). F201s AEP response was further measured while running click stimuli at two rates (600 Hz and 1 kHz). There was significa nt destructive cancel lation with the 1 kHz click rate, so only the 600 Hz click rate was analyzed. F201s hearing was compared to mean threshold responses from 29 free-ranging Sarasota Bay resident females (Cook, 2006). In-air AEP measurements for the 29 females used for comparison were collect ed and analyzed by Mandy Hill Cook (2006) over five health assessments in Sarasota Bay: June 2003, February 2004, June 2004, February 2005, and June 2006. 3.2.2 Post-Release Monitoring F201 was closely monitored via a VHF radio transmitter and sighted 20 times (12.4 hours of behavioral observations) betw een the release date on 28 March 2007 and final observation on 1 May 2007. During post-re lease monitoring, F201s vocalizations were recorded with a HTI-96 hydrophone (s ensitivity -164 dBV/Pa; 2 Hz-37 kHz) and Creative Nomad Jukebox 3 at a sample rate of 48 kHz. I acquired acoustic recordings on four different days and successfully reco rded the dolphin whistling and echolocating. Recordings were only taken when there were no other dolphins within viewing distance. Data were visually inspected using Adobe Audition 2.0.
94 3.3 Results Results indicated that F201 was hearing-impaired at high frequencies. When compared to baseline values for the Sarasota dolphin females, F201 was found to have hearing loss of approximately 50 dB or more at 40 kHz, 80 kHz, and 120 kHz (Cook 2006) (Figures 3.2). Hearing loss was confir med by the results of the second test which showed hearing loss of 25 dB at 40 kHz a nd even weaker responses at 80 and 120 kHz when compared to mean Sarasota female dolphin thresholds. In comparison, F201s hearing thresholds at 20 kHz appeared norma l (Figure 3.2). F201s click threshold was 112 dBpeak. The lowest detected AEP signal was at 112 dBpeak. In all recordings, F201 was found to be either whistling or echolocating (Figure 3.3). On 17 April 2007, the animal was docum ented as scanning (a behavior usually indicative of echolocation) and the recording confirmed that the dolphin was indeed echolocating (Figure 3.3). In the 20 sightings post-release, F201 was recorded as with boat in 45% of sightings. The dolphin was often observed milling near fishing boats and in areas such as small canals, yacht club basins, or boat rental facilities. The dolphin was seen foraging in 25% of sightings, but was only confirmed to successfully capture fish during one sighting. Post release, F201 remained mostly solitary, and wa s only sighted with at least one other dolphin in 15% of sightings. On 13 April 2007, F201 again acquired monofilament, this time around the radio tag. The monofilament was not entangled aroun d the dolphin and was relatively short, less than 40 cm in length. However, on 1 May 2007, F201 acquired more monofilament,
95 this piece trailing about 30 cm behind her fl uke. F201 has not been sighted since 1 May 2007. 3.4 Discussion The results from this study provide preliminary evidence for a link between dolphin associations with angl ers and boats and hearing loss. F201 was severely hearing impaired at higher frequencies and based on her behavior and enta nglement history, it appears the dolphin selected ha bitats with high ri sk of monofilament entanglements and concentrations of anglers and boaters. It is still to be determined if higher h earing thresholds are a cause or result of angler interaction behavior. Dolphins could damage their hearing temporarily or permanently as a result of depredation if th ey select areas with more boats and thus greater anthropogenic noise (Au et al 1999, Erbe 2002, Nachtigall et al 2004). Noise from engines and depth finders may be aff ecting the dolphins hearing. An acoustic model created by Erbe (2002) for killer whales ( Orcinus orca ) predicted that engine noise would cause a temporary threshold shift (TTS) of 5 dB if the animal was within 450 m of a fast moving inflatable boat or 20 m of a slow moving boat for 30-50 minutes. Hearing was expected to return to normal within 24 hours if the killer whale avoided further contact with boats (Erb e 2002). Such avoidance would be nearly impossible for Sarasota Bay dolphins based on findings from a previous study showing that dolphins come within 100 m of a boat every six minutes during daylight hours (Nowacek et al 2001).
96 Furthermore, the model showed that longterm exposure to noise from fast-moving vessels within 1 km or to slow moving vessels within 50 m could cause a permanent threshold shift (PTS) of 25 dB (Erbe 2002). The Erbe (2002) model and Nowacek et al (2001) study suggest that depredating dolphins in Sarasota Bay have the potential to be effected by both TTS and PTS. However, in the case of F201, hearing impairment was most likely not the result of angler interaction behavior but rather the cause. F201 was particularly young to have such a substantial hearing loss, thus sugges ting that hearing loss was probably a genetic predisposition (Houser and Finneran 2006 b). Dolphins, like F201, may resort to depredation and feeding on dead, injured, or provisioned fish because their hearing is already diminished and they are not successful at prey pursuit and capture using passive listening or echolocation. Evidence that F201 was struggling to find or capture prey is potentially supported by the f act that the dolphin vomite d plastic upon admission to the hospital. The effects of F201s hearing loss on effective comm unication range was calculated assuming a cylindric al spreading loss model (10*log(1/distance)) and taking into account the two-way travel of echoloca tion (from dolphin to target; from target to dolphin). A dolphin with a hear ing loss of 50 dB, normally able to echolocate an object 100 m away, would now only be able to detect the same object at 0.3 m distance. In terms of energy cost, depredation is a more ef ficient foraging strategy in that it requires little effort on the part of the dolph in, especially if provisioned. Regardless of whether impaired hearing is a cause or eff ect of depredation, dolphin-angler interactio ns may increase the rate at whic h hearing is impaired. Optimal
97 foraging theory would predict that the benefit and available o pportunities of being provided easy prey must outweigh the cost of foraging opportunities that will be lost as a result of hearing damage to the dolphi n (Emlen 1966, MacArthur and Pianka 1966). However, depending on the dolphins rate of depredation and angler-interactions, it is probable that significant hearing impairment coul d take months to years to actually show noticeable effects to the dol phin, therefore making depred ation a non-optimal strategy with high short-term energetic gains. F201 was an example of how high frequenc y hearing loss can potentially increase a dolphins probability of becoming entangled or involved in angler interactions. F201s high frequency hearing loss allowed only limited echolocation abilities over short distances and communication with conspecifics. It remains unclear the degree to which F201 could decipher received echoes especially over long distances. It may be possible that F201 was able to decrease the peak frequency of echoloca tion clicks in order to shift it within her hearing range as has been demonstrated with captive Tursiops and other odontocete species (Au et al 1985, Moore and Pawloski 1990, Houser et al. 1999). Anecdotally, in visual inspections of some echolocation recordings, it appeared that F201 had a disproportionate amount of click energy focused in the lower frequencies than is typical for bottlenose dolphin echolocation. The suspected relationship between hear ing loss and angler in teraction behavior as exemplified with F201 warrants further i nvestigation as depredation is becoming a more prevalent behavior in Florida dolphi n populations. To determine the relationship between higher thresholds, probability of depredation behavior, and entanglement risk, hearing of dolphins needs to be assessed on a mo re fine scale basis. In an ideal scenario,
98 measuring a dolphins hearing thresholds ever y year to two years while also closely monitoring individual depredat ion behavior would help determine the relationship between hearing impairment and depredation. In Cooks (2006) analyses of hearing in the Sarasota Bay dolphin community, only one female (F195) showed possible signs of hearing loss suggesting that hearing may be so valuable that if lost, animals ma y not survive and are quickly selected out of the population. F195, tested in 2005, died tw o years later. Due to its history of entanglement, premature separation from its mother, and high fre quency hearing loss, F201 is presumed to have followed the same fate as F195. If death quickly follows hearing loss, depredation may be a last reso rt behavior that is employed in order to successfully acquire food when foraging strate gies based on sensory modalities, such as passive listening and echolocat ion, fail. The circumstances surrounding the case of F201 offer preliminary evidence that hearing loss is linked to angler inte raction behavior and death for wild dolphins. This preliminary evidence should prompt scientists and managers to continue to pursue this hypothe sis in order to better understand the possible physiological causes or impacts of dolphin depredation behavior.
Figure 3.1 F201 photographed by the Sarasota Dol phin Research Program on 19 January 2007. The dolphins lower peduncle is entangled in monofilament encrusted with algae. Boat strike scars on the upper pedunc le are also evident in this photo. 99
100 40 50 60 70 80 90 100 110 120 130 140 0102030405060708090100110120130 Male FemaleSPL (dB re 1 Pa) Threshold (dB re 1Pa) Fr equency (kHz) Figure 3.2 Comparison of F201 audito ry evoked potential audiogram to the mean ( SD) audiograms measured for 32 male and 29 fema le free-ranging bottlenose dolphins in Sarasota Bay, Florida (Cook 2006). Frequency (kHz) F201 Females Males
Frequency (Hz) 5000 10000 20000 15000 0 0 0.5 1.0 1.5 2.0 Time (s) Figure 3.3 Spectrogram of F201 echolocation click train followed by a whistle recorded on Ap ril 17, 2007 (post-release). Sample rate was 48 kHz. 101
102 4. Depredation Monitoring using Passive Acoustics 4.1 Introduction Passive acoustic observation is an e xpanding technique for cetacean monitoring. The technique uses an instrument to reco rd sounds from the environment (Mellinger et al 2007). Passive acoustic record ing devices in the forms of mobile hydrophones, fixed recorders, or attachable tags have been used to determine presence, abundance and distribution patterns ( e.g. Stafford et al. 1998, McDonald and Fox 1999, Oswald et al. 2003, Mellinger et al. 2004, Barlow and Taylor 2005, Rankin et al. 2007), characterize a species vocalizations ( e.g Stafford et al. 2001, Nowacek 2005, Parks and Tyack 2005), and reconstruct dives for a number of cetacean species (e.g. Zimmer et al. 2003, Lammers et al. 2006, Tyack et al 2006, Watwood et al. 2006, Stimpert et al. 2007, Aguilar Soto et al 2008). Passive acoustic monitori ng offers advantages over visual surveys by giving researchers the ability to monitor cetacean presence throughout night hours and poor environmental conditions as well as to accurately estimate the number of animals in the near vicinity (McDona ld and Fox 1999, Barlow and Taylor 2005, Mellinger et al. 2007, Rankin et al 2007).
103 Depredation behavior am ong odontocete cetaceans is increasing in frequency around the world (Read 2008). Depredation behavior creates an economic loss for fishermen, and endangers the life of the anim al through risk of gear entanglement, gear ingestion or retaliation by anglers (Wells and Scott 1994, Gorzelany 1998, Read 2008, Wells et al. 1998, 2008). The ability to monitor depr edation behavior in a cost-effective manner is essential for researchers and mana gement agencies. Traditional behavioral observation methods are costly in terms of time, manpower and general boat costs ( i.e fuel, maintenance) when compared to passive acoustic techniques. Studies comparing visual and acoustic e ffort typically found that acoustic surveys were more successful at detecting or determining numbers of cetaceans present (McDonald and Fox 1999, Mellinger et al. 2004, Barlow and Taylor 2005) and could be used to accurately identi fy cetacean species (Oswald et al. 2003, Rankin et al 2007). Results from these studies show potential for effective passive ac oustic monitoring of depredation at a particular site. However, one challenge of continuous passive acoustic observation is the huge volume of data that is generated and must then be analyzed in a timely manner. Automated sound detection is of ten considered to be the best option for dealing with this volume of data. A number of different techniques for automated detection have been developed such as image processing of spectrograms (Gillespie 2004) and energy calculations wi thin a selected band (Oswald et al. 2004). Passive acoustic observation of a problematic fishing pier was investigated in terms of reliability in monitoring the severity and frequency of the depredation behavior. This study also tested the efficiency and accuracy of the new MATLAB-based software, DSGLab, for building an automated echoloca tion detector. Dolphi n presence around the
104 fishing pier was monitored by determining the number of echolocation clicks detected by the automated program. Trends in echolocati on click rates were examined for variations by month and hour and compared to visual sighting data. This method has the potential to create an inexpensive, universal op tion for assessing depredation by odontocetes. 4.2 Methods 4.2.1 Hydrophone Deployment and Setup In June 2007, an HTI-96 hydrophone (sen sitivity -164 dBV/Pa; 2 Hz-37 kHz) was mounted at Anna Maria City Pier. Con tinuous recordings were taken through June 2008 with the exception of breaks in Oc tober 2007, November 2007, and March 2008 when the hydrophone was being repaired following vandalism, damage from fishing gear, or equipment malfunction. Data were recorded continuously as 10 min files at a sample rate of 44.1 kHz onto a computer with an external hard drive using LoggerheadDT software. The hydrophone was susp ended underneath, ne ar the center of the pier at a depth of approxi mately 1.3 m (of total depth of 2.8 m) above a sandy bottom. The recording system was calibrated by recording 0.1 V-peak sine waves from 100 Hz to 22000 Hz and measuring the levels recorded in the wav files in MATLAB. The system had a flat response from 100 Hz to 22 kHz.
105 4.2.2 DSGLab Automatic Echolocation Detection Data were analyzed automatically using the MATLAB (version 6.5-7.4) based program DSGLab, which allows the user to customize an automated sound detector. An echolocation detector was built rather than a whistle detector because preliminary visual scans of over 19 hours of data (115 files) show ed that echolocation occurred more often than whistles. Due to the infrequent presen ce of dolphins at the pier, the goal of the detector was to reduce the number of false detections. In general, the echolocation detector wo rked by detecting click trains. Recording files were processed in 10 s segments. The detector first filtered the data with a 10 kHz high pass filter which eliminated much of th e boat engine noise from files. The signal was further rectified, enveloped and then wa s thresholded at 2.5 times the root mean square (RMS) of the signal. This resulte d in the detection of both echolocation and snapping shrimp clicks. Recordings were furt her processed by calculating the run length of possible echolocation trains and counting the number of clicks within the specified time frame. The program then output the number of times the criteria were met. If the output value was six or greater, the file was marked as with echolocation. Due to the regularity of clicks, snapping shrimp noise was mostly eliminated and echolocation was detected. A trial run of the detector was conduct ed on 50 files with no echolocation present and 50 files with clicks present. Files in both categories were selected to include recordings with boat engine noise It was determined that th e click train needed to have at least six clicks to ensure the eliminati on of false detection on boat noise or snapping shrimp. During trial testing, the detector was successful at detecti ng 12% of files with
106 known click trains, with no false detections. As a further check, after the detector had completed analysis, 100 files with output values of six or more were randomly selected and visually inspected for ec holocation. The program false detected echolocation clicks in six of the 100 inspec ted files. Detected click trains were thos e that had the highest signal-to-noise ratio. Recordings from June 2007 to June 2008 were sub-sampled at 5% by analyzing one out of every 20 10-min files. The number of files with a maximum value of six or greater (i.e. those with echolocation) was to taled and then the percent of files with echolocation was calculated. A regression in StatSoft Statistica 6.1 was used to determine the predictability of monthly a nd hourly angler interactions rates based on echolocation click detections. Angler inte ractions were measured by the number of depredation events witnessed during pier surveys and were standardized by monthly or hourly effort (i.e. the average monthly or hourly depredation rate ). A total of 11 depredation events by the male dolphin F106 was used to create the depredation curve for Anna Maria City Pier. Since only pier surv ey data with corres ponding recordings were included, data were only available during daylight hours from 07:00 to 15:00, and so were not comparable across night hours. A Spearman R rank correlation in Statistica was used to compare hourly click detections to the mean number of dolphins sighted and mean fishing effort (mean number of lines) also determined during visual pier surveys. 4.2.3 Visual Surveys vs. Passive Acoustic Recordings Dolphins were detected by visual surv eys 27 times at the pier when sound data were also available. The acoustic recordi ngs were analyzed manually to estimate the
107 effectiveness of the passive acoustic method in detecting dol phin presence at a fishing pier, specifically when dolphins were e ngaging in angler inte raction behavior. Corresponding 10-min acoustic recordings were compared to the 14.6 hours of visual sighting data from six days in the summer of 2007: 27-30 June, 3 Jul y, and 11 July. Pier survey observation effort totaled 30.6 hours. All recordings were manually inspected for whistles and echolocation clicks in Adobe Audition 2.0. Mann-Whitney U tests in Statistica were used for comparisons of whis tle and echolocation ra tes during depredation and non-depredation sightings. 4.3 Results 4.3.1 DSGLab Automatic Echolocation Detection The DSGLab automated echolocation dete ctor analyzed 87,746 segments of 10 s recordings. Of processed f iles, 0.13% met the criteria for echolocation. The month with the greatest percentage of files with echol ocation was June 2007 (Figure 4.1). April and June 2008 had no files with detected echolocat ion (Figure 4.1). The correlation between peaks in echolocation and monthly depred ation activity was marginally significant (R2=0.65, p= 0.053). Detections were also anal yzed on an hourly timescale. The highest percentages of echolocation clicks detected per hour were at 04:00 and 12:00 (Fi gures 4.2). Times 05:00, 07:00, 08:00, 21:00, 22:00 had no clicks detected (Figures 4.2). There was no correlation between detected hour ly echolocation rates and dol phins visually sighted per
108 hour (R=-0.34, p=0.366) and a marginally signif icant correlation between hourly echolocation rates and fish ing effort per hour (R=0.66, p=0.053) (Figures 4.3, 4.4). Detected echolocation rates were found to si gnificantly correlate to the increase in depredation during mid-day hours (R2=0.53, p=0.027) (Figure 4.2). 4.3.2 Visual Surveys vs. Passive Acoustic Recordings In 63% of 27 sightings, there was acoustic evidence (whistle or echolocation) of dolphin presence. In total, 103 whistles we re recorded in 22% of sightings and echolocation was detected manua lly in 56% of sightings. Of the total 27 sightings, 21 included the 26-year-old male dolphin, F106. F106 was solitary in 81% of these si ghtings and engaged in angler interaction behaviors in 91% of sightings. Acoustic evidence of the dolphi ns presence was detected in 63% of the sightings when F106 was engaging in angler interaction behaviors. Some of the whistles recorded during the F106 sightings were conf irmed to be the animals signature whistle when compared to recordings from the Sarasota Dolphin Whistle Catalog (Courtesy of Laela Sayigh and Vincent Janik) (Figure 4.5). Furthermore, 36% of F106 depredation events were accompanied by echolocation with in the minute surrounding the event. The dolphin produced echolocation clicks wh ile engaging in both provisioning and scavenging. Mann-Whitney U tests did not find significantly more whistles ( p=0.450) or echolocation clicks ( p=0.719) during the 19 sightings w ith dolphin-angler interaction behavior than the nine sightings wi thout this behavior (Table 4.1).
109 4.4 Discussion Since sightings with dolphin-angler interaction behaviors were not found to contain significantly more whis tles or echolocation than si ghtings without the behavior, passive acoustics cannot be used exclusivel y for determining depredation events at a fishing pier. Dolphins are most likely to wh istle when socializing and the frequency of whistles also increases with larger group size (Cook et al. 2003). When F106 was engaging in depredation behaviors around the pier, the dolphin was usually solitary and because it was being provided with easy p rey, the dolphin would not always have to seek it acoustically or coordinate with othe r dolphins to capture prey. These factors would reduce the dolphins necessity to whis tle or echolocate, thus explaining the nonsignificance of the occurrence of whistles or echolocation between angler interaction and non-angler interaction sightings. F106 was detected acoustica lly in 63% of sightings containing dolphin-angler interaction behavior and was found to echolocate during some provisioning and scavenging events. For dolphins that comm only depredate, like F106, depredation even with echolocation may be as successful, and at equal or lesser energetic cost as other more natural feeding strategies. To more clearly understand the role of echolocation in depredation, the use of echoloc ation during dolphin-angler inte raction behavior should be further examined to include a greater number of known depredating do lphins at different piers. Also, sub-sampling the available data at a greater rate (i.e 10%) will provide a larger sample size that may further increas e the predictability of echolocation in determining dolphin presence a nd depredation behavior.
110 Peaks in echolocation were correlated to both monthly and hourly depredation events, suggesting that when dolphins are present around the pier there is a greater probability that depredation or angler inte ractions will occur. The hourly percent detected echolocation clicks show a small peak in the very early morning hours, but show the greatest peak at 12:00. This peak at noon was the height in F106 depredation activity at the pier. Dolphin echolocat ion is typically associated w ith foraging (Jones and Sayigh 2002, Nowacek 2005) and findings from previous studies show that dolphins feed more often in early morning hours (Shane et al. 1986, Shane 1990, Waples 1995, Allen et al. 2001). This shift away from the expected echolocation feeding pa tterns further suggests that depredation behavior changes th e dolphins natural ac tivity pattern. If further studies reveal that echolocation or whistles are produc ed in the majority of odontocete depredation foraging incidents, passive acoustics could serve as a method for reducing commercial fisheries depredat ion and possibly bycatch. Active acoustic methods using pingers that emit tone pulses ha ve been explored in attempting to reduce bycatch of marine mammals by deterring or aler ting animals to the presence of fisheries nets (Kraus et al. 1997, Stone et al. 1997, Dawson et al 1998, Bordino et al. 2002, Barlow and Cameron 2003, Cox et al. 2003, Monteiro-Neto et al. 2004). However, concerns arise with pinger use, since the constantly emitted pulses can act as a dinner bell leading to increased depredation from the nets, and poten tially causing hearing damage for net interac tion animals (Richardson et al. 1995, Bordino et al. 2002, Cox et al 2003). A recorder with the ability to broadcast in real time attached to long-lines, gillnets, or other types of fisher ies gear could alert fishermen to the presence of odontocete
111 cetaceans and potential acts of depredation. Fi shers could either pull gear and terminate a set early or avoid the area wh en next setting fishing gear. Detection of whistles or echolocation could be automated therefore just alerting the fishermen to cetacean presence by alarm when echolocation or whis tle rates crossed a threshold level. The advantages of developing such a device ra ther than using the traditional hydrophone are that fishermen do no have to be trained or hire additional personal to monitor the data in real time. Also, the device can be left on so aking gear but still br oadcast information to fishermen that may be some distance away. Such a device may be expensive, however if proven successful, the cost of the device ma y outweigh the economic loss of damage to catch and the harm to marine mamma ls as a result of depredation. I recommend both visual and acoustic su rveys when monitoring dolphin presence around a fishing pier and assessing the rate of angler or fisheries in teractions. Initially, visual surveys proved valuable when evaluati ng the severity of dolphin-angler interaction behavior at the pier and for identifying problem dolphi ns. However, passive acoustic monitoring was less expensive and time intensiv e than visual surveys, detected problem animal presence more than 50% of the time, and correlated echolocation with peaks in monthly and hourly depredation activity. Peak s in echolocation can alert scientists and managers to increases in dolphin presence th at might be indicative of dolphin-angler interactions at the specific site If cause for concern, scientists and managers could then employ more intense and costly monitoring methods (e.g. visual surveys) for precisely determining the extent and rate of depr edation behaviors during times of peak echolocation. Passive acoustic monitoring has the potential to serve as an inexpensive and preliminary assessment of depredation behavi ors at a specified site. More studies in
112 other areas with similar dolphinangler interaction issues woul d be beneficial in further assessing the reliability of passive acoustics to monitoring depredation behavior.
113 Table 4.1 Summary of visual dolphin sighti ngs and corresponding ac oustic recordings over six days at Anna Maria City Pier. Dol phins were not found to whistle or echolocate significantly more in sightings with angl er interaction behavior than without ( p=0.450 for whistles; p=0.719 for echolocation). Date Angler Interaction Behavior? Sighting No. Animal(s) Present Whistle? Echolocation? 27-Jun-07 Y 101 F106 27-Jun-07 Y 102 F106 X 27-Jun-07 Y 103 F106, FB10 X 27-Jun-07 Y 104 F106 28-Jun-07 N 101 No photos X 28-Jun-07 N 102 FB36 X 28-Jun-07 N 103 No photos 29-Jun-07 Y 101 F106 X 29-Jun-07 Y 102 F106 X X 29-Jun-07 N 103 F106, FB65, C654 X 29-Jun-07 Y 104 F106, FB65, C654 X 29-Jun-07 Y 105 F106 X 29-Jun-07 Y 106 F106, FB10 X X 30-Jun-07 N 101 FB93, C934, F196 30-Jun-07 N 102 FB93, C934, F196 30-Jun-07 Y 103 F106 30-Jun-07 Y 104 F106 30-Jun-07 Y 105 F106 X X 30-Jun-07 Y 106 F106 X 30-Jun-07 Y 107 F106 3-Jul-07 Y 101 F106 X X 3-Jul-07 Y 102 F106 3-Jul-07 Y 103 F106 3-Jul-07 N 104 Unknown X 3-Jul-07 Y 105 F106 X 3-Jul-07 Y 106 F106 X 11-Jul-07 N 48 F106 X
0 0.5 1 1.5 2 2.5 3 3.5 Jun07 Jul07 Aug07 Sep07 Oct07 Nov07 Dec07 Jan08 Feb08 Mar08 Apr08 May08 Jun08 Jul08 DateAvg. no. of depredation events0.0% 0.1% 0.2% 0.3% 0.4% 0.5% 0.6% 0.7% 0.8% 0.9% Depredation % detect echoesPercent detected echolocation Figure 4.1 The percentage of 10 s segments containing echolocation click trains per month at Anna Maria City Pier compared to the monthly depredati on rate (determined by pier surveys) for Sarasota Bay, FL (R2=0.65, p=0.053). 114
0.0% 0.1% 0.1% 0.2% 0.2% 0.3% 0.3% 0.4% 0.4% 0.5% 0.5%0 : 00 1 : 0 0 2 : 0 0 3 : 0 0 4 : 0 0 5:00 6:00 7:00 8:00 9:00 10 : 0 0 11 : 0 0 1 2: 0 0 1 3:00 14:00 15:00 16:00 17:00 18 : 0 0 1 9: 0 0 2 0:00 2 1:00 2 2:00 2 3:00HourPercent detected echolocation0 0.05 0.1 0.15 0.2 0.25 0.3 % detect echo DepredationAvg. no. of depredation events Figure 4.2 The percentage of 10 s segments containing echolocation click trains per hour at Anna Maria City Pier compared to the hourly depredation rate (determined by pier surveys) for Sarasota Bay, FL (R2=0.53, p=0.027). 115
0.0% 0.1% 0.1% 0.2% 0.2% 0.3% 0.3% 0.4% 0.4% 0.5% 0.5%0 :0 0 1 : 00 2:00 3:0 0 4 : 00 5 : 00 6:0 0 7:0 0 8 : 00 9:00 10:0 0 1 1 :0 0 1 2 : 00 13:0 0 14:0 0 1 5 :0 0 16:00 17:0 0 1 8 :0 0 1 9 : 00 20:0 0 21:0 0 2 2 :0 0 2 3 : 00HourPercent detected echolocation0 0.5 1 1.5 2 2.5 3 % detect echo Avg. dolphinsAvg. no. of dolphins Figure 4.3 The percentage of 10 s segments containing echolocation click trains compared to the mean number of dolphins sighted per hour (determined by pier surveys) at Anna Maria City Pier. Data sets were not correlated (R=-0.34, p=0.366). 116
0.0% 0.1% 0.1% 0.2% 0.2% 0.3% 0.3% 0.4% 0.4% 0.5% 0.5%0: 0 0 1 : 0 0 2:00 3:00 4 : 0 0 5 : 0 0 6:00 7 : 0 0 8 : 0 0 9:00 1 0 : 0 0 1 1 : 0 0 12:00 13: 0 0 1 4 : 0 0 15:00 16: 0 0 1 7 : 0 0 18:00 19:00 2 0 : 0 0 21 : 0 0 22:00 2 3 : 0 0HourPercent detected echolocation0 2 4 6 8 10 12 14 % detect echo Avg. fishing linesAvg. no. of fishing lines Figure 4.4 The percentage of 10 s segments with echolocation click trains compared to the mean number of total fishing lines per hour (determined by pier surveys) at Anna Maria City Pier. Data sets were marginally correlated (R=0.66, p=0.053). 117
35000 30000 25000 20000 Frequency (Hz) 15000 10000 5000 0 0.5 0.9 Time (s) 118 Figure 4.5 Comparison of F106 signature whis tles from the Sarasota Dolphin Whistle Catalog (top) (courtesy of L. Sayigh and V. Janik) and from pier recordings (bottom). 20000 22000 15000 Frequency (Hz) 10000 5000 0 0.5 0.9 1.0 Time (s)
119 Literature Cited Aguilar Soto, N., M.P. Johnson, P.T. Madsen, F. Daz, I. Domnguez, A. Brito, and P. Tyack. 2008. Cheetahs of the deep sea: deep foraging sprints in short-finned pilot whales off Tenerife (Canary Islands ). Journal of Animal Ecology, 77:936-947. Allen, M.C., A.J. Read, J. Gaudet, L.S. Sayigh. 2001. Fine-scale habitat selection of foraging bottlenose dolphins Tursiops truncatus near Clearwater, Florida. Mar Ecol Prog Ser, 222:253-264. Altmann, J. 1974. Observation study of beha vior: sampling methods Behaviour, 49:227265. Au, W.W.L. 1993. The Sonar of Dolphi ns. Springer-Verlag, New York. 277p. Au, W.W.L. 2004. Echolocation sign als of wild dolphins. Acoustical Physics, 54(4):454462. Au, W.W.L., C.A. Carder, R.H. Penner, B.L. Scronce. 1985. Demonstration of adaptation in beluga whale echolocation signals. J Acous Soc Am, 77:726-730. Au, W.W.L, P.E. Nachtigall, and J.L. Pa wloski. 1999. Temporary threshold shift in hearing induced by an octave band of con tinuous noise in the bottlenose dolphin. J Acous Soc Am, 106(4):2251. Ballance, L.T. 1992. Habitat use patterns and ra nges of the bottlenose dolphin in the Gulf of California, Mexico. Mar Ma mmal Science, 8(3):262-274. Barlow, J. and G.A. Cameron. 2003. Field expe riments show that acoustic pingers reduce marine mammal bycatch in the California drift gillnet fishery. Mar Mammal Science, 19:265-283. Barlow, J. and B.L. Taylor. 2005. Estima tes of sperm whale abundance in the northeastern temperate Pacific from a co mbined acoustic and visual survey. Mar Mammal Science 21(3):429-445. Barros, N.B. and R.S. Wells. 1998. Prey a nd feeding patterns of bottlenose dolphins ( Tursiops truncatus ) in Sarasota Bay, Florida. Journal of Mammalogy, 79(3):10451059.
120 Bejder, L., A. Samuels, H. Whitehead, a nd N. Gales. 2006. Interpreting short-term behavioural responses to disturbance with in a longitudinal perspective. Animal Behaviour, 72:1149-1158. Bordino, P., S. Kraus, D. Albareda, A. F azio, A. Palmerio, M. Mendez, and S. Botta. 2002. Reducing incidental mortality of Franciscana dolphin Pontoporia blainvilei with acoustic warning devices attached to fishing nets. Mar Mammal Science, 18:833-842. Bossart, G.D. 2006. Marine mammals as sentin el species for oceans and human health. Oceanography, 19:134-137. Bossart, G.D., D.G. Baden, R.Y. Ewing, B. Roberts, S.D. Wri ght. 1998. Brevetoxicosis in manatees ( Trichechus manatus latirostris ) from the 1996 epizootic: gross, histologic, and immunohistochemical features. Toxicol Pathol, 26:276-282. Broadhurst, M.K. 1998. Bottlenose dolphins, Tursiops truncatus removing bycatch from prawn-trawl codends during fishing in New South Wales, Australia. Marine Fisheries Review, 60(3):9-14. Brotons, J.M., A.M. Grau, and L. Rendell. 2008 Estimating the impact of interactions between bottlenose dolphins and artisanal fisheries around the Balearic Islands. Mar Mammal Science, 24(1):112-127. Buckstaff, K.C. 2004. Effects of watercraft noise on the acoustic behavior of bottlenose dolphins, Tursiops truncatus in Sarasota Bay, Florida. Mar Mammal Science, 20(4):709-725. Chilvers, B.L, and P.J. Corkeron. 2001. Trawling and bottlenose dolphins social structure. Proc Roy Soc Lond, B Biol Sci, 268(1479):1901-1905. Coleman, F.C., W.F. Figueira, J.S. Uela nd, and L.B. Crowder. 2004. The impact of United States recreational fisheries on marine fish populations. Science, 305(5692):1958. Constantine, R., D.H. Brunton, and T. Denni s. 2004. Dolphin-watching tour boats change bottlenose dolphin ( Tursiops truncatus ) behaviour. Biological Conservation, 117(3):299-307. Cook, M. L. H., L.S. Sayigh, J.E. Blum, and R.S. Wells. 2004. Signature-whistle production in undisturbed free-ra nging bottlenose dolphins (Tursiops truncatus ). Proc R Soc Lond B, 271:1043-1049.
121 Cook, M.L.H., R.A. Varela, J.D. Goldstein, S. D. McCulloch, G.D. Bossart, J.J. Finneran, D. Houser, and D.A. Mann. 2006. Beaked whale auditory evoked potential hearing measurements. J Comp Physiol A, 192:489-495. Cook, M.L.H. 2006. Behavioral and auditory evoked potential (AEP) hearing measurements in odontocete ce taceans. Ph.D. thesis. University of South Florida. Cox, T.M., A.J. Read, D. Swanner, K. Uria n and D. Waples. 2003. Behavioral responses of bottlenose dolphins, Tursiops truncatus to gillnets and acous tic alarms. Biological Conservation, 115:203-212. Cunningham-Smith, P., D.E. Colbert, R.S. We lls, and T. Speakman. 2006. Evaluation of human interactions with a provisioned wild bottlenose dolphin ( Tursiops truncatus ) near Sarasota Bay, Florida, and efforts to curtail the interactions. Aquatic Mammals, 32:346-356. Danil, K., D. Maldini, and K. Marten. 2005. Patterns of use of Makua Beach, Oahu, Hawaii, by spinner dolphins ( Stenella longirostris ) and potential effects of swimmers on their behavior. Aquatic Mammals, 31(4):403-412. Dawson, S.M., A.J. Read, and E. Slooten. 1998. Pingers, porpoises and power; uncertainties with using pingers to redu ce bycatch of small cetaceans. Biological Conservation, 84:141-146. Donoghue, M., R.R. Reeves, and G. Stone. 2002. Report on the workshop on interactions between cetaceans and longline fisheries he ld in Apia, Samoa. November 2002. New England Aquatic Forum Se ries Report 03-1. 44p. Emlen, J.M. 1966. The role of time and en ergy in food preference. The American Naturalist, 100(916):611-617. Erbe, C. 2002. Underwater noise of whale-watc hing boats and potential effects on killer whales ( Orcinus orca ), based on an acoustic impact model. Mar Mammal Science, 18(2):394-418. Finn, H., R. Donaldson, and M. Calver. 2008. Fe eding flipper: a case study of a humandolphin interaction. Pacific C onservation Biology, 14:215-225. Finneran, J.J., and D.S. Houser. 2006. Comp arison of in-air evoked potential and underwater behavioral hear ing thresholds in four bottlenose dolphins (Tursiops truncatus ). J Acous Soc of Am, 119:3181-3192. Fire, S.E., D. Fauquier, L.J. Flewelling, M. Henry, J. Naar, R. Pierce, and R.S. Wells. 2007. Brevetoxin exposure in bottlenose dolphins ( Tursiops truncatus ) associated with Karenia brevis blooms in Sarasota Bay, Florida. Mar Biol, 152:827-834.
122 Fire, S.E., L.J. Flewelling, J. Naar, M.J. Twiner, M.S. Henry, R.H. Pierce, D.P. Gannon, Z. Wang, L. Davidson, and R.S. Wells. 2008. Pr evalence of brevetoxins in prey fish of bottlenose dolphins in Sarasota Bay, Florida. Mar Ecol Prog Ser, 368:283-294. Flewelling, L.J., J.P. Naar, J.P. Abbott, D.G. Baden, N.B. Barros, G.D. Bossart, M.Y.D. Bottein, D.G. Hammond, E.M. Haubold, C.A. Heil, M.S. Henry, H.M. Jacocks, T.A. Leighfield, R.H. Pierce, T.D. Pitchford, S. A. Rommel, P.S. Scott, K.A. Steidinger, E.W. Truby, F.M. Van Dolah, and J.H. Landsberg. 2005. Red tides and marine mammal mortalities. Nature, 435:755-756. Florida Charts. 2007. Trends in population gr owth. Florida Legislature, Office of Economic & Demographic Research; Florida Office of Vital Statistics. http://www.floridacharts.com/charts Gannon, D.P., N.B. Barros, D.P. Nowacek, A. J. Read, D.M. Waples, and R.S. Wells. 2005. Prey detection by bottlenose dolphins, Tursiops truncatus : an experimental test of the passive listening hypothesis. Animal Behaviour, 69(3):709-720. Gannon, D.P., E.J. Berens, S.A. Camilleri, J.G. Gannon, M.K. Brueggen, A. Barleycorn, V. Palubok, G.J. Kirkpatrick and R.S. Wells. 2009. Effects of Karenia brevis harmful algal blooms on nearshore fish communities in southwest Florida. Marine Ecology Progress Series, 378:171-186. Gillespie, D. 2004. Detection and classificati on of right whale calls using an edge detector operating on a smoothed spectrogr am. Canadian Acoustics, 32:39-47. Gorzelany, J.F. 1998. Unusual deaths of tw o free-ranging Atlantic bottlenose dolphins ( Tursiops truncatus ) related to ingestion of recrea tional fishing gear. Mar Mammal Science, 14(3):614-617. Heithaus, M.R. and L.M Dill. 2002. Food avai lability and tiger shark predation risk influence bottlenose dolphin habi tat use. Ecology, 83(2):480-491. Houser, D.S., D.A. Helweg, and P.W. Moore. 1999. Classification of dolphin echolocation clicks by energy and frequency distributions. J Acous Soc Am, 106(3):1579-1585. Houser, D.S., and J.J. Finneran. 2006a. A comparison of underwater hearing sensitiviy in bottlenose dolphins (Tursiops truncatus ) determined by electrophysiological and behavioral methods. J Acous Soc America, 120:1713-1722. Houser, D.S. and J.J. Finneran. 2006 b. Variation in the hearin g sensitivity of a dolphin population obtained through the use of e voked potential audiometry. J Acous Soc America, 120:4090-4099.
123 Houser, D.S., A. Gomez-Rubio, and J.J. Fi nneran. 2008. Evoked potential audiometry of 13 Pacific bottlenose dolphins ( Tursiops truncatus gilli ). Mar Mammal Science, 21(1):28-41. Johnson, C.S. 1967. Sound detection thresh olds in marine mammals. Pp. 247-260. In: W.N. Tavaloga (ed.), Marine bioaco ustics, Vol. 2. Pergamon, New York. Jones, G.J. and L.S. Sayigh. 2002. Geographic variation in rates of vocal production of free-ranging bottlenose dolphins. Mar Mammal Science, 18(2):374-393. Kirkpatrick, B., L.E. Fleming, D. Squicciarini, L.C. Backer, R. Clark, W. Abraham, J. Benson, Y.S. Cheng, D. Johnson, R. Pierce, J. Zaias, G.D. Bossart and D.G. Baden. 2004. Literature review of Florida red tide: implications for human health effects. Harmful Algae, 3(2):99-115. Kraus, S.D., A.J. Read, A. Solow, K. Baldwin, T. Spradlin, E. Anderson, J. Williamson. 1997. Acoustic alarms reduce porpoise mortality. Nature, 388:525. Lammers, M.O., M. Schotten, and W.W.L A u. 2006. The spatial context of free-ranging Hawaiian spinner dolphins ( Stenella longirostris ) producing acoustic signals. J Acous Soc Am, 119(2):1244-1250. Lauriano, G., C.M. Fortuna, G. Moltedo, and G. Notarbartolo Di Sciara. 2004. Interactions between common bottlenose dolphins ( Tursiops truncatus ) and the artisanal fishery in Asinara Island National Park (Sardinia): assessment of catch damage and economic loss. J Cetacean Res Manage, 6(2):165-173. Lusseau, D. 2003. Effects of tour boats on th e behavior of bottlenose dolphins: using Markov chains to model anthropogenic imp acts. Conservation Biology, 17(6): 17851793. Lusseau, D. 2004. The hidden cost of tourism: detecting long-term effects of tourism using behavioral informati on. Ecology and Society, 9(1): http://www.ecologyandsociet y.org/vol9/iss1/art2. Lusseau, D. 2006. The short-term behavioral reactions of bottlenose dolphins to interactions with boats in Doubtful Sound, New Zealand. Mar Mammal Science, 22(4):802-818. MacArthur, R.H. and E.R. Pianka. 1966. On optimal use of a patchy environment. The American Naturalis t, 100(916):603-609. Mann, J. 1999. Behavioral sampling methods fo r cetaceans: a review and critique. Mar Mammal Science, 15(1):102-1222.
124 Mann, J. and B. Smuts. 1999. Behavioral development in wild bottlenose dolphin newborns ( Tursiops sp.). Behaviour 136(5):529-566. Mann, J. and B. Sargeant. 2003. Like moth er, like calf: the ontogeny of foraging traditions in wild Indian ocean bottlenose dolphins ( Tursiops sp.) Pp. 236-266 In : D.M. Fragaszy and S. Perry (eds.), The biology of traditions; models and evidence. Cambridge Univ. Press, New York. 478p. McDonald, M.A. and C.G. Fox. 1999. Passive acoustic methods applied to fin whale population density estimation. J Acous Soc Am, 105(5):2643-2651. Mellinger, D.K., K.M. Stafford, S.E. Moor e, L. Munger, and C.G. Fox. 2004. Detection of North Pacific right whale ( Eubalaena japonica ) calls in the Gulf of Alaska. Mar Mammal Science, 20:872-879. Mellinger, D.K., K.M. Stafford, S.E. M oore, R.P. Dziak, and H. Matsumoto. 2007. An overview of fixed passive acoustic observation methods for cetaceans. Oceanography, 20(4):36-45. Monteiro-Neto, F.J.C. vila, T.T. Alves-Jr., D.S. Silva Arajo, A. Alves Campos, A.M.A. Martins, C. Leite Parente, M. A. Andrade Furtado-Neto, J. Lien. 2004. Behavioral responses of Sotalia fluviatilis (Cetacea, Delphinidae) to acoustic pingers, Fortaleza, Brazil. Mar Mammal Science, 20(1):145-151. Moore, W.B. and D.A. Pawloski. 1990. Investigations of the control of echolocation in the dolphin ( Tursiops truncatus ). Pp. 305-316 In : J. Thomas and R. Kastelein (eds.), Sensory abilities of cetaceans: laborator y and field evidence. Plenum, New York. 710p. Myers, R.A., and B. Worm. 2003. Rapid wo rldwide depletion of predatory fish communities. Nature, 423: 281-284. Nachtigall, P.E., A.Y. Supin, J. Pawloski and W.W.L. Au. 2004. Temporary threshold shifts after noise exposure in the bottlenose dolphin ( Tursiops truncatus ) measured using evoked auditory potentials Mar Mammal Science 20:673-687. Nachtigall, P.E., T.A. Mooney, K.A. Tayl or. M.M.L Yuen. 2007. Hearing and auditory evoked potential methods applied to odont ocete cetaceans. Aquatic Mammals, 33(1):6-13. NOAA. 2006a. Bottlenose Dolphins Increase in depredatory (stealing) behavior and deaths associated with recreational fishi ng gear. NOAA Fisheries Services, Southeast Regional Office, October 2006.
125 NOAA. 2006b. F/SER32:KT. NOAA Fisheries Servi ces, Southeast Regional Office, 10 August 2006. Noke,W.D. and D.K. Odell. 2002. Interacti ons between the Indian River Lagoon blue crab fishery and the bottlenose dolphin, Tursiops truncatus Mar Mammal Science, 18: 819-832. Nowacek, D.P. 1999. Sound use, sequential beha vior and ecology of foraging bottlenose dolphins, Tursiops truncatus Ph.D.thesis. Massachusetts Institute of Technology/ Woods Hole Oceanographic Institution. Nowacek, D.P. 2002. Sequential foraging behaviour of bottlenose dolphins, Tursiops truncatus in Sarasota Bay, Florida. Behaviour, 139(9):1125-45. Nowacek, D.P. 2005. Acoustic ecology of foraging bottlenose dolphins (Tursiops truncatus ), habitat-specific use of three sound types. Mar Mammal Science 21(4):587-602. Nowacek, S.M., R.S. Wells, and A.R. Solow. 2001. Short-term effects on boat traffic on bottlenose dolphins, Tursiops truncatus in Sarasota Bay, Florida. Mar Mammal Science, 17(4):673-688. Owen, E.C.G., R.S. Wells, and S. Hofmann. 2002. Ranging and association patterns of paired and unpaired adult male Atlantic bottlenose dolphins, Tursiops truncatus in Sarasota, Florida, provide no evidence for alternative male strategies. Can J Zool, 80:2072-2089. Oswald, J.N., J. Barlow, and T. Norris. 2003. Acoustic identification of nine delphinid species in the eastern tropical Pacifi c Ocean. Mar Mammal Science, 19:20-37. Oswald, J.N., S. Rankin, and J. Barlow. 2004. The effect of recording and analysis bandwidth on acoustic identification of delphinid species. J Acous Soc Am, 116:3178-3185. Parks, S.E. and P.L. Tyack. 2005. Sound production by North Atlantic right whales ( Eubalaena glacialis ) in surface active groups. J Acous Soc Am, 117:3297-3306. Peddemors, V. 2001. A review of cetacean intera ctions with fisheries and management thereof in South Africa. Inte rnational Whaling Commission. 41p. Read, A.J. 2008. The looming crisis: Interactions between marine mammals and fisheries. J. of Mammalogy, 89(3):541-548.
126 Reeves, R.R., A.J. Read, and G. Notarbarto lo di Sciara (editors). 2001. Report of the workshop of interactions between dolphins and fisheries in the Mediterranean: evaluation of mitigation alternatives. Istitu to Centrale per al Ricerca Applicata al Mare. Rome, Italy. Richardson, W.J., C.G. Greene Jr., C.I. Ma lme, D.H. Thomson. 1995. Marine mammals and noise. Academic Press, San Deigo. 204p. Ridgway, S.H., T.H. Bullock, D.A. Carder, R.L. Seeley, D. Woods, and R. Galambos. 1981. Auditory brainstem response in dolphins. Neurobiology, 78:1943-1947. SDRP. 2006. Manual for field research and la boratory activities. Sarasota Dolphin Research Program. 62p. pdf download av ailable from www.sarasotadolphin.org Secchi, E.R., and T. Vaske, Jr. 1998. Killer whale ( Orcinus orca ) sightings and depredation on tuna and swor dfish longline catches in southern Brazil. Aquatic Mammals, 24(2):117-122. Shane, S.H. 1990. Behavior and ecology of the bottlenose dolphin at Sanibel Island, Florida. Pp. 245-265. In S. Leatherwood and R.R. Reeves, (eds.). The bottlenose dolphin. Academic Press, Sand Diego, CA. Shane, S.H., R.S. Wells, and B. Wrsig. 1986. Ecology, behavior, and social organization of the bottlenose dolphin: a review Mar Mammal Science, 2(1):34-63. Sigler, M.F., C.R. Lunsford, J.M. Stra ley, and J.B. Liddle. 2008. Sperm whale depredation of sablefish l ongline gear in the northeast Pacific Ocean. Mar Mammal Science, 24(1):16-27. South Pacific Region Environment Program (S PREP). 2002. Plan of action and priorities for research to reduce depr edation on longlines by cetaceans. SPREP Technical Workshop, November 2002 in Apia, Somoa. Supin, A.Y., V.V. Popov, and V.O. Klishi n. 1993. ABR frequency tuning curves in dolphins. J Comp Physiol A, 173:649-656. Supin, A.Y., and V.V. Popov. 1995. Envel ope-following response and modulation transfer function in the dolphins aud itory system. Hearing Research, 92:38-46. Stafford, K.M., C.G. Fox, and D.S. Clark. 1998. Long-range de tection and lo calization of blue whale calls in the northeast Pacifi c using military hydrophone arrays. J Acous Soc Am, 104:3616-3625.
127 Stafford, K.M., S.L. Nieukirk, and C.G. F ox. 2001. Geographic and seasonal variation of blue whale calls in the North Pacific. J ournal of Cetacean Research and Management, 3:65-76. Steidinger, K.A., G.A. Var go, P.A. Tester, and C.R. To mas. 1998. Bloom dynamics and physiology of Gymnodinium breve with emphasis on the Gulf of Mexico. Pp. 133-153 In: D.A. Anderson, A.D. Cembella, and G.M. Hallegraeff (eds.), Physiological ecology of harmful algal bloom s. Springer-Verlag, Berlin. Stimpert, A.K., D.N. Wiley, W.W.L. A u, M.P. Johnson, and R. Arsenault. 2007. Megapclicks: acoustic click trains and buzzes produced during night-time foraging of humpback whales ( Megaptera novaeangliae ). Biol Lett, 3:467-470. Stone, G., S. Kraus, A. Hutt, S. Martin, A. Yoshinaga, and L. Joy. 1997. Reducing bycatch: Can acoustic pingers keep Hectors dolphins out of fishing nets? Marine Technology Society Journal, 31(2):3-7. Sutinen, J.G. and R.J. Johnston. 2003. Angling management organizations: integrating the recreational sector in to fishery management. Marine Policy, 27(6):471-487. Szymanski, M.D., D.E. Bain, K.K. K. Ki ehl, S. Pennington, S. Wong, and K.R. Henry. 1999. Killer whale ( Orcinus orca ) hearing: Auditory brainstem response and behavioral audiograms. J Acous Soc Am, 106:1134-1141. Tester, P.A., J.T. Turner, and D. Shea. 2000. Vectorial transport of toxins from the dinoflagellate Gymnodinium breve through copepods to fish. J Plankton Res, 22:4761. Tyack, P.L., M. Johnson, N. Aguilar Soto, A. Sturlese, and P.T. Madsen. 2006. Extreme diving of beaked whales. The Journa l of Experimental Biology, 209:4238-4253. U.S. Dept. of the Interior, Fish and Wildlife Service, and U.S. Dept. of Commerce, and U.S. Census Bureau. 2006. 2006 National Surv ey of Fishing, Hunting, and WildlifeAssociated Recreation. 162p. Van Voorhees, D., and E.S. Pritchard. 2008. Fi sheries of the United States. July 2008. National Marine Fisherie s Service, Office of Science and Technology 2007 Preliminary Report on Commercial and Recreational Fisheries of the United States. 103pp. Waples, D.M. 1995. Activity budgets of free-ranging bottlenose dolphins ( Tursiops truncatus ) in Sarasota Bay, Florida. MS thesis University of California Santa Cruz.
128 Watwood, S.L., P.J.O. Miller, M. Johnson, P.T. Madsen, and P.L. Tyack. 2006. Deepdiving foraging behavior of sperm whales ( Physeter macrocephalus). Journal of Animal Ecology, 75:814-825. Wells, R.S. 1991. The role of long-term study in social structure of a bottlenose dolphin community. Pp.199-225 In : K. Pryor and K.S. Norris (eds.), Dolphin Societies: Discoveries and Puzzles. Univ. of California Press, Berkeley. 397 pp. Wells, R.S., A.B. Irvine, and M.D. Scott. 1980. The social ecology of inshore odontocetes. Pp. 263-317 In : Herman, L.M. (ed.) Cetacean Behavior: Mechanisms and Functions. New York, J Wiley & Sons. 463 pp. Wells, R.S., M.D. Scott and A.B. Irvine. 1987. The social structure of free-ranging bottlenose dolphins. Pp. 247-305 In : Genoways, H. (ed.), Current Mammalogy, Vol. 1. New York, Plenum Press. Wells, R.S. and M.D. Scott. 1994. Incidence of gear entanglement for resident inshore bottlenose dolphins near Sarasota, Florida. Page 629 In: W.F. Perrin, G.P. Donovan, and J. Barlow (eds.), Gillnets and Cet aceans, Rep Int Whal Commn (Special Issue 15). Wells, R.S., S. Hofmann, and T.L. Moors. 1998. Entanglement and mortality of bottlenose dolphins, Tursiops truncatus in recreational fishing gear in Florida. Fish. Bull., 96:647-650. Wells, R.S., D.J. Boness and G.B. Rathbun. 1999. Behavior. Pp. 324-422 In : J.E. Reynolds, III and S.A. Rommel, (eds.), Biology of Marine Mammals. Smithsonian Institution Press, Washington, DC. 578 pp. Wells, R.S. 2003. Dolphin social complexity : Lessons from long-term study and life history. Pp. 32-56 In: F.B.M. de Waal and P.L. Tyack, eds., Animal Social Complexity: Intelligence, Culture, and Indivi dualized Societies. Harvard University Press, Cambridge, MA. 616p. Wells, R.S., K. Bassos-Hull, J. Allen, N. Barros, and D. Fauquier. 2006. Impacts of human activities on a long-term reside nt community of bottlenose dolphins on Floridas west coast. Defenders of Wildli fe, Carnivores Conference, St. Petersburg, FL, November 12-15, 2006. Wells, R.S., J.B. Allen, S. Hofmann, K. Bassos-Hull, D.A. Fauquier, N.B. Barros, R.E. DeLynn, G. Sutton, V. Socha, and M.D. Scott. 2008. Consequences of injuries on survival and reproduction of common bottlenose dolphins ( Tursiops truncatus ) along the west coast of Florida. Ma r Mammal Science, 24(4):774-794.
129 Whitehead, H.L., L. Rendell, R.W. Osborne, and B. Wrsig. 2004. Culture and conservation of non-humans with reference to whales and dolphins: review and new directions. Biological Conservation, 120:427. Williams, R., D. Lusseau, and P.S. Hammond. 2006. Estimating relative energetic costs of human disturbance to killer whales ( Orcinus orca ). Biological Conservation, 133:301-311. Wrsig, B., and M. Wrsig. 1977. The photographic determination of group size, composition, and stability of coastal porpoises ( Tursiops truncatus ). Science, 198(4318):755-756. Zimmer, W.M.X, M.P. Johnson, A. DAmico, and P.L. Tyack. 2003. Combining data from a multisensor tag and passive sonar to determine the diving behavior of a sperm whale (Physeter macrocephalus ). IEEE Journal of Oceanic Engineering, 28(1):13-28. Zollett, E.A. and A.J. Read. 2006. Depred ation of catch by bottlenose dolphins ( T. truncatus ) in Florida King mackerel ( Scomberomorus cavalla ) troll fishery. Fishery Bull, 104:343-349.