1 AGE, GROWTH, AND DIET OF LIONFISH ( PTEROIS SPP. ) FROM LITTLE CAYMAN ISLAND, B.W.I. BY MORGAN EDWARDS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCI ENCE UNIVERSITY OF FLORIDA 2012
2 2012 Morgan Edwards
3 To my Dad, Mom, Joni and Drew
4 AKNOWLEDGEMENTS I thank my committee: Dr. Frazer, Dr. Jacoby, and Dr. Allen for imparting their advice and expertise throughout this project. Special than ks go to Dr. Jacoby for always fitting me in his very busy schedule when his time is in such high demand. I thank James Morris and Jennifer Potts for generously agreeing to provide training in otolith removal and preparation, as well as answering my numero us questions. I thank Dr. Parkyn and Dr. Murie for the expert advice they provided on ageing fish that are not always easy to read, and many other matters. I thank Dr. Selena Heppell for her input lab mates: Darlene, Jessica, Sky, Savanna, and Pat for their help in logistical matters as well as fish work up and processing. I thank Christian Barientos and Drew Dutterer for doing the second and third reads on the otoIiths. I thank the Central Caribbe an Marine Institute for providing room, board, and equipment, as well as their wonderful staff. I thank the resorts of Little Conch Club for providing boats and fuel for the c ulls. I thank all of the staff from the resorts for generously donating their free time to aid in the culls and so graciously donating the fish to my project. I would never have been able to complete this project to this extent without them, and I am extre mely grateful. I am forever grateful to my family for their support, love, and unwavering faith in me. I am so grateful to Drew for his love Last but not least, I thank my friends for their support and encouragement
5 TABLE OF CONTENTS page AKNOWLEDGEMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 GENERAL INT RODUCTION ................................ ................................ .................. 11 2 AGE STRUCTURE AND GROWTH OF LIONFISH OFF LITTLE CAYMAN ISLAND ................................ ................................ ................................ ................... 14 Introduction ................................ ................................ ................................ ............. 14 Study Site ................................ ................................ ................................ ............... 15 Methods ................................ ................................ ................................ .................. 16 Fish Collection and Processing ................................ ................................ ........ 16 Otolith Processing and Counts of Annuli ................................ .......................... 16 Marginal Increment Analysis ................................ ................................ ............ 17 Counts and Validation of Da ily Rings ................................ ............................... 17 Modeling Fish Growth ................................ ................................ ...................... 18 Results ................................ ................................ ................................ .................... 19 Lionfish Size St ructure and Length Weight Relationship ................................ .. 19 Marginal Increment Analysis ................................ ................................ ............ 20 Analysis of Daily Rings ................................ ................................ ..................... 20 Estimates of Growth Rates ................................ ................................ ............... 20 Discussion ................................ ................................ ................................ .............. 21 3 DIET OF LIONFISH OFF LITTLE CAYMAN ISLAND ................................ ............. 36 Introduction ................................ ................................ ................................ ............. 36 Methods ................................ ................................ ................................ .................. 38 Collection and Processing ................................ ................................ ................ 38 Calculation of Alimentary Indices ................................ ................................ ..... 39 Results ................................ ................................ ................................ .................... 40 General Observations ................................ ................................ ...................... 40 Alimentary Index ................................ ................................ ............................... 41 Discussion ................................ ................................ ................................ .............. 42 4 CONCLUSION ................................ ................................ ................................ ........ 52 LIST OF REFERENCES ................................ ................................ ............................... 54
6 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 60
7 LIST OF TABLES Table page 2 1 Akaike Information Cr iteria (AIC) for model comparisons: (h) = a separate ................................ .................. 29 2 2 von Bertalanffy growth parameters estimated for all lionfish in the sampled population, males, and females. ................................ ................................ ......... 29 3 1 Taxa found in stomachs of lionfish from Little Cayman Island, with associated frequencies of occurrence, alimentary indices, and estimates of abundance.. ... 47 3 2 Comparison of the alimentary index of three different prey items for all lionfish and three different age classes, age 0, a ge 1 2 and age 3 4 lionfish. .... 48
8 LIST OF FIGURES Figure page 2 1 Photograph courtesy of the Cayman Islands Department of Tourism (2012). .... 29 2 2 Length frequency distribution of lionfish from Little Cayman Island in 2011. ...... 30 2 3 Length weight data for lionfish collected in 2011 from Little Cayman Island. y = total weight (g); x = total length (mm). ................................ ............................. 30 2 4 Mean index of completion ( 1 SE) for 1 5 year old lionfish from Little Cayman Island. ................................ ................................ ................................ ... 31 2 5 Otolith showing a light colored band resulting from treatment with oxytetracycline hydrochloride (OTC) 10 11 days before the otolith was processed and photographed ................................ ................................ ............ 31 2 6 Number of fish hatched per month. Hatch dates were calculated using daily ring counts from age 0 otoliths. ................................ ................................ .......... 32 2 7 Sagittal otoliths of similarly sized lionfish (245 mm TL and 254 mm TL respectively) ................................ ................................ ................................ ...... 32 2 8 Condition factors for a ged fish and unaged fish ................................ .................. 33 2 9 von Bertalanffy growth cur ves fit to raw size at age data ................................ ... 34 2 10 Age frequency distribution for lionfish from Little Cayman Island in 2011. .......... 35 3 1 Cumulative species curve for lionfish prey from Little Cayman Island. ............... 49 3 2 Proportion of lionfish diets containing fish or shri mp according to age. .............. 49 3 3 Typical stomach contents from younger lionfish and a mature lionfish .............. 50 3 4 Body depth of prey, ver tical gape and horizontal gape for lionfish of differing total lengths. ................................ ................................ ................................ ....... 50 3 5 Dry weights of diets versus dry weights of lionfish. ................................ ............. 51
9 Abstract of T hesis Prese nted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science AGE, GROWTH AND DIET OF LIONFISH ( PTEROI S SPP ) FROM LITTLE CAYMAN ISLAND B.W.I. By Morgan Edwards D ecember 2012 Chair : Thomas Frazer Major: Fisheries and Aquatic Sciences The introduction of lionfish ( Pterois spp. ) to the Western Atlantic Ocean and Caribbean Sea cause s concern among natural resource managers. I mpacts may include competition with nativ e species for prey and habitat, direct predation on native species, and altered native fish community structure Managing these impacts requires improved information on trophic biology and life history parameters used in models predicting removal rates tha t will reduce lionfish densities and impacts successfully This study validate d annuli and daily rings in sagittal otoliths, documented ages, estimated Brody growth coefficient s (K) and asymptotic maximum lengths (L ) and described diets for lionfish from Little Cayman Island Annual increment formation was confirmed with marginal increment analysis and daily ring deposition in juveniles was confirmed using an oxytetracycline hydrochloride marker S agittal otoliths (N = 499 ) from l ionfish collected weekly between January and December 2011 indicated that males were 0 5 years old and females were 0 3 years old. v on Bertalanffy equations estimated K as 0.42 for the population, 0.38 for males, and 0.57 for females Estimates of L were 349 mm for the population, 382 mm for males and 286 mm for females
10 Lionfish consumed over 50 species, and frequencies of occurrence and dry weights indicated that shrimp were an important component in diets of juveniles and adults became more pisci vorous Overall, t his study showed that lionfish exert predation pressure on a broad range of species, and their relatively rapid growth will pose challenges for removal programs.
11 CHAPTER 1 GENERAL INTRODUCTION I mpacts caused by invasive species can be w idespread and in certain cases devastating. Impacts can be ecological, economic, or social. Ecological impacts include : c ompetition with native species and predation on native species (Gordon 1998). Examples of e conomic impacts include damage to structure s reduced abundances of economically important native species, and decreased tourism (Bax et al. 2003). S ocial impacts also are possible, e.g., loss of jobs and detrimental effects on human health (Bax et al. 2003). In many cases, introduced species compou nd the effects of other a nthropogenic alterations to the natural environment. Habitats become less suitable for native species when altered, and added stress arises from new competitors. For example, the invasion of the Southeastern United States by fire a nts and subsequent negative impacts on native taxa may have been facilitated by previous anthropogenic alteration of habitats (Todd et al. 2008). Approximately 50,000 introduced species generate $137 billion dollars per year in economic and ecological dam ages in the United States alone (Pimentel et al. 2000). At least 241 species have become established in marine systems with some having significant impacts (Semmens et al. 2004). The North Pacific seastar ( Asterias amurensis ) now dominates biomass in Port Phillip Bay, Australia An invasive ctenophore in the Black Sea and an Asian clam in San Francisco have been implicated in the collapse of fisheries (Bax et al. 2003). Currently, the introduction and establishment of lionfish ( Pterois spp ) in the Western Atlantic Ocean, Caribbean Sea and Gulf of Mexico generates prominent concerns for natural resource managers in these regions
12 Two species of lionfish ( Pterois volitans and Pterois miles ) were introduced to Florida waters in the 1980s (Morris & Whitfield 2 009). They rapidly spread up the east coast of the United States and by 2000, lionfish were found as far north as Cape Hatteras, North Carolina (Schofield 2010). Lionfish are now found throughout the Caribbean Sea and Gulf of Mexico (Aguilar Perera & Tuz Sulub 2010). The major factor limiting their current distribution is believed to be lower temperature s at the higher latitudes (Kimball et al. 2004). Thus climate change and future increase in oceanic water temperatures may result in further range expansi on (Schofield 2010 ). Lionfish inhabit a wide variety of habitat s including coral reefs, seagrass es and mangrove s (Barbour et al. 2011, Darling et al. 2011 ). In addition to the use of multiple habitats, t heir broad environmental tolerance generalist diet rapid growth early sexual maturation frequent spawning, and high fecundity make them ideal invader s as evidenced by their rapid and widespread establishment and tendency to occur in high densities (Morris & Whitfield 2009). In the Bahamas for example, densities of lionfish have been reported to approach 400 fish per hectare (Green & Cte 2009). Lionfish reached densities of close to 600 fish per hectare four years after they were first reported off Little Cayman Island in 2008 (Frazer et al. 2012). Suc h densities have the potential to significantly reduce the biomass of native prey species. Albins and Hixon (2008), for example, found that lionfish on experimental patch reefs in the Bahamas reduced recruitment of native reef fish by ~79%. Management of i mpacts from lionfish will benefit from diet studies that determine which species are subject to the highest risk of predation and if those risks shift during ontogeny.
13 Current management consists of l ocalized efforts to control lionfish numbers at discrete locations (Albins & Hixon 2011). For example, Bermuda instituted a culling program in 2008 that provided training to those interested in aiding with lionfish removal (Morris et al. 2009). However, the effectiveness of removal efforts is cur rently in dispu te due in part, to the lack of accurate estimates for parameters characterizing growth in different regions of the invasion. Estimates of growth have been produced thus far only for North Carolina, the northernmost region in which lionfish are able to ove rwinter in the invaded range. As yet, there are no growth estimates for lionfish in the Caribbean Sea (Barbour et al. 2011) The objectives for this study were to fill gaps in information on age structure, growth and diets of lionfish in the Caribbean by ( 1) validating the periodicity of annual and daily otolith increments, (2) providing estimates of growth for lionfish from Little Cayman, (3) determining whether growth varied between males and females, and (4) characterizing food habits and determining the presence or absence of ontogenetic shifts in diet.
14 CHAPTER 2 AGE STRUCTURE AND GR OWTH OF LIONFISH OFF LITTLE CAYMAN ISLAND Introduction Models can be useful tools to assess the potential effectiveness of lionfish removal efforts. A s tage based population model (using larvae, juvenile, and adult stages) predicted that approximately 27% of the lionfish population would need to be removed on a monthly basis in order to curb in creasing abundance s (Morris et al. 2010) However, this estimate was surrounded by a broad range of uncertainty due to a lack of precise estimates for key life history parameters i.e., growth and natural mortality (Morris et al. 2010). Barbour et al. (2011) produced an age structured model for lionfish off the coast of North Carolina an d estimated that 35 65% of the population would need to be removed annually in order to induce recruitment overfishing. As part of their evaluation, Barbour et al. (2011) suggested that growth parameters for lionfish in warmer climates may differ from thos e estimated for lionfish from North Carolina, and that age and growth information should be collected for lionfish at lower latitudes. K ey demographic information for lionfish from different regions of the invaded range (temperate, subtropical, and tropica l) is necessary to inform management decisions about removal strategies. In addition to reducing the level of uncertainty pertaining to growth age based studies provide a first step toward estimating natural mortality due to its relationship with growth (Pauly 1980) Improved estimates of growth also lead to more accurate measures of yield per recruit (Lai & Gunderson 1987) and better estimates of whether the number of lion fish being removed is adequate to ind uce recruitment overfishing ( Barbour et al. 20 11)
15 Ages of fish often are estimated from annuli or rings that form in their sagittal otoliths. Growth can be modeled by fitting a von Bertalanffy growth equation to size at age data. In temperate and sub tropical areas, assigning ages to sagittal otolith s can be relatively straightforward because annuli typically are readily apparent. In t ropical regions assigning ages can be more difficult as opaque and hyaline bands in otoliths are less evident as a consequence of reduced seasonal variation in temperatu re and/or photoperiod (Green et al. 2009 ). For example, a change in water temperature of 3C was enough to promote the form ation of distinguishable rings in pomacentrids from the Western Atlantic but, otoliths from the same taxa collected near the equator were more difficult to read (Caldow and Wellington 2003) In spite of the potential challenges of using otoliths to age fish in the tropics, t his study relies on ages read from sagittal otoliths, total lengths and von Bertalanffy growth equations to estima te parameters characterizing growth that will i mprove models predicting the percentage of Caribbean lionfish population s that would need to be removed in order to decrease local abundance s Thus, the objectives of this study were to (1) validate the annual periodicity of otolith increment formation and the rate of daily ring deposition (2) provide estimates of growth for lionfish from Little Cayman Island and (3) determine if growth varied significantly between males and females. Study Site Little Cayman Island Figure 2 1). The island is home to approximately 150 permanent residents. Human impacts on the surrounding waters are minimal. In addition to low levels of anthropogenic stress, 20% of the reefs s urrounding Little Cayman are protected in no take marine reserves (Coelho & Manfrino 2007)
16 Methods Fish Collection and Processing The majority ( 93 %) of lionfish collected for this study were removed from reef walls in Bloody Bay Marine Park during commun ity culling events. Divers donated the fish for research purposes. Culls occurred on a weekly basis from January to December 2011 at approximately 16:30, until day length increased, at which point collection time shifted to 17:00. Lionfish feeding activity increases during crepuscular periods, which makes dusk a suitable period for collecting (Green et al. 2011). Collectors used Hawaiian sling style spears and SCUBA with fish placed into a sealed bucket for transport back to the boat. C ollection s generally occurred between 15 and 30 m The efficacy of spears decreased when fish were less than ~ 75 mm TL; therefore smaller size classes typically were collected separately using nets. Upon return to the dock, fish were placed on ice and transported to the Litt le Cayman Research Centre, where processing occurred the next day. Processing yielded sizes and sexes of lionfish, along with otoliths for use in estimating ages. Total length to the nearest 1 mm and weight to the nearest 0.1 g were recorded for each fish. Sexes were determined macroscopically and sagittal otoliths were removed. Otoliths were rinsed in fresh water, dried, and stored in numbered plastic vials that unique ly identified each fish Otolith Processing and Counts of Annuli Whole sagittae were sub mersed in freshwater on a black background and read under reflected light using a dissecting micro scope at 40 60x magnification Both the right and left sagittae were viewed in a search for the most distinct and readable annuli An age in years was assign ed according to the number of opaque zones present. To
17 ascertain within reader agreement, one reader analyzed all otoliths twice with the two readings separated by several months A second experienced reader examined the otoliths for which the two indepen dent readings agreed. From these two sets of reads, a percentage of agreement between reader s was estimated The opinion of a third experienced reader settled disagreements between the other two readers. To test for possible improvement in estimates of age 250 otoliths were sectioned to compare with readings from whole otoliths Sections proved to be more difficult to read than whole otoliths; therefore, these data were not used in subsequent analys e s Marginal Increment Analysis M arginal increment analysi s was employed to validate production of one annulus per year for lionfish off Little Cayman Island, and to also determine the timing of increment formation. These analyses were based on an index of completion (Eq uation 2 1): C = W n /W n 1 ( 2 1) where C is t he index of completion, W n represents the width of the marginal increment and W n 1 represents the width of the preceding complete annulus (Murie and Parkyn 2005). M easurements of W n and W n 1 were made from photographs of whole otoliths using ImageJ softwa re (Rasband 2012) All age classes were combined for this analysis and mean indices of completion were plotted against months of collection; the presence of a single minimum, was interpreted as evidence of annuli (Murie and Parkyn 2005). Counts and Valid ation of Daily Ring s T o improve modeling of growth by assigning ages based on partial years and to determine an approximate time of spawning otoliths of age 0 fish were prepared for
18 analysis of daily rings. Histamount was used to fix the otoliths to glass slides. The otoliths were then wet ground and polished until the nucleus plane was reached and daily rings became visible. I did not attempt to count daily rings for fish older than ~110 days due to decreasing accuracy (Sweatman & Kohler 1991). Otoliths w ere viewed through a compound microscope at 100 200x magnification. Two counts of daily increments separated by four weeks were performed by a single reader. The date of spawning was calculated by subtracting the daily ring count from the date when the fis h was collected, and an age as a fraction of a year, was assigned to the fish by dividing the number of daily rings by 365. Average daily growth rates were calculated by dividing the standard length by the daily ring count in order to estimate mm of growth d 1 Five, age 0 lionfish were used to validate daily increment formation by immersing them in 500 mg l 1 oxytetracycline hydrochloride (OTC) for 24 h. Ten days after exposure, otoliths were prepared as described for reading of daily rings. The fluoresce nt band produced by OTC exposure and the number of daily rings produced subsequently were identified at 400x magnification using a compound microscope equipped with a blue filter. Modeling Fish Growth Growth was modeled with the von Bertalanffy growth equa tion (Equation 2 2): L t = L [1 e K(t t 0 ) ] (2 2) where L t = length at age (t), L = asymptotic maximum length, K = the Brody growth coefficient (which describes the rate of growth toward L (Newman et al. 1996)) and t 0 = the theoretical time at which t he fish would be 0 mm in length (von Bertalanffy 1957). A
19 maximum likelihood framework was used to estimate von Bertalanffy parameters, L K, and t 0 for the population. Growth parameters also were calculated for males and females separately, with one hal f of the juvenile, unsexed lionfish randomly allocated to each sex. In both cases, t 0 was assumed to be the same as that for the complete sample population Akaike information criteria (AIC) were used to determine the most appropriate model for the data. I compared four different hypotheses: (1) separate parameter estimates for males and females should be used to model the lionfish population in Little Cayman, allowing all parameters (L K, and sigma, the variance around mean size at age) to vary; (2) one set of parameter estimates for the population should be used in models, assigning the same values to all parameters for males and females, (3) the same estimate of K should be used in both models, allowing the other parameters to vary independently, and (4) the same estimate of L should be used in both models, allowing the other parameters to vary independently. Since t 0 is constrained in order to produce the separate estimates for mal es and females, it was not included in the AIC analysis. Models with AIC values that differed by < 2 from the model with independent estimates for all three parameters were considered potentially appropriate replacements, differences of 4 7 indicated less support for the alternative model, and differences >10 indicated no support for the alternative model (Burnham & Anderson 2004). Results Lionfish Size Structure and Length Weight Relationship From January to December 2011, 1979 lionfish were collected from the waters surrounding Little Cayman Island Fish ranged from 27 to 391 mm in total length (Figure
20 2 2) and total length accounted for approximately 97% of the variation in total weight (Figure 2 3). Marginal Increment Analysis Marginal increment analysis confirmed annual increment formation. Increment width increased from November until April and then gradually decreased throughout the rest of the year (Figure 2 4 ). On average, annulus formation was least complete in October at which point formation was considered to begin. The gradual change between maximum and minimum indices of completion likely reflects the effects of relatively constant photoperiod and water temperatures in the tropics, as contrasted with more disjunct data from other fishes in tempe rate areas (e.g., Murie and Parkyn 2005). Analysis of Daily Rings Otoliths treated with OTC exhibited a fluorescent mark under a blue light. Beyond the OTC band, an additional 10 11 d of growth was noted, confirming the formation of daily increments (Figur e 2 5). Daily ring s were counted for 29 fish ranging from 27 to 199 mm total length which translated to ages of 23 to 110 days. Growth rates calculated using standard lengths ranged from 0 .73 to 1.70 mm per day with an average of 1.03 mm p er day. Back cal culated dates for hatching provided evidence that lionfish recruit to Little Cayman Island year round (Figure 2 6) Estimates of Growth Rates Of the 1979 fish collected, 1188 pairs of otoliths were read, and of those 499 were successfully aged by counting the number of opaque rings. Otolith clarity varied significantly from fish to fish and only 42% of otoliths pro duced discernible rings (Figure 2 7a and b). Of the fish yielding readable otoliths, 238 could be sexed, with 110 being
21 males and 128 being fema les. This approximate 1:1 ratio was consistent with that observed for all sexed lionfish collected off Little Cayman Island in 2011(i.e., 533 females and 563 males). To discern the condition of lionfish that produced interpretable rings and those that did not, condition factors (K n ) were calculated as follows (Equation 2 3): K n = W o /aL b (2 3) where W o is the observed weight, L is the total length, and a and b are constants based on the length weight relationship y = aL b The average condition factors for t he aged fish and those that were not able to be aged were 1.15 and 1.17 respectively (Figure 2 8). G rowth curve s were fit to data for all fish and data for males and females (Figure 2 9a, b and c) C omparison of AIC values supported the use of a model wher e all parameters varied independently (Table 2 1). Females grew 1.5 faster than males and reached a maximum length that was only 0.75 that of males ( Table 2 2) In all cases, rapid initial growth decreased when fish reached two years old (Figure 2 9a, b and c) For both males and females, there was a large overlap in size for fish that were two years old and older, which suggested growth had begun to asymptote (Figure 2 9b and c). In combination with the fact that two year old fish represented the largest age class in this population, this result also indicated the presence of a strong year class in 2009 (Figure 2 10 ) Discussion This study provides the first insight s into the age structure of the invasive lionfish population established off Little Cayman Island the first successful validation of daily ring deposition for lionfish, and the first estimates of key metrics needed to model
22 growth of lionfish in the Caribbean. Such information can be used to improve management of this invasion and reduce impac ts on Caribbean coral reefs No effort was made to differentiate between P. miles and P. volitans in this study The morphologies of the two species are so similar that g enetic analys e s are needed to differentiate them. In addition, available data indicate that relatively low numbers of P. miles are found throughout the invaded range (Hamner et al. 2007) and that P. miles and P. volitans are physiologically similar (Hare & Whitfield 2003). Annuli in sagittal otoliths indicated that t he population of lionf ish off Little Cayman Island comprised 0 5 year old males and 0 3 year old fe males. In general, these results match the reported timeline of the invasion; because lionfish were first reported from Little Cayman in 2008 (Schofield 2009). The single 5 year o ld male provided evidence that lionfish arrived before 2008, but remained undetected Introduced species often remain undetected until their densities increase. For example, pythons in the Everglades probably escaped detection initially as a consequence of low densities but after densities increased and the population became established, their presence became known (Wilson et al. 2011). Although annuli in otoliths from tropical fish can be difficult to interpret due, in large part, to the narrow range of temperature experienced and less pronounced seasonality (Caldow & Wellington 2003, Green et al. 2009, Marriott & Mapstone 2006) whole otoliths provided useful estimates of age for lionfish from Little Cayman Island and marginal increment analysis validat ed an annual deposition rate for rings I found that the visibility of annual rings in ot oliths varied from fish to fish, but approximately 50% of the otoliths examined supported age determination with sufficiently distinct opaque and
23 hyaline bands. As fou nd by Vilizzi and Walker (1999) for common carp sectioning did not increase interpretability The inability to age approximately half of the lionfish otoliths collected in this study may be a consequence of differential growth. Hoyer et al. (1985) discus s the issues associated with using whole reads to age fish that grow slower than average. Hoyer and colleagues suggest that slower growing fish are more difficult to read due to crowding of annuli. It may be the case that the lionfish otoliths producing di scernible rings came from fish experiencing faster growth. Should this be true, the resulting estimates of growth may be biased towards the faster growing individuals of the population. C ondition factors calculated for the two groups of fish in this study, i.e., fish that could be aged and those that could not be aged, provide no evidence for differential growth. Daily ageing validated by incorporation of oxytetracycline hydrochloride improved estimates of growth. Results also provided evidence of y ear roun d spawning consistent with the findings reported by Morris & Whitfield (2009 ). Combining ages in days with sizes yielded an average daily growth rate of 1.03 mm d 1 This estimate exceeds the average daily growth rate of 0.46 mm d 1 reported by Jud and Lay man (2012) for lionfish from the lower Loxahatchee River estuary, Florida. Standard length was used in this portion of the analysis in order to make comparisons between this study and Jud and Layman (2012).The size ranges of fish used to generate the two i ndependent estimates overlapped (21 156 mm standard length here and 45 185 mm for Jud & Layman 2012), although Jud and Layman (2012) used a mark recapture technique to measure growth It is likely, however, that other factors, e.g., temperature, salinity a nd/or dissolved oxygen concentration, played a more important role in generating the growth differences
24 between juvenile lionfish from the two study locations. Additional factors, such as diet also may have played a role (Baltz et al. 1998, Moyle & Light 1 996). Models based on combinations of ages and sizes yielded estimates of asymptotic maximum lengths (L ) and Brody growth coefficients (K) for the population and separately for males and females. The L values for all lionfish, 349 mm, males, 382 mm, and females, 286 mm, were smaller than the 455 mm calculated by Potts et al. (2010) and 425 mm calculated b y Barbour et al. (2011) for lionfish from North Carolina. The K values for the population, 0.42, males, 0.38, and females, 0.57, were nearly as high or higher than the values of 0.32 and 0.47 calculated for lionfish from North Carolina (Barbour et al. 2011 Potts et al. 2010). Thus, lionfish from Little Cayman Island appear to grow as fast or faster than those from North Carolina, but they may not reach the same maximum size. Similarly, female lionfish from Little Cayman Island grew faster than males, but t hey did not reach the same maximum size. Faster growth for a tropical population is not unexpected given the warmer water temperatures, with Beamesderfer and North (1995), for example, reporting faster growth rates for largemouth bass in southern waters as compared to northern populations. Both populations of lionfish appear to grow faster than other native, medium bodied predators in the Caribbean. For example, growth coefficient s estimated for yellowtail snapper and graysby grouper are approximately 0.1 0 and 0.35, respectively ( Froese & Pauly 2012 ). Eventually, the relatively rapid growth exhibited by lionfish may be moderated by other biotic and abiotic factors, such as food availability and density dependence (Rogers & Ruggerone 1993, Rose 2000).
25 Estimat es of L and K may have been affected by the fact that the relatively recently established population of lionfish off Little Cayman Island lacks older and larger fish, which results in a truncated growth curve. Potts et al. (2010) found that the maximum size of l ionfish taken from a 10 year old population of lionfish off North Carolina was 464 mm TL and the maximum age was 8 years old, with these fish representing larger and older fish than the 391 mm TL and 5 year old fish from Little Cayman Island. Nevertheless, growth appears to begin to asymptote at age 5 in the North Carolina population (Barbour et al. 2011), which is the oldest age class found off Little Cayman Island. For this reason, I believe that the most critical portion of the curve is represented in th e data from Little Cayman Island. To explore issues associated with modeling an immature population, I truncated the size at age data from the North Carolina study and examined the resulting effects on L and K. Differences in the estimates produced by mod els that incorporated a consistent t 0 (using the value of 1.01 calculated for the Little Cayman Island population) were minor, with the 0 8 age range yielding an L of 483 mm TL and a K of 0.30 and the 0 5 age range yielding an L of 495 mm TL and a K of 0.29. From this information, I can infer that key parameters describing growth should not change considerably as the population matures. Based on the reduced frequency of fish in the 60 to 120 mm TL size class, it is not unreasonable to assume that strong size dependent vulnerability exists when using spears to capture lionfish. An underrepresentation of smaller size classes can result in negative bias for estimates of t 0 and K. Gwinn et al. (2010) evaluated methods to reduce bias in growth parameters due t o size selective sampling. They indicated that understanding the sampling vulnerability schedule of the species for a particular gear
26 type is necessary in order to determine the best method to use to correct gear bias. However, a vulnerability schedule for lionfish and spearing was not evaluated in this study. Thus, corrective measures to resolve issues associated with gear bias remain unknown. Attempting to solve for von Bertalanffy parameters separately for males and females produced unreasonable estimat es when t 0 was allowed to vary based on the data for each sex. Therefore, t 0 was constrained to the value calculated for the whole population when estimating K and L for the separate sexes. Vilizzi and Walker (1999) also employed this method to produce se x specific growth and maximum length parameters for similar reasons. However, fixing t 0 inevitably affects estimates of L and K. This study of Caribbean lionfish provided important information not included in Potts et al. (2010) and Barbour et al. (2011) studies. The lionfish population off Little Cayman Island exhibited sexually dimorphic growth, with males attaining larger sizes than females. Females, however, achieved their asymptotic length at a faster rate than males. Sexually dimorphic growth is a re latively common phenomenon in fish populations and has been reported for a number of taxa including snappers and white grunt (Murie & Parkyn 2005, Newman et al. 1996). In addition, Mesa et al. (2005) documented sexually dimorphic growth for Scorpaena mader ensis a Mediterranean scorpaenid with males attaining a larger size and having slightly slower growth than females. The growth coefficients and values of L for males and females of this species equaled 0.23 and 0.44 and 96 mm and 148 mm, respectively W it h growth coefficients of 0.38 for males and 0.57 for females, lionfish off Little Cayman Island grow at a faster
27 rate than the Mediterranean scorpaenid. Data from larger numbers of 0 1 year old fish from Little Cayman Island may increase the accuracy of growth parameters estimate d for males and females The relatively fast and sexually dimorphic growth of lionfish combined with evidence of year round spawning generate several implications for managers attempting to reduce the impacts of this invasive spec ies. F or example, f requent spawning will likely complicate efforts to control lionfish at broad spatial scales due to a relatively constant supply of recruits. In fact, t he most sustainable fisheries are based on fish populati ons that exhibit fast growth, high natural mortality and high fecundity ( Schindler et al. 2002 ) with restrictions placed on harvests of fish exhibiting slower growth in order to keep those fisheries from collapsing (Bennet et al. 1991). Unfortunately for those working to reduc e dens ities of lionfish their relatively fast growth translates in to more effort being necessary to achieve that goal Management implications with regards to sexually dimorphic growth typically concern spawning stock biomass (Marshall et al. 2005) which depen ds on the number of females present in the population and their fecundity. For some fish populations that exhibit sexually dimorphic growth, fishing pressure may skew sex ratios by removing the larger individuals. If the larger individuals are female s the spawning stock biomass could decrease drastically leading to a collapse of the fishery For lionfish, the pattern of s exually dimorphic growth could lead to removal of more males because spearfishers have been shown to remove larger lionfish first (Frazer et al. 2012) If spearing does not remove a sufficient number of females, then removal efforts may be compromised. In this study, t he sex ratio
28 produced by spear fisher s was approximately 1:1 but managers should remain vigilant to ensure the ratio does no t change.
29 Table 2 1 A kaike I nformation C riteria (AIC) for m odel c omparison s: (h) = a separate parameter for males and females; = parameter shared by both sexes; = change in Akaike Information Criterion Table 2 2 v on Bertalanffy growth par ameters estimated for all lionfish in the sampled population males, and females. Group L (mm) K t 0 N Range of a ges Population 349 0.42 1.01 499 0 5 Males 3 82 0. 38 1.01 110 0 5 Females 2 86 0. 57 1.01 128 0 3 Figure 2 1. Little Cayman Island an d the dive sites ( ) where lionfish were collected. Photograph courtesy of the Cayman Islands D epartment of Tourism (2012 ).
30 Figure 2 2. Length frequency distribution of lionfish from Little Cayman Island in 2011 Figure 2 3 Length weight data fo r lionfish collected in 2011 from Little Cayman Island. y = total weight (g); x = total length (mm).
31 Figure 2 4. Mean index of completion ( 1 SE) for 1 5 year old l ionfish from Little Cayman Island Figure 2 5. Otolith showing a light colored ban d resulting from treatment with oxytetracycline hydrochloride (OTC) 10 11 days before the otolith was processed and photographed The hatch marks indicate the number of rings produced after treatment.
32 Figure 2 6. Number of fish hatched per month. Hat ch dates were calculated using daily ring counts from age 0 otoliths. (a) (b) Figure 2 7 Sagittal otolith s of similarly sized lionfish (245 mm TL and 254 mm TL respectively) ( a) shows t wo clearly visible opaque rings and (b) displays no visible rings.
33 Figure 2 8. Condition factors for aged fish and unaged fish.
34 (a) (b) (c) Figure 2 9 von Bertalanffy growth curve s fit to raw size at age data (a) von Bertalanffy curve for all lionfish, (b) for females and (c) for males
35 Figure 2 10. Age frequency distribution for lionfish from Little Cayman Island in 2011.
36 CHAPTER 3 DIET OF LIONFISH OFF LITT LE CAYMAN ISLAND Introduction Introduced aquatic species have the potential to alter the way in which ecosystems function (Pimentel et al. 200 0). From an ecological viewpoint, the most destructive of the problematic invaders disrupt community structure and change food webs (Molnar et al. 2008). Certain high profile cases, such as the addition of Nile perch ( Lates niloticus ) to Lake Victoria, exe mplify such disruptions. Nile perch were implicated in the extinction of ~65% of the endemic haplochromine cichlids in Lake Victoria. This reduction in planktivores may have promoted algal blooms occurring in that system (Goldschmidt et al. 1993). Other e xamples include the carp ( Cyprinus carpio ) in Australia and the peacock hind grouper ( Cephalopholis argus ) in Hawaii; both became the dominant fish biomass in some areas (Dierking et al. 2009, Koehn 2004). Carp remove aquatic vegetation and in the process, increase turbidity. This leads to a reduction in light availability with consequences for photosynthesis and native fish that feed visually (Koehn 2004). The peacock hind feeds on a broad range of prey species and not just those that are locally abundant (Dierking et al. 2009). In general, understanding the threats associated with introduced species is paramount to sustaining our native biota (Cucherousset & Olden 2011). The introduction of Indo Pacific lionfish to the Western Atlantic, Caribbean and Gulf of Mexico provides such a challenge. Lionfish are considered to be voracious, generalist predators with the capacity to reduce recruitment of other reef fish on experimental patch reefs by 79% (Albins & Hixon 2008). Novel predation strategies, such as blow ing directed jets of water at their
37 prey, make them extremely efficient predators on nave prey (Albins & Lyons 2012). For example, Ct and h 1 for lionfish. Predation at this rate may further reduce densities of Caribbean reef fish, which are already declining due to overexploitation and habitat degradation (Paddack et al. 20 09). In fact, lionfish densities have reached potentially unsustainable levels in certain areas of the invaded range. Green and Ct (2009) state that lionfish populations on the majority of 12 Bahamian coral reefs could not be sustained given the density of available prey. In addition, c onsumption of economically important species such as grouper and yellowtail snapper ha s been observed; however, these species have not comprise d a significant portion of lionfish diet s (Morris & Akins 2009). Species that do comprise a significant portion of lionfish diets, such as small bodied and juvenile reef fish, render ecological services necessary for the health of coral reef ecosystems. Lionfish may reduce populations of native herbivores needed to control algal over growth of coral ( Morris & Whitfield 2009, Morris 2009). Evidence of such a phase shift was noted on mesophotic reefs in the Bahamas, with the abundance of herbivores decreasing after the introduction of lionfish and a subsequent shift to increased algal co ver (Lesser & Slattery 2011). As these changes could not be explained by other factors, e.g., disease, overfishing, or damage from hurricanes, lionfish were implicated as the primary cause (Lesser & Slattery 2011). Overall, impacts from the lionfish invas ion will vary depending on local fish communities, the density of lionfish, and the system being invaded (Morris et al. 2008). Diets of lionfish have been characterized in the Bahamas (Green et al. 2011, Morris & Akins 2009), but such work has yet to be do ne in the Caribbean. Information regarding
38 predation on native species should be collected in various regions of the invaded range to better inform bioenergetic and trophic models that predict consumption rates for lionfish (Cerino 2010). With the informat ion currently available for environmental conditions and observed growth, Cerino (2010) predicted daily consumption rates of 2.186 kg prey d 1 for an area with densities of 393 lionfish ha 1 In an effort to improve understand ing of impacts from the invasi on off Little Cayman Island the objectives for this study were to (1) examine stomach content s of lionfish to identify the diversity of prey, (2) determine the apparent importance of prey items (3) determine whether differences existed in average weight between male and female diet items, and (4) determine whether lionfish display an ontogenetic shift in prey consumption. Methods Collection and Processing The majority ( 93 %) of lionfish collected for this study were removed from reef wall habitat in Bloo dy Bay Marine Park during community culling events. Culls occurred on a weekly basis from January to December 2011 at approximately 16:30, until day length increased, at which point culls began at 17:00. Lionfish feeding activity increases at dawn and dusk making the early evening hours suitable for finding stomachs containing identifiable prey (Green et al. 2011). Collection s were performed by SCUBA divers using Hawaiian sling style spears and captured fish were placed into sealed bucket s for transport. Depth s of collection s ranged from approximately 15 to 30 m and dive time s ranged from 30 to 45 min After docking fish were placed on ice and transported to the Little Cayman Research Centre, where processing occurred the next day. I recorded the total length, total weight, gape, and sex of each fish before removing stomachs. Stomachs were placed in plastic bags with 95% ethanol to stop
39 digestion and degradation of prey They were later dissected, and any contents were classified to the lowest possible t axonomic level using a dissecting microscope at 40 60x magnification and Humann and De L oach (2002) as an identification guide. Stomach contents were then placed into separate pre weighed tin s and dried in an oven at 24C for three days prior to recording t heir dry weight. Calculation of Alimentary Indices To determine the importance of various prey items, I calculated alimentary indices. These indices were based on data from stomachs containing a single type of prey. The index of choice (Equation 3 1) was p reviously used by Valentim et al. (2008): IA i = (3 1) where IA i = the relative importance of food item i expressed on a scale of 1 to 100, %FO i = the relative commonness of item i expressed as a percentage occurrence in the total number of st omachs examined, and W i% = the relative biomass contributed by item i expressed as 100 times the sum of the dry weights for the item divided by the total dry weights of all items. For example, creole wrasse occurred in 68 lionfish stomachs. To determine %F O, divide 68 by the sum of all diet item occurrences and multiply by 100. Similarly, the sum of the creole wrasse diet items equaled 33 g dry weight. To determine the %W, divide 33 by the sum of all diet item dry weights and multiply by 100. Finally, to de termine %IAi, take the value of %FO creole wrasse x %W creole wrasse and divide that by the sum of the %FO x %W for all diet items and multiply by 100. Combining frequency of occurrence and gravimetric data provided the opportunity to make unbiased inference s about the importance of different prey species. Indices based solely on weight can overestimate the importance of single, heavy prey items.
40 Frequency of occurrence was employed to avoid misrepresentation associated with attempts to discern the number of individual prey items in highly digested stomach contents (Hyslop 1980). To avoid overestimating the importance of numerous, smaller prey items, such as shrimp, dry weight was factored into the estimates of importance to capture potential caloric contribut ions. A cumulative prey curve was used to assess how fully the diet of lionfish was characterized (PRIMER 6; Anderson et al. 2008). The Chao 1 index was applied to counts of prey items in order to capture the influence of rare species. A thousand permutati ons of samples yielded estimates of the expected number of prey items in a fully characterized diet. Results General Observations In total, 1599 stomachs were analyzed, and of tho se, 475 stomachs were empty (~30 %). Of the stomachs containing prey, 50% held fish tissue, 25% held shrimp tissue, and 25% comprised unidentifiable items, crabs, bivalves or plant matter. The average dry wei ght of male diets (0.21 g) was twice that of female diet s (0.10 g). This difference was due to males eating both more prey (a mean of 1.3 items for males versus a mean of 0.7 items for females) and larger prey (a mean total length of 28.1 mm for prey of males versus a mean total length of 10.1 mm for prey of females). Over 50 species of fish and crustaceans were observed in the d iets of lionfish suggesting a generalist, carnivorous diet. The cumulative prey curve suggested an additional 25 species eventually may be found in lionfish diets. The cumulative prey curve began to asymptote around sample number 400 at 75 unique prey item s (Figure 3 1). Thus, 1599
41 samples characterized approximately two thirds of the diet for lionfish from Little Cayman Island. Alimentary Index Although the complete diet of lionfish from Little Cayman included over 50 species, s everal species dominated ac cording to alimentary indices (Table 3 1). In total, ~33 g (dry weight) of c reole wrasse ( Clepticus parrae ) were found in 68 stomachs ~7 g (dry weight) of fairy basslets ( Gramma loreto ) were found in 86 stomachs and ~8 g (dry weight) of red night shrimp ( Cinetorhynchus manningi ) were found in 138 stomachs. Although red night shrimp had the highest frequency of occurrence, creole wrasse ranked highest in overall importance of identified items after weights were considered (Table 3 1). These data suggested that creole wrasse contribute the most calories to lionfish diets off Little Cayman Island. Age data linked to stomach contents via a unique identifier for each was used to identify ontogenetic shifts in diets. A s lionfish aged, the proportion of shrimp f ound in their diets decreased steadily; thus, older lionfish became more piscivorous (Figure s 3 2; 3 3a and b). In fact, the alimentary index for red night shrimp, creole wrasse, and fairy basslets changed depending on the age of the fish. For example, red night shrimp had an alimentary index of 7.97 for age 0 lionfish, 1.4635 for lionfish 1 2 years old, and 0.4308 for lionfish 3 4 years old. Thus, the importance of red night shrimp in lionfish diets decreased as they aged. In contrast, creole wrasse had an alimentary index of 0.0000 for age 0 lionfish, 4.4775 for lionfish 1 2 years old, and 34.0980 for lionfish 3 4 years old. Therefore, the importance of creole wrasse in lionfish diets increased as lionfish aged. Lastly, alimentary index for fairy basslets was highest for lionfish that were 1 2 years old (Table 3 2).
42 Lionfish gape size increases as body size increases, allowing lionfish to eat a wider variety of prey of different sizes. Increased lionfish size does not, however, preclude them from continuing to eat smaller prey (Figure 3 4). In fact, all prey body depths fell below the gape measurement of the lionfish that ate them, and prey depth ranged from 5 to 77% of the gape size (Figure 3 4). The relationship between diet weight and lionfish weight show ed that lionfish ate up to 9% of their body weights (Fig ure 3 5 ). Discussion Managing and mitigating impacts of invasive lionfish will benefit from knowledge of their effects on local prey fish assemblages. For example, an understanding of prey selection c an help managers develop metrics of success for removal programs. Arrington et al. (2002) found that the percentage of empty stomachs can be associated with trophic status and, to some degree, g eographic location. Piscivores that consume whole prey, such a s lionfish, and fish that provide parental care had the highest overall percentage of empty stomachs. When geographic location was taken into account, Arrington et al. (2002) found that the proportion of empty stomachs in Central and S outh America was just under 30%. Results from Little Cayman Island corroborate these findings The large percentage of highly digested material may be due to the timing of feeding, with fish for this study collected in the early evening so any prey items consumed in the mornin g would have been at least partially digested (Morris & Akins 2009) Male diets weighed twice that of females on average. This difference is most likely due to males attaining larger body sizes than females. The largest male lionfish was 391 mm TL, and th e largest female was 333 mm TL. In his study of bioenergetics and
43 trophic impacts of lionfish, Cerino (2010) found that males gained twice as much mass and consumed 21% more prey than females. He suggested that reduced somatic growth for females may be due to a reallocation of energy stores to gamete production. Cerino (2010) also stated that lionfish consumed the most prey at 30C; therefore, consumption probably will vary in temperate regions and remain higher and more constant in the tropics. Lionfish fr om Little Cayman Island consumed similar proportio n s of fish and crustaceans as lionfish from the Bahamian a rchipelago (Morris and Akins 2009) Lionfish from both locations exhibited an ontogenetic shift from shrimp to fish In this study, the shift was li nked to age classes As lionfish control efforts progress, the older and larger individuals in the population will likely be fished down first, truncating the size distribution (Berkeley et al. 2004, Frazer et al. 2012). A population comprised of smaller i ndividuals may exert less pressure on native fish populations due to their propensity to consume smaller taxa comprised of shrimp (Frazer et al. 2012). A similar ontogenetic shift is exhibited by native groupers, such as the Nassau grouper, which raises co ncerns about competition with lionfish. In a study done by Eggleston et al. (1988), Nassau grouper were shown to eat primarily crabs and other crustaceans in the juvenile stage and primarily fish in the adult stage. Crabs were not a significant part of juv enile or adult lionfish diets; however, competition may exist between adult lionfish and early adult grouper that feed on fish of similar sizes. Muoz et al. (2011) found lionfish from hard bottom reefs in the southeastern U.S. generally were piscivorous and p rey items were consumed in proportion to their local abundance Although we did not survey prey populations, we can infer that lionfish in
44 the Caribbean consume prey in the same manner Results from surveys conducted by others (S. Barry unpublished d ata) ranked creole wrasse and fairy basslets (two common prey items for lionfish) as the second and fourth most common species out of 50 species, which confirms the premise that lionfish prey on locally abundant species from Little Cayman Island In addit ion, qualitative abundance estimates for species consumed by lionfish in the Caribbean (Humann & DeLoach 2002) indicate that red night shrimp, creole wrasse, and fairly basslets are abundant. Species occurring less frequently in lionfish stomachs generally were described as common to occasional. While these qualitative estimates are not based solely on assemblages off Little Cayman Island, they do describe what is available in the Caribbean. Morris and Akins (2009) further corroborate this pattern, with res ults showing that lionfish in their study fed readily on abundant teleost fishes and crevice dwelling species. Red night shrimp and fairy basslets both inhabit reef crevices, and creole wrasse often are found on outer reefs and drop offs (Humann and DeLoac h 2002) along with lionfish. The abundances of these species and the habitat overlap make them especially vulnerable to predation by lionfish. Increases in lionfish gape as they grow in length allow them to eat a wider variety and wider size range of prey. Sharf et al. (2000) noted that larger predator gape does not necessarily mean that the predator will consume only larger prey items. Asymptotic predator size prey size distributions exist due to larger predators continuing to eat small prey in conjunction with larger prey. St John (1999) found such a relationship between predator gape and prey depth for the grouper, Plectropomus leopardus
45 Lionfish diets weighed up to 9% of their dry body weights. Estimates of consumption by lionfish based on observations of foraging at dusk and an assumed 2 h of foraging at that rate per day yielded a mean consumption standard error of approximately 9 5% for lionfish in the Bahamas (Green et al. 2011). Lionfish captured near dusk off Little Cayman Island did not achie ve this level of consumption in 95% of 1599 cases, which may indicate they had not foraged for 2 h. Nevertheless, all available data indicate that lionfish can consume considerable amounts of food. Competition for food continues to be one of the perceived impacts frequently associated with the lionfish invasion (Barbour et al. 2011, Morris & Whitfield 2009). Albins (2012) attempted to provide a causal link between the presence of lionfish on experimental patch reefs, decreased abundance of native prey, decr eased species richness for prey, and decreased fitness for a similar sized native piscivore, the coney grouper. By tracking prey assemblages and measuring growth of the two predators added to the experimental reefs in combination and separately, he determi ned that lionfish growth surpassed that of the coney grouper in all cases. Although the presence of lionfish did not appear to inhibit growth of coney grouper over an 8 week period, Albins (2012) stated that perhaps coney grouper sharing reefs with lionfis h were forced to consume subpar prey, which may negatively impact growth in long run. In addition, the more rapid growth of lionfish may lead to direct effects through predation on coney grouper and other native piscivores that recruited in the same cohort This study provides insight into the feeding habits of lionfish in the Caribbean. Seasonal bias should not be present in our diet analysis since fish were collected every week for a year. Feeding habits and important prey species most likely will change with
46 the availability of prey, and diet studies should be expanded throughout the invaded area. It also is important to monitor changes in diet over time because such changes may indic ate changes in the availability of prey in the invaded range In additi on, prey composition may change in areas with targeted culling. Surveys of potential prey would complement diet studies and provide a more direct assessment of availability.
47 Table 3 1. Taxa found in stomachs of lionfish from Little Cayman Island, with a ssociated frequencies of occurrence, alimentary indices, and estimates of abundance. Abundance = qualitative abundance in the Caribbean according to Humann and DeLoach (2002); Rank = rank of mean abundances from surveys of three sites with few lionfish off Little Cayman Island (S. Barry, unpublished data). Taxon Frequency %IA i Abundance Rank Unknown fish 595 85.4834 Clepticus parrae 68 7.66047 Abundant 2 Unknown shrimp 382 4.08280 Cinetorhynchus manningi 138 1.42562 Abundant Gramma loreto 86 0.54014 Abundant 4 Chromis multilineata 7 0.23862 Abundant 31 Holocentridae 11 0.13167 Common Chromis cyanea 11 0.12204 Abundant 3 Thalassoma bifasciatum 15 0.05066 Abundant 7 Serranidae 13 0.04737 Common to occasional Synodus synodus 4 0.03935 O ccasional Malacoctenus triangulatus 32 0.03162 Common to occasional Scaridae 9 0.02826 Common to occasional Neogonodactylus oerstedii 17 0.02222 Occasional Liopropoma rubre 19 0.01925 Common 46 Apogonidae 20 0.01778 Common Aulostomus macula tus 7 0.01532 Common Monacanthus tuckeri 13 0.01248 Occasional Labrisomidae 12 0.01077 Common to occasional Epinephelus cruentatus 1 0.01000 Common 39 Halichoeres garnoti 12 0.00902 Common 13 Sparisoma viride 4 0.00673 Common to occasional 11 A pogon townsendi 7 0.00533 Common Mysidae 6 0.00493 Stegates partitus 8 0.00352 Common 17 Neoniphon marianus 2 0.00319 Occasional 42 Unknown crab 14 0.00289 Hypoplectrus spp. 1 0.00237 Occasional Neogonodactylus curacaoensis 5 0.00192 Occa sional Scarus iserti 2 0.00152 Common Serranus tigrinus 3 0.00062 Common Coryphopterus spp. 5 0.00057 Abundant to common 1 Stegastes planifrons 1 0.00056 Abundant to common Stegastes variabilis 1 0.00053 Occasional 25 Microgobius carri 2 0.0 0034 Occasional Apogon maculatus 1 0.00023 Common Haemulon spp. 2 0.00021 Common Lucayablennius zingaro 2 0.00007 Occasional
48 Table 3 1. Continued Taxon Frequency %IA i Abundance Rank Acanthurus bahianus 2 Common to occasional 45 Chaetodon aculeatus 2 Common to occasional 43 Chaetodontidae 2 Common to occasional Pomacentridae 4 Common Serranus baldwini 1 Common to occasional Liopropoma eukrines 1 Not reported Gramma melacara 1 Common 9 Sparisoma aurofrenatum 1 Com mon to occasional 16 Scarus taeniopterus 4 Common to occasional 22 Gobiidae 16 Common to occasional Gobiosoma genie 1 Common Malacoctenus boehlkei 1 Occasional 36 Malacoctenus versicolor 1 Occasional Amblycirrhitus pinos 3 Common to occasional Periclimenes yucatanicus 1 Common Stenopodidae 2 Portunus spinimanus 1 Common Portunidae 1 Common to occasional Unknown gastropod 1 Table 3 2. Comparison of the alimentary index of three different prey items f or all lionfish and three different age classes, age 0, age 1 2 and age 3 4 lionfish. Species IA i Overall Age 0 Age 1 2 Age 3 4 Cinetorhynchus manningi 1.4256 7.9796 1.4635 0.4308 Clepticus parrae 7.6605 0.0000 4.4775 34.0980 Gramma loreto 0.5401 0.00 00 0.9471 0.0000
49 Figure 3 1. Cumulative species curve for lionfish prey from Little Cayman Island. Figure 3 2. Proportion of lion fish diets containing fish or shrimp according to age Age 0 fish = 64 282 mm; age 1 fish = 138 288 mm; age 2 fish = 169 370 mm; age 3 fish = 205 372 mm; age 4 fish = 356 392 mm.
50 (a) (b) Figure 3 3. Typical stomach contents from younger lionfish and a mature lionfish. (a) shows stomach contents of a younger lionfish entirely comprised of small shrimp and ( b) shows stomach contents of an adult lionfish comprising a parrotfish, two slender filefish, a trumpet fish, and a harlequin bass. Figure 3 4 Body depth of prey, vertical gape and horizontal gape for lionfish of differing total lengths.
51 Figure 3 5 Dry weights of diets versus dry weights of lionfish.
52 CHAPTER 4 CONCLUSION Although the growth and reproduction of lionfish complicate management of this invasive species, c ulling efforts in localized areas have been shown to be effective in reducing densities of lionfish, and they remain the best available option for control (Frazer et al. 2012). Jud and Layman (2012) also make a case for localized control by demonstrating that lionfish exhibit high site fidelity which suggests replacement will be th rough recruitment and not im migration. Another factor influencing removal strategies involves the kind of effort and gear required to capture lionfish. A lthough there have been reports of hook and line fishermen occasionally catching lionfish t raditional hook and line methods will not target only lionfish, and in fact, are not very effective (Barbour et al. 2011, Morris and Whitfield 2009). Lionfish also have been caught in lobster traps (Morris and Whitfield 2009), but trapping may yield bycatch and dama ge to coral reefs. To date the least damaging and most efficient and effective means of catch ing lionfish are spearing or netting while on SCUBA. Limitations associated with this method include the restrictions on bottom time and depth limits Therefore efforts typically focus o n the shallower portion of the depth range inhabited by lionfish, which may leave a source population that supplies recruits Nevertheless, i f managers can identify areas of special concern and focus removal efforts there, they may be able to reduce impacts of lionfish in the shallow water coral communities that harbor the highest diversity and density of fish or threatened and endangered species (Albins & Hixon 2008, Brokovich et al. 2008, Frazer et al. 2012). While a great deal of work has been done to improve our understanding of lionfish ecology and biology, questions still remain. Further research on lionfish could include a
53 reanalysis of the proportion of P. miles and P. volitans present in different regions of the invaded rang e, which would clarify the composition of the invasive populations. Reassessing growth once the Little Cayman Island population has matured may increase the accuracy of estimated growth parameters. Furthermore, steps should be taken to determine natural mo rtality. Natural mortality remains one of the most elusive parameters to obtain, and accurate estimates typically require a population that has been established for some time. Managers would also benefit from knowledge regarding the degree of compensatory growth and reproduction in fished populations. Once this information becomes available, it will increase the accuracy of models that predict population dynamics of lionfish and the amount of fishing pressure needed to maintain reduced densities of this inv asive species.
54 LIST OF REFERENCES Aguilar Perera A, Tuz Sulub A ( 2010 ) Non native, invasive r ed lionfish ( Pterois volitans [Linnaeus, 1758]: Scorpaenidae), is first recorded in the southern Gulf of Mexico, off the northern Yucatan Peninsula, Mexico. Aqua t Invasions 5:S9 Albins MA, Hixon MA (2008) Invasive Indo Pacific lionfish Pterois volitans reduce recruitment of Atlantic coral reef fishes. Mar Ecol Prog Ser 367:233 238 Albins MA, Hixon MA ( 2011 ) Worst case scenario: potential long term effects of Invas ive predatory lionfish ( Pterois volitans ) on Atlantic and Caribbean coral reef communities. Environ Biol Fish doi: 10.1007/s10641 011 9795 1 Albins MA, Lyons PJ (2012) Invasive red lionfish Pterois volitans blow directed jets of water at prey fish. Mar Eco l Prog Se 448:1 5 Anderson MJ, Gorley RN, Clarke KR (2008) PERMANOVA + for PRIMER: guide to software and statistical methods. PRIMER E, Plymouth, United Kingdom gy 83:2145 2151 Baltz DM, Fleeger JW, Rakoeinski CF, McCall JM (1998) Food, density, and microhabitat: factors affecting growth and recruitment potential of juvenile saltmarsh fishes. Environ Biol Fish 53:89 103 Barbour AB, Allen MS, Frazer TK, Sherman KD (2011) Evaluating the potential efficacy of invasive lionfish ( Pterois volitans ) removals. PLoS ONE 6(5):e19666. doi:10.1371/journal.pone.0019666 Bax N, Williamson A, Aguero M, Gonzalez E, Geeves W (2003) Marine invasive alien species: a threat to global b iodiversity. Mar Policy 27:313 323 Beamesderfer RCP, North JA (1995) Growth, natural mortality, and predicted response to fishing for largemouth bass and smallmouth bass populations in North America. North Am J of Fish Manage 15:688 704 Bennet DH, Hatch DR Liter MD (1991) Managing largemouth bass in the northwest: a game of recruitment, protection, and patience. U.S. Forest Service General Technical Report RM 207:313 322 Berkeley SA, Hixon MA, Larson RJ, Love MS (2004) Fisheries sustainability via protecti on of age structure and spatial distribution of fish populations. Fisheries 29:23 32 Brokovich E, Einbinder S, Shashar N, Kiflawi M, Kark S (2008) Descending to the twilight zone: changes in coral reef fish assemblages along a depth gradient down to 65 m. Mar Ecol Prog Ser 371:253 262
55 Burnham KP, Anderson DR (2004) Multimodel inference: understanding AIC and BIC in model selection. Colorado Cooperative Fish and Wildlife Research Unit Technical Memorandum Caldow C, Wellington GM (2003) Pattern of annual incr ement formation in otoliths of pomacentrids in the tropical western Atlantic: implications for population age structure examination. Mar Ecol Prog Ser 265:185 195 Cerino DS (2010) Bioenergetics and trophic impacts of invasive Indo Pacific lionfish. Master s Thesis, East Carolina University, Greenville, N orth C arolina Coelho VR, Manfrino C (2007) Coral community decline at a remote Caribbean island: marine no take reserves are not enough. Aquat Conserv: Mar Freshwat Ecosyst 17:666 685 010) Predation rates of Indo Pacific lionfish on Bahamian coral reefs. Mar Ecol Prog Ser 404:219 225 Cucherousset J, Olden JD (2011) Ecological impacts of nonnative freshwater fishes. Fisheries 36:215 230 t IM (2011) In do Pacific lionfish are larger and more abundant on invaded reefs: a comparison of Kenyan and Bahamian lionfish populations. Biol Invasions 13:2045 2051 Dierking J, Williams ID, Walsh WJ (2009) Diet composition and prey selection of the introduced grouper species peacock hind ( Cephalopholis argus ) in Hawaii. Fish Bull 107: 464 476 Eggleston DB, Grover JJ, Lipcius RN (1998) Ontogenetic diet shifts in Nassau grouper: trophic linkages and predatory impact. Bull Mar Sci 63:111 126 Frazer TK, Jacoby CA, Edwards MA, Barry SC, Manfrino CM (2012) Coping with the lionfish invasion: can targeted removals yield beneficial effects? Res Fish Sci 20:1 8 Froese R, Pauly D (2009) FishBase. Accessed 12 Sept www.fishbase.org Goldschmidt T, Witte F, Wanink J (1993) Cascading effects of the introduced Nile perch on the detritivorous/planktivorous species in the sublittoral areas of Lake Victoria. Conserv Biol 7:686 700 Gordon DR (1998) Effects of invasive, non indigenous plant species on an ecosystem processes: lessons from Flo rida. Ecol Appl 8:975 989 Green BS, Mapstone BD, Carlos G, Begg GA ( 2009 ) Introduction to otoliths and fisheries in the tropics. Reviews: methods and technologies in fish biology and fisheries doi: 10.1007/978 1 4020 5775 5_1
56 Green SJ, Akins JL, Ct IM (2 011) Foraging behavior and prey consumption in the Indo Pacific lionfish on Bahamian coral reefs. Mar Ecol Prog Ser 433:159 167 Green S, Ct I (2009) Consumption potential of invasive lionfish ( Pterois volitans ) on Caribbean coral reefs. Annu Proc Gulf Ca ribb Fish Inst 62 :358 359 Gwinn DC, Allen MS, Rogers MW (2010) Evaluation of procedures to reduce bias in fish growth parameter estimates resulting from size selective sampling. Fish Res 105: 75 79 Hamner RM, Freshwater DW, Whitfield PE ( 2007 ) Mitochondria l cytochrome b analysis reveals two invasive lionfish species with strong founder effects in the western Atlantic. J Fish Biol 71:214 222 Hare JA, Whitfield PE (2003) An integrated assessment of the introduction of lionfish ( Pterois volitans/miles complex) to the western Atlantic Ocean. NOAA Technical Memorandum NOS NCCOS 2:1 21 Hoyer MV, Shireman JV, Maceina MJ (1985) Use of otoliths to determine age and growth of largemouth bass in Florida. T Am Fish Soc 114: 307 309 Humann P, DeLoach N (2002) Reef fish i dentification: Florida Caribbean Bahamas. New World Publications, Inc., Jacksonville, Florida Humann P, DeLoach N (2002) Reef creature identification: Florida Caribbean Bahamas. New World Publications, Inc., Jacksonville, Florida Hyslop EJ (1980) Stomach c ontents analysis a review of methods and their application. J Fish Biol 17:411 429 Jud ZR, Layman CA (2012) Site fidelity and movement patterns of invasive lionfish, Pterois spp., in a Florida estuary. J Exp Mar Biol Ecol 414 415:69 74 Kimball ME, Miller J M, Whitfield PE, Hare JA (2004) Thermal tolerance and potential distribution of invasive lionfish ( Pterois volitans/miles complex) on the east coast of the United States. Mar Ecol Prog Ser 283:269 278 Koehn JD (2004) Carp ( Cyprinus carpio ) as a powerful in vader in Australian waterways. Freshwater Biol 49: 882 894 Lai HL, Gunderson DR (1987) Effects of ageing errors on estimates of growth, mortality and yield per recruit for walleye pollock ( Theragra chalcogramma ). Fish Res 5:287 302 Lesser MP, Slattery M (2 011) Phase shift to algal dominated communities at mesophotic depths associated with lionfish ( Pterois volitans ) invasion on the Bahamian coral reef. Biol Invasions 13:1855 1868
57 Marriot RJ, Mapstone BD (2006) Geographic influence on and the accuracy and pr ecision of age estimates for the red bass, Lutjanus bohar (Forsskal 1775): A large tropical reef fish. Fish Res 80:322 328 Marshall CT, Needle CL, Thorsen A, Kjesbu OS, Yaragina NA (2005) Systematic bias in estimates of reproductive potential of Atlantic c od ( Gadus morhua ) stock : implications for stock recruit theory and management. Can Fish Aquat Sci 63:980 994 Mesa ML, Mesa GL Micalizzi M ( 2005 ) Age and growth of madeira scorpionfish, Scorpaena maderensis Valenciennes, 1833, in the central Mediterranean. Fish Res 74:265 272 Molnar JL, Gamboa RL, Ravenga C, Spalding MD (2008) Assessing the global threat of invasive species to marine biodiversity. Front Ecol Environ 6:485 492 Morris JA Jr (2009) The biology and ecology of the invasive Indo Pacific lionfish PhD dissertation, North Carolina State University, Raleigh, N orth Carolina Morris JA Jr., Akins JL (2009) Feeding ecology of invasive lionfish ( Pterois volitans ) in the Bahamian archipelago. Environ Biol Fish 86:389 398 Morris JA Jr., Akins JL, Barse A, Cerino D, Freshwater DW, Green SJ, Muoz RC, Paris C, Whitfield PE (2009) Biology and ecology of the invasive lionfishes, Pterois miles and Pterois volitans Annu Proc Gulf Caribb Fish Inst 61:1 6 Morris JA Jr., Whitfield PE (2009) Biology, ecology, control and management of invasive Indo Pacific lionfish: an updated integrated assessment. NOAA Technical Memorandum NOS NCCOS 99:1 57 Morris JA Jr Shertzer KW, Rice JA ( 2010 ) A stage based matrix population mo del of invasive lionfish with impl ications for control. Biol Invasions 13:7 12 Moyle PB, Light T (1996) Fish invasions in California: do abiotic factors determine success? Ecology 77:1666 1670 Muoz RC, Currin CA, Whitfield PE (2011) Diet of invasive lionf ish on hard bottom reefs of the southeast USA: insights from stomach contents and stable isotopes. Mar Ecol Prog Ser 432:181 193 Murie DJ, Parkyn DC (2005) Age and growth of the white grunt ( Haemulon plumieri ): a comparison of two populations along the Wes t coast of Florida. Bull Mar Sci 76:73 93 Newman SJ, Williams DM, Russ GR (1996) Age validation, growth and mortality rates of tropical snappers (Pisces: Lutjanidae) Lutjanus adetti (Castelnau, 1873) and L. quinquelineatus (Bloch, 1790) from the central gr eat barrier reef, Australia. Mar Freshw Res 47:575 584
58 Paddack MJ, Reynolds JD, Aguilar C, Appledoorn RS, and others (2009) Recent region wide declines in Caribbean reef fish abundance. Curr Biol19:590 595 Pauly D (1980) On the interrelationships between n atural mortality, growth parameters, and mean environmental temperature in 175 fish stocks. J Cons Int Explor Mer 39:175 192 Pimentel D, Lach L, Zuniga R, Morrison D (2000) Environmental and economic costs of nonindigenous species in the United States. Bio science 50:53 65 Potts JC (2010) Age and growth of lionfish from the western north Atlantic. Annu Proc Gulf Caribb Fish Inst 63 :314 Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/, 1997 2012 Ro gers DE, Ruggerone GT (1993) Factors affecting marine growth of Bristol Bay sockeye salmon. Fish Res 18:89 103 Rose KA (2000) Why are quantitative relationships between environmental quality and fish habitat so elusive? Ecol Appl 10:367 385 Schindler DE, E ssington TE, Kitchell JF, Boggs C, Hilborn R (2002) Sharks and tunas: fisheries impacts on predators with contrasting life histories. Ecol Appl 12:735 748 Schofiel d PJ ( 2009 ) Geographic extent and chronology of the invasion of non native lionfish ( Pterois volitans [Linnaeus 1758] and P. miles [Bennet 1828]) in the Western North Atlantic and Caribbean Sea. Aquat Invasions doi 10.3391/ai.2009.4.3 Schofield PJ (2010) Update on geographic spread of invasive lionfishes ( Pterois volitans [Linnaeus, 1758] and P. m iles [Bennet, 1828]) in the western north Atlantic Ocean, Caribbean Sea and Gulf of Mexico. Aquat Invasions 5:S117 S122 Semmens BX, Buhle ER, Salomon AK, Pattengill Semmens CV (2004) A hotspot of non native marine fishes: evidence for the aquarium trade as an invasion pathway. Mar Ecol Prog Ser 266:239 234 Sharf FS, Juanes F, Rountree RA (2000) Predator size prey size relationships of marine fish predators: interspecific variation and effects of ontogeny and body size on trophic niche breadth. Mar Ecol Prog Ser 208:229 248 St John J (1999) Ontogenetic changes in the diet of the coral reef grouper Plectropomus leopardus (Serranidae): patterns in taxa, size and habitat of prey. Mar Ecol Prog Ser 180:233 246
59 Sweatman JJ, Kohler CC (1991) Validation of daily oto lith increments for young of the year white crappies. North Am J Fish Manage 11:499 503 Todd BD, Rothermel BB, Reed RN, Luhring TM, Schlatter K, Trenkamp L, Gibbons JW (2008) Habitat alteration increases invasive fire ant abundance to the detriment of amph ibians and reptiles. Biol Invasions 10:539 546 Valentim MFM, Caramaschi EP, Vianna M (2008) Feeding ecology of monkfish Lophius gastrophysus in the south western Atlantic ocean. J Mar Biol Assoc UK 88:205 212 Vilizzi L, Walker KF (1999) Age and growth of t he common carp, Cyprinus carpio in the River Murray, Australia: validation, consistency of age interpretation, and growth models. Environ Biol Fish 54:77 106 von Bertalanffy L ( 1957 ) Quantitative laws in metabo lism and growth. Q Rev Biol 32:217 231 Wilson JD, Dorcas ME, Snow RW ( 2011 ) Identifying plausible scenarios for the establishment of invasive Burmese pythons ( Python molurus ) in Southern Florida. Biol Invasions 13:1493 1504
60 BIOGRAPHICAL SKETCH Morgan Edwards was born and raised in Gainesville, Flor ida. After graduating from high school in 2002, she attended the University of Florida to study wildlife ecology and conservation. In 2008 she began work as a technician in the Department of Fisheries and Aquatic Sciences at the University of Florida. Once fully immersed in the world of field work; traveling all over Florida to help study riverine and near shore ecosystems, and even spending a three week stint at the bottom of the Grand Canyon on a research project, she decided to pursue a graduate degree w ith the Frazer lab in the Fisheries department. Her graduate work focused on the invasive lionfish population of Little Cayman, and while there, she completed a teaching internship with the Little Cayman Research Centre where she resided for six months. He r experience in graduate school provided her with unique challenges that she hopes have helped her to become a better scientist. Morgan plans to continue working with the Frazer lab after graduation, working in the place where she first fell in love with t he ocean, the Florida coast.