SALAMANDRA 46(3) 141–146 20 AMuigcurosht a2b0i1ta0t seleIcStSioNn 0i0n3 C6–te3n3o7s5aura bakeri Microhabitat selection in the spiny-tailed iguana Ctenosaura bakeri on Utila Island, Honduras Ulrich Schulte1 & Gunther Köhler2 1) Present address: Ulrich Schulte, University of Trier, Department of Biogeography, Am Wissenschaftspark 25-27, 54295 Trier, Germany 2) Present address: Gunther Köhler, Senckenberg Forschungsinstitut und Naturmuseum, Department of Herpetology, Senckenberganlage 25, 60325 Frankfurt am Main, Germany Corresponding author: Ulrich Schulte, e-mail: [email protected] Manuscript received: 06 March 2009 Abstract. We studied the habitat use in the Utila iguana (Ctenosaura bakeri Stejneger, 1901) in a heterogeneous mangrove forest site on the north coast on Utila Island using 162 sightings of iguanas along a line-transect. Our results suggest on- togenetic size-related shifts of habitat use in this species. Hatchlings preferred the fringe area of the red mangrove zone and perched on the ground or at low heights (mean ± SD = 0.8 ± 1.28 m). Similarly, juveniles were encountered most frequently in the red mangrove zone at mean tree heights of 3.6 ± 2.83 m, whereas adult iguanas preferentially selected the white man- grove zone. Males perched higher (4.9 ± 2.7 m) than females (3.75 ± 3.15 m). The habitat choice of hatchlings is seen as a predator avoidance strategy. Adult C. bakeri may select the white mangrove as the best opportunity to reach preferred body temperature for microbial digestion and fermentation processes. Sightings of juvenile and adult iguanas were associated with the presence of suitable cavities and trunks of mangrove trees as important sleeping and hiding-places. Key words. Squamata, Iguanidae, autecology, line-transect, mangrove, perch height. Introduction congeners of Ctenosaura, which mainly inhabit exposed open habitats in arid and subhumid lowland areas from Habitat selection has been studied in a variety of lizard spe- Mexico to Panama (Gutsche & Köhler 2008). The lat- cies in their natural habitat as a function of thermal ecology, ter authors propose that the extraordinary lifestyle of C. food and mate acquisition, and protection from predators bakeri is the result of ecological exclusion from more suit- (Beck & Jenning 2003, Reaney & Whiting 2003, Quirt able habitats by the larger, more widespread and highly et al. 2006). Within a species, different size classes can show adaptable C. similis. Nowadays, the species can be found in differences in interspecific habitat use due to the physio- open habitats due to the vast destruction of the mangrove logical structure of the environment, their physiology, food for land use. Meanwhile, different aspects of the life histo- availability, and predator avoidance (Jenssen et al. 1998, ry (reproduction, Kuttler 2000, feeding and population Shaw 2005, Nemes et al. 2006). Primary concerns for ecto- ecology, Gutsche 2005a) and phylogenetic position and therms are thermoregulatory requirements (microclimates) hybridization of the species have been studied (Köhler et and protection from predators (Beck & Jenning 2003). For al. 2000, Gutsche & Köhler 2004, Gutsche & Köhler the spiny-tailed iguana Ctenosaura hemilopha, Blízquez & 2008, Pasachnik et al. 2009). Gutsche (2005b) demon- Rodríguez-Estrella (2006) provided evidence that pre- strated a high fidelity of adults to tree cavities as retreats. dation risk is the strongest factor influencing habitat choice Nevertheless, little is known about the microhabitat prefer- in Baja California Sur. The species select cardon cacti that ences of this endangered species. To focus conservation ef- offer woodpecker holes as retreat sites. Information about forts and understand the distribution patterns of this igua- reptile habitat preferences in tropical tidal mangrove for- na, it is crucial to identify those preferences within the het- ests is scanty. On the one hand there are only few examples erogeneously distributed and threatened mangrove habi- of reptiles inhabiting mangroves (Gutsche 2005b). On the tats on the island. The purpose of this study was to char- other hand there is a lack of information due to the difficult acterize microhabitat selection in different age groups of accessibility of highly structured mangrove forests. Ctenosaura bakeri in its natural habitat. The spiny-tailed iguana, Ctenosaura bakeri, is endem- ic to only about 8 km2 of mangrove swamp on the island of Utila, which is the smallest (41 km2) of the three major Material and methods Bay Islands off the Caribbean coast of Honduras. This is- Site description land hosts one of the largest mangroves in the Caribbean (Clauss & Wild 2002). The humid, periodically flooded Our study was conducted at Iron Bound situated on the mangrove swamp contrasts the main habitats of all other north coast of Isla de Utila (UTM, WGS84 0510681/1782319). © 2010 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT), Rheinbach, Germany All articles available online at http://www.salamandra-journal.com 141 Ulrich Schulte & Gunther Köhler The area is characterized by sharp-edged coral rocks with General study procedures lichen, cacti (Mammelaria spec.) and succulents (Sesuvium portulacastrum). Landwards, a beach forest belt on sandy Fieldwork was conducted during two periods from 20 No- ground (essential for egg-laying of Ctenosaura bakeri) dis- vember to 14 December of 2004, and from 12 June to 13 Au- perses with typical pioneer species (Ipomoea pres-caprae, gust of 2005 on Isla de Utila, Honduras, with a minimum Sesuvium portulacastrum, Coccoloba uvifera, Cocos nuci- of two persons in the mangrove. One permanently marked fera, Terminalia catappa). The more elevated area (20 m line transect with a total length of 556 metres, which con- above sea level) of the coralline platform supports a belt of tained all vegetation types that are important for Ctenosau- Caribbean dry forest dominated by species such as Bursera ra bakeri’s life history traits (Fig. 1), was established in the simaruba, Begonia cf. lindleyi and Thrinax radiata (Clauss (Iron Bound) area. As we detected only on two occasions & Wild 2002). Farther landwards, after passing mangrove an individual of C. bakeri in the Caribbean dry forest and associates (Acrostichum aureum, Hibiscus tiliaceus, Cono- none in the beach forest, we analysed lizards sighting fre- carpus erectus, Tabebuia rosea, Batis maritima), the man- quency in relation to microhabitat availability exclusively grove basin of the island is reached (elevation above sea in the three different mangrove zones (442277,,33 mm,, fffooorrr pppooorrr--- level: 2-8 metres), which is subjected to the periodical tions of zones see Table 1). While setting the transect, veg- tides. As a result of different adaptations in their tolerance etation sampling was conducted in addition to available to tidal inundation and salinity, Utila’s mangrove swamp data (Clauss & Wild 2002). As it is known that tree cavi- contains three rather heterogeneous mangrove species. ties are important resources as sleeping and hiding plac- The red mangrove (Rhizophora mangle, Fig. 2A), with its es for this species and therefore control population den- characteristic aerial roots, can be found on the lowest el- sity (Gutsche 2005b), all tree cavities along the transect evation, on frequently flooded sites, especially around the line (10 m at each site) were noted. Tree cavities typical- channel and lagoons. The black mangrove (Avicennia ger- ly result from core-putridity with increasing age in black minans, Fig. 2B), which is less salt water tolerant, grows mangroves. White mangroves require a higher age to de- at humid sites which are not flooded too often. The white velop tree cavities, and in red mangroves bigger cavities are mangrove (Laguncularia racemosa, Fig. 2C) occurs on the scarce. landward fringe of the mangrove communities and readily On an average distance of 12 metres, large trees were colonizes disturbed sites (Hogarth 1999). permanently marked with 44 plastic tags. We subdivided Figure 1. Delineation of the line transect, containing all important vegetation types for Ctenosaura bakeri’s life history traits on the Iron Bound coast on Utila Island. 142 Microhabitat selection in Ctenosaura bakeri the study site into adjacent sampling units. Geographic co- ordinates of all marked trees were determined with a GPS (Garmin Geko 201, UTM WGS84) and transformed with ArcView 3.2. software into GIS format (Fig. 1). On 28 sur- veys, distance sampling was conducted in a standardised manner from 12 June to 13 August 2005 by two persons from 10:00-11:30 in the morning. Iguanas were counted within a strip of 10 m wide to both sides of the transect line. Once an iguana was located, its age group (hatchling: SVL (snout vent length) > 55 mm, juveniles: SVL > 100 and < 150 mm, adults: SVL > 150 mm, according to Gutsche 2005a), sex, tree number, distance from the transect line, and perch height were recorded. Additionally, possible predators and other reptile species were recorded. The deviation (D) of expected iguana sightings at each mangrove unit was calculated by the following formula: D = (O – E) / E, where O is the observed proportion of sight- ings (%) and E the proportion of vegetation units on the transect. Data are between -1 and 0, if an age group avoids a particular habitat, whereas data > 0 indicate a preference for a habitat, respectively (Čeirāns 2004). Age group-spe- cific (hatchlings: n = 55; juveniles: n = 51; adults: n = 56) and sex-specific differences in mean perch heights of adult in- dividuals (♂ = 18; ♀ = 32) were visualized using box plots with SPSS 13.0 for windows. We opted for this approach, because we could not find any differences in the distribu- tion of both sexes in the mangrove zones, which have dif- ferent growth-heights. Considering that juvenile and adult Ctenosaura bakeri occupy and defend tree cavities perenni- ally (Gutsche 2005b), sightings of these cohorts (n = 107) were compared with the presence of tree cavities. Results Mangrove preferences A total of 162 observations were made (55 hatchlings, 51 ju- venile and 56 adult iguanas.) We found conspicuous dif- ferences in the distribution of all age classes, whereas we could not observe sex-specific habitat differences. Strik- ingly, the majority (69.1%, n = 38) of all hatchling sightings occurred in a mangrove zone dominated by the red man- grove, which has the highest prevalence of cavities. 23.6% (n = 13) of all sightings were made in the black mangrove, and only very few hatchlings were seen in the white man- grove (Table 1). Sightings of juvenile iguanas were com- mon in the red mangrove (43.1%, n = 22), as well as in the black mangrove (37.2%, n = 19). Adult C. bakeri preferred Figure 2. A) Red mangrove (Rhizophora mangle), B) Black man- the white mangrove zone (41.1%, n = 23), while sighted at grove (Avicennia germinans), C) White mangrove (Laguncularia about the same rate in the red (30.3%, n = 17) and black racemosa). mangrove (28.6%, n = 16). Perch heights, tree cavities and predators Surprisingly, the black mangrove unit (36.4%, n = 20) Sightings of hatchlings (n = 55) were mostly at perch harboured fewer tree cavities than the red mangrove unit heights (± SD) of 0.8 ± 1.28 m (Fig. 3). Juvenile iguanas (n (41.8%, n = 23), although in the latter unit most tree cavities = 51) perched at an average height of 3.6 ± 2.83 m. Sightings were recorded in the few scattered black mangrove trees. of adult C. bakeri (n = 56) occurred at an average perch This may be due to the very heterogeneous distribution of height of 4.3 ± 3.16 m. Males (n = 18) were sighted at an av- mangrove species in the red mangrove zone. The occur- erage height of 4.9 ± 2.7 m, whereas females (n = 32) were rence of juvenile and adult iguana sightings suggests a cor- seen at 3.75 ± 3.15 m (Fig. 4). relation with the occurrence of tree cavities. 143 Ulrich Schulte & Gunther Köhler Table 1. Lengths of transects (TL) in the three different mangrove units and deviation (D) from the expected data of iguana sightings. Data are between -1 and 0 if an age group avoids a particular habitat, whereas data > 0 indicate a preference for a habitat, respectively. Additionally, mean perch heights and standard deviations are presented. Mangrove type TL (m) hatchlings juveniles adults tree cavities (n = 55) (n = 51) (n = 56) (n = 55) 139.7 23 Dominance of red mangrove 1.17 0.35 -0.1 (32.7%) (41.8%) 192.6 20 Dominance of black mangrove -0.48 -0.16 -0.37 (45.1%) (36.4%) 95 12 Dominance of white mangrove -0.75 -0.19 0.85 (22.2%) (21.8%) Perch height (mean ± SD) (m) 0.8 ± 1.28 3.6 ± 2.83 4.3 ± 3.16 Most avian predators (Cathartes aura, Buteogallus an- with laterally indicated dark staining and crossbands on the thracinus, Coccyzus minor palloris, Melanerpes aurifrons, tail. This coloration is strikingly different to the greenish Bubulcus ibis, Fregata magnificens) were observed in the coloration of freshly hatched and juvenile C. similis (Köh- black mangrove. As a potential predator of hatchlings and ler 1998). Considering behavioural (perching near the juveniles, Ctenosaura similis reaches high densities in the forest ground) and habitat characteristics (soil coloration), Caribbean dry forest and was occasionally recorded in the coloration of C. bakeri hatchlings is an excellent cryptic the white or red mangrove. Single observations of Basilis- coloration and thermoregulatory feature. Numerous aerial cus vittatus (a possible predator of hatchlings) were made roots and dead wood at the swamp edge in combination in the red mangrove. In the black mangrove, the colubrid with a shrub stratum of Batis maritima and mangrove ferns snake Leptophis mexicanus as well as the southern opos- (Acrostichum speciosum) provide the best predator avoid- sum (Didelphis marsupialis) were detected. ance refuges and highest structural heterogeneity of all mangrove zones. Most hatchlings were observed perching on the mangrove soil or at heights of 1-3 m. Similar obser- Discussion vations were made on juveniles of C. melanosterna, which were far more terrestrial than adult individuals within hill The first hatchlings of 2005 were observed in the first week forests and wind scrubs on Cayo Menor, Honduras (Shaw of June in open sandy areas of the more elevated beach 2005). Great numbers of small hiding and safe perching forest (L. Dirksen pers. comm. 2005). Immediately after places at suitable heights for hatchlings are only available hatching, they migrate into the periphery of the mangrove in the periphery of the red mangrove (although not quan- swamp. We assume that avoidance of Ctenosaura similis, tified in our study), and neither in the more exposed tall adult conspecifics (cannibalism) and avian predators could black mangrove nor in the white mangrove. Although all explain the observed preference for the mangrove edge (red age groups are predominantly herbivorous, the herbal con- mangrove vegetation) of the hatchlings. Ctenosaura bakeri tingent increases with age (Gutsche 2005a), indicating an hatchlings are coloured relatively consistently grey-brown ontogenetic dietary shift in a manner similar to that known m) m) ht ( ht ( g g ei ei h h h h c c er er p p hatchlings juveniles adults females males age group sex Figure 3. Perch heights for the different age groups (hatchlings: Figure 4. Perch heights for both sexes (♂ = 18; ♀ = 32). n = 55; juveniles: n = 51; adults: n = 56). 144 Microhabitat selection in Ctenosaura bakeri from C. pectinata (Durtsche 2000). Hatchlings and juve- dazzo for continuing the monitoring of C. bakeri. Many niles predate more frequently on insects, and start to pre- thanks go to the volunteers of the station (especially Den- date on fiddler crabs (Uca rapax) with a total length of 64- nis Baulechner, Stesha Pasachnik and Sebastian 70 mm (Gutsche pers. comm. 2006), which reproduce in Schütze) for their field assistance during the study. We the summer (da Silva Castiglioni & Negreiros-Fran- also thank two anonymous reviewers for valuable com- sozo 2005). Predating on insects might be easier and more ments on an earlier draft of this manuscript. efficient at the terrestrial fringe of the red mangrove. None of the age groups preferred the homogeneous black mangrove with its nearly barren ground and lin- References ear growth, although it represented the largest vegeta- tion portion of the whole transect. Our data differ from Beck, D. D. & R. D. Jenning (2003): Habitat use by Gila Mon- sters: the importance of shelters. – Herpetological Mono- Kuttler’s (2000) observations that pure black mangrove graphs, 17: 111–129. stands would host the highest iguana densities (172 indi- viduals/ha), followed by red mangrove (131 individuals/ha) Blízquez, M. C. & R. Rodríguez-Estrella (2006): Factors in- fluencing the selection of basking perches on cardon cacti by and white mangrove zones (73 individuals/ha). However, spiny-tailed iguanas (Ctenosaura hemilopha). – Biotropica, 29: the very heterogeneous composition of all three mangrove 344–348. species on the Iron Bound coast makes comparisons be- Čeirāns, A. (2004): Reptiles in sub-boreal forests of Eastern Eu- tween these studies difficult. In the black mangrove, we rope: patterns of forest type preferences and habitat use in An- found the highest densities of avian predators. Gutsche’s guis fragilis, Zootoca vivipara and Natrix natrix (Squamata: (2005a) finding that iguana density is correlated with the Sauria: Anguidae, Lacertidae; Serpentes: Colubridae). – Her- occurrence of tree cavities is consistent with our observa- petozoa, 17: 65–74. tions. We could not find obvious differences or limitations Clauss, S. & C. Wild (2002): Planning of land area use and pro- in the availability of crabs (Uca rapax) as the most impor- tected areas as means for sustainable development on the is- tant protein-source (content: 32.7%, Gutsche 2005a) or in land of Utila, Honduras and possibilities and bounds of sus- deciduous mangrove leaves and flowers between the man- tainable tourism on the island of Utila, Honduras. – Com- grove zones. bined Diploma Thesis translated by A. J. Gutman, M. Frank, Most adult Ctenosaura bakeri sightings were recorded N. Schütz, K. Gees & A. Hoffmann. – Departments of in the white mangrove. The occasionally turquoise body Landscape Ecology and Vegetation Ecology, Technical Uni- versity of Munich-Weihenstephan, Germany. coloration of adult iguanas represents an adaptation to- wards the stronger pronounced arboreal mode of life in da Silva Castiglioni, D. & M. L. Negreiros-Fransozo (2005): the mangrove canopy. The observed sex-specific difference Comparative population biology of Uca rapax (Smith, 1870) (Brachyura, Ocypodidae) from two subtropical mangrove in perch heights is most likely the result of higher territo- habitats on the Brazilian Coast. – Journal of Natural History, riality in males, which survey their home ranges and dis- 39: 1627–1640. play visual communication. Large lizards require high in- Durtsche, R. D. (2000): Ontogenetic plasticity of food habits in solation for maintaining their preferred body temperature. the Mexican spiny-tailed iguana, Ctenosaura pectinata. – Oec- Large adult iguanas depend more than other age groups ologia, 124: 185–195. on perching places with high solar radiation intensity. The Gutsche, A. (2005a): Freilanduntersuchungen zur Populations- digestion of the scytophyllous mangrove leaves is highly und Nahrungsökologie des Utila-Leguans (Ctenosaura bakeri temperature-dependend because it relies on the microbi- Stejneger 1901). – Dissertation Humboldt-Universität Berlin. al fermentation pathway in the colon. We assume that the Gutsche, A. (2005b): Distribution and habitat utilization of white mangrove zone with its marginal foliation (high in- Ctenosaura bakeri on Utila. – Iguana, 12: 143–151. solation) and brightly lit sandy ground (little absorption) Gutsche, A. & G. Köhler (2004): A fertile hybrid between is selected, because it offers the best conditions for ther- Cteno saura similis (Gray, 1831) and C. bakeri Steijneger, 1901 moregulation to adult iguanas. Additionally, white man- (Squamata: Iguanidae) on Isla de Utila, Honduras. – Salaman- grove leafs (Laguncularia racemosa) have the highest con- dra, 40: 201–206. tent of sugar, water and amylum of all species (Gutsche Gutsche, A. & F. Köhler (2008): Phylogeography and hybrid- 2005a), and crabs are easier to predate on the dryer ground ization in Ctenosaura species (Sauria, Iguanidae) from Car- (reduced searching costs). ibbean Honduras: insights from mitochondrial and nuclear However, further studies on habitat selection in C. ba- DNA. – Zoosystematics and Evolution, 84: 245–253. keri should be conducted during different seasons to eval- Hogarth, P. J. (1999): The Biology of Mangroves. – Oxford Uni- uate the validity of our results. Nevertheless, it becomes versity Press Inc., New York. clear that protecting the heterogeneous composition of Jenssen, T. A., K. A. Hovde & K. G. Taney (1998): Size-related Utila’s mangrove and beach forest is the primary key to the habitat use by non-breeding Anolis carolinensis lizards. – Co- conservation of this species. peia, 1998: 774–779. Köhler, G. (1998): Further additions to the known herpetofauna of Isla de Utila (Islas de la Bahia, Honduras) with notes on Acknowledgements other species and a key to the amphibians and reptiles of the island. – Senckenbergiana Biologica, 77: 139–145. We thank the former and present directors of the Igua- Köhler, G., W. Schroth & B. Streit (2000): Systematics of the na Research and Breeding Station, Lutz Dirksen, Au- Ctenosaura group of lizards (Reptilia: Sauria: Iguanidae). – rel Heidelberg, Jimena Castillo and Angela Ran- Amphibia-Reptilia, 21: 177–191. 145 Ulrich Schulte & Gunther Köhler Kuttler, H. P. (2000): Untersuchungen zur Populationsökologie und zum Ortsverhalten von Ctenosaura bakeri. – Diplomar- beit. J. W. Goethe Universität, Frankfurt am Main. Nemes, S., M. Vogrin, T. Hartel & K. Öllerer (2006): Habitat selection at the sand lizard (Lacerta agilis): ontogenetic shifts. – North-Western Journal of Zoology, 2: 17–26. Pasachnik, S., B. M. Fitzpatrick, T. J. Near & A. C. Echter- nacht (2009): Gene flow between an endangered ende mic iguana, and its wide spread relative, on the island of Utila: when is hybridization a threat? – Conservation Genetics, 10: 1247–1254. Quirt, K. C., G. Blouin-Demers, B. J. Howes & S. C. Lougheed (2006): Microhabitat selection of five-lined skinks in northern peripheral populations. – Journal of Herpetology, 40: 335–342. Reaney, L. T. & M. J. Whiting (2003): Picking a tree: habitat use by the tree agama, Acanthocercus atricollis atricollis, in South Africa. – African Zoology, 38: 273–278. Shaw, H. J. (2005): Habitat selection and distribution of reptiles on Cayo Menor, Cayos Cochinos, Honduras. – Undergraduate Thesis, Staffordshire University. 146