Note brève Autumnal home range in radio-tracked TORTOISES (Testudo hermanni) from a semi-arid Mediterranean environment Luca Luiselli 1 *, Lorenzo Rugiero 1, Stefano Celletti 2, Roberto Papi 2, Giulia Gracceva 3, Manuela Stacchiotti 2, Federica Mancini 2, Giuseppe Berretta 2, Laura Berretta 2, Giovanni Bombara 2, Riccardo Fiaschetti 2, Emanuele Lucioli 2, Maria Grazia Trionfetti 2 & Andrea Ungaro 2 Résumé. Domaine vital automnal de tortues (Testudo hermanni) suivies par radio-pistage dans un environnement méditerranéen semi-aride. La phase d activité automnale est cruciale pour la plupart des reptiles européens qui restent inactifs plusieurs mois durant l hiver et qui doivent se préparer à cette période de latence en orientant leurs activités vers l acquisition de nourriture et de réserves énergétiques. Néanmoins les caractéristiques écologiques de cette phase d activité automnale demeurent relativement inconnues pour la majorité des espèces reptiliennes, y compris les chéloniens. Dans la présente étude nous avons suivi par radio-pistage quatre tortues (Testudo hermanni), deux mâles et deux femelles, durant la période d activité automnale afin de comprendre les patterns de déplacement et l étendue des domaines vitaux. L étude de terrain a été conduite dans une zone méditerranéenne semi-aride du centre de l Italie. Les domaines vitaux ont été déterminés par la méthode du polygone convexe minimal et les taux de chevauchement des domaines individuels ont été calculés. Les domaines dépassaient en moyenne les deux hectares et se sont avérés légèrement plus grands chez les mâles que chez les femelles, bien que la différence n ait pas été statistiquement significative. Il est apparu une substantielle différence interindividuelle dans la taille des domaines au sein d un même sexe, confirmant ainsi des patterns déjà observés ailleurs sur des tortues. Les chevauchements des domaines se sont avérés très faibles entre les individus, particulièrement entre les deux mâles. Ces données sont discutées en rapport avec les principaux types d activité que montrent les tortues durant la phase automnale. The annual activity cycle of European reptiles is characterized by a more or less prolonged inactive phase during the winter months because of the cold external temperatures (e.g., Bruno & Maugeri, 1977). Thus, for several reptile species the phase of autumnal activity is crucial in order to re-build the energy reserves necessary to spend the winter and for reproduction to take place the following year (e.g., Luiselli, 1992; Bonnet et al., 1998). Nonetheless, despite its importance for the ecology of most temperate zone reptiles, the autumnal phase of activ- 1 F.I.Z.V. (Ecology) and Environmental Studies Centre Demetra s.r.l., via Olona 7, 00198 Roma, Italy. E-mail: lucamlu@tin.it 2 Parco Naturale Regionale Marturanum, Piazza G. Marconi 21, 01010 Barbarano Romano (VT), Italy. E-mail: marturanum@parchilazio.it 3 Dipartimento di Biologia, Universita di Roma Tre, viale G. Marconi 442, 00152 Roma, Italy * Corresponding author: lucamlu@tin.it Rev. Écol. (Terre Vie), vol. 64, 2009. 73
ity remains badly known. For instance, whereas plenty studies have monitored movements of reptiles during spring- and summertime, i.e. during the mating season and the oviposition period (e.g., see Calzolai & Chelazzi, 1991; Madsen, 1984; Luiselli, 1995; Zuffi & Odetti, 1998), very few research has been done in order to determine the homing patterns of temperate reptiles during autumn. Thus, studies focused on determining homing patterns and movements of European reptiles in autumn will provide a fertile field of investigation in the years to come, and will likely allow scientists to better understand optimization strategies for homing and gaining energy in this crucial period of the activity cycle. Studies on homing patterns are not only decisive to understand more in depth the autumnal activity of reptiles, but may also be useful to help conservation planning for threatened species. The Hermann s Tortoise (Testudo hermanni Gmelin, 1789) is increasingly threatened in Western Europe and in Italy because of habitat loss, pollution, and illegal removal of freeranging specimens for the pet trade (Corbett, 1989; Hailey, 2000; Mazzotti, 2004). Hence, a detailed knowledge of the biology of this species is necessary for conservation and appropriate management of the various populations, especially in areas where large populations can still be present because of the presence of suitable climatic and environmental conditions Our aim in this study is to address preliminary data on movements and homing patterns of Testudo hermanni during the autumnal months, at a Mediterranean area of central Italy characterized by semi-arid climate and sparse vegetation cover. MATERIALS AND METHODS The study area was included in the Parco Naturale Regionale Marturanum (Barbarano Romano, Province of Viterbo, Latium, Italy), a over 1200 hectares protected area with altitudes ranging between 175 and 541 m a.s.l. The geographic position of this territory is very interesting, because it is situated between the temperate and the Mediterranean zones (Blasi et al., 1993). Thus, various kinds of woodlands are present: from mixed oak forests with predominance of Quercus cerris, or of Quercus roverella, to zones with bushy and scrub vegetation (with dominance of Spartium junceum, or of Pliurus spina-christi, Pyrus amygdaliformis and Rubus sp.). A wide variety of grassy pastures, due to a different combination of exposition, topography and human impact (xerophilous; mesophilous; meso-hygrophilous), are also found. Tortoises are still relatively common in the study area (Luiselli et al., unpublished data). Radio-tracking was conducted in an area characterized by semiarid climate conditions, with vegetation including a mosaic of bushes or sparse brush of Pliurus spina-christi, Pyrus amygdaliformis and Rubus sp. Grass between the bushes was usually very low because of intense grazing, and the ground was also characterized by sparse but widespread stones. Other reptile species at the radio-tracking site include Podarcis muralis, Podarcis sicula, Lacerta bilineata, Hierophis viridiflavus, Coronella girondica, Elaphe quatuorlineata, Zamenis longissimus and Vipera aspis. For this study we radio-tracked two free-ranging adult females (respectively, T1 was 16.3 cm linear carapace length, and T3 was 13.0 cm), and two free-ranging adult males (T2, 13.0 cm long, and T4, 14.0 cm long). T3 had the carapace partially damaged, and got a veterinary treatment on the shell before being radio-implanted. Radio-transmitters (SIKA collar cable-tie type, modified in order to fit with tortoise morphology, and weighing less than 5% of the tortoise mass) were externally implanted on the dorsal part of the carapace by silicon glue, in a position that was particularly convenient to allow normal movements and activity to the animals. Tortoises were kept in captivity for 2-5 days after transmitter attachments before being released to the capture site. For radio-tracking, we used a SIKA Biotrack Radiotracking Receiver equipped with a flexible four-element Yagi antenna. The four tortoises were radio-tracked through the whole autumn season, i.e. from early September to early December 2007. The time elapsed between two successive fixes varied depending on the phase of the research monitoring: every day during the first weeks and every 2-3 days during the period of least activity of the tortoises (November and early December). In total, over 900 hours were spent in radio-tracking tortoises in the wild. We recorded by GPS 60 Garmin the geographic coordinates (UTM grid) of each data fix for every tortoise individual. On each visual sighting, the type of activity exhibited by tortoises was recorded (these data will be used for a forthcoming article). To calculate home range size, here we used the minimum convex polygon method (Hayne, 1949). This method has been recently recommended as the best method for calculating home ranges in reptiles (Row & Blouin-Demers, 2006) despite the frequent use of the kernel method (Powell, 2000). Home range areas (in m 2 ) were calculated using a GIS software (ESRI ArcView, version 3.2), either for all individuals together or separated by sexes. We retained 95% of the point locations per individual, using CALHOME software (Kie et al., 1996). At least 30 independent fixes were retained for analysis for each individual, with an interval between two consecutive fixes of at least 2 days to avoid data pseudoreplication. Overlap area between the home ranges of two different individuals was measured in m 2 using GIS (ESRI ArcView, version 3.2), and the percentages relative to the means of the two areas considered were then calculated (Ferner, 1974; Brown et al., 1995). In practice, percent overlap is the percent of focal individual s home range shared with one or more other individuals of the specified sex: Percent overlap = ((X Y 1 ) (X Y 2 ) (X Y n ))/ X 74
where we define X, Y 1,, Y n to indicate the home range area of the focal individual X and the overlapping tortoises Y 1 through Y n, and where the symbols in the above-formula define intersection ( ) and union ( ) (see also Abell, 1999). These overlap percentages were used for the descriptive analyses. Overlaps were calculated for the following three groups: (i) male vs. male, (ii) female vs. female, and (iii) male vs. female. Inter-group differences were analysed, due to the very small sample sizes, by a Monte Carlo randomisation procedure of Mann-Whitney U-test, with 10,000 permutations (Gotelli & Graves, 1996). All statistical tests were performed using a STATISTICA 7.1 software, with all tests being two-tailed and alpha set at 5%. RESULTS Mean home range area was 21 100 m 2 after pooling males and females. Home range area averaged larger in males than in females (Tab. I), but there was a substantial inter-individual difference in home range size even within sexes. Overall, the home range sizes of males and females were not significantly different (Monte Carlo Z = -0.775; exact p = 0.667, range of confidence limits for p after 10,000 randomisations = 0.657-0.669), although this may have obviously depended on the very small sample sizes. The smallest home range was recorded for T3 that had a damaged carapace. This sub-optimal physical condition may have affected her homing performance, thus explaining why her home range was much smaller than that of the other three individuals. In figure 1 we show the GIS home range profiles for the two males and the females, including the relative overlaps. The highest overlap was observed between the home ranges of Table I Descriptive statistics for the home range areas (m 2 ) of male and female Testudo hermanni N indiv. Mean Min Max Variance S.D. S.E. total 4 21140 3766 47226 388.04 19.70 9.85 males 2 27765 8304 47226 756.46 27.52 19.46 females 2 14515 3766 25265 231.10 15.20 10.75 S.D. = Standard deviation; S.E. = Standard error. Mcpt3.shp Mcpt2.shp Mcpt4.shp Mcpt1.shp N 200 0 200 400 Meters Figure 1. Home ranges, and their relative overlaps, among the various radio-tracked tortoises in the study area. Individuals named T1 and T3 were females, and those named T2 and T4 were males. Symbols: Mcpt1.shp = home range of T1; Mcpt2.shp = home range of T2; Mcpt3.shp = home range of T3; Mcpt4.shp = home range of T4. 75
the two males (T2-T4) with an overlapped area of 830 m 2 (Fig. 2). The T4 home range was also overlapped with those of the two females, and especially with that of T1 (226 m 2, but only 11 m 2 with T3; Fig. 2). The percent overlaps among the three categories considered in this study (Tab. II) were on average very low (0.7%), this evidence being clear also with regard to malemale overlaps (0.3%). Figure 2. Histogram showing the overlaps (m 2 ) between home range areas of the various individuals radio-tracked during the present study. Individuals: T1 and T3 were adult females, T2 and T4 were adult males. Table II Descriptive statistics for the overlaps (%) between pairs of home ranges, subdivided by categories N Mean Min Max Variance S.D. S.E. TOT 6 0.73 0 2.98 1.33 1.15 0.47 C1 1 2.98 2.98 2.98 / / / C2 1 0 0 0 / / / C3 4 0.36 0 0.76 0.15 0.39 0.19 Category symbols: C1 (Male-Male); C2 (Female-Female); C3 (Male-Female). S.D. = Standard deviation; S.E. = Standard error. DISCUSSION A first evidence of this study is that male autumnal home ranges were slightly larger than female ones, although not at a statistical significant level. This relative inter-sexual difference in home range area mirrors evidence already got from other reptiles, with sex being one of the main determinants of inter-individual differences in home range area (e.g., Rose, 1982; Perry & Garland, 2002 for reviews). With regard to Testudo hermanni, literature data are quite contrasting. Some studies reported no difference between sexes (Longepierre et al., 2001; Hailey, 1989; Swingland & Stubbs, 1985), whereas others showed larger home ranges in males (Swingland & Stubbs, 1985; Calzolai & Chelazzi, 1991; Mazzotti et al., 2002). Conversely, in other tortoises home range sizes are largest in the females (Lagarde et al., 2003). The home ranges of the two males were also clearly different from each other, thus suggesting the existence of diverging homing patterns between these males (as also revealed by 76
the overlap data). Indeed, male T4 had a substantial home range overlap with the two females, and was sometimes observed in the vicinity of these and other females (also trying to copulate), whereas the male T2 did not. Thus, we suggest that T4 was reproductively active during autumn, whereas the same was not true for T2. The mating season for these tortoises is generally described in May and in August-September (Mazzotti & Vallini, 2000), but it has also been suggested to be continuous for the whole active season, from spring to autumn (Vetter, 2006). This latter is certainly the case in Mediterranean central Italy (Luiselli et al., unpublished data). Overall, this study also confirms that there is a huge inter-individual difference in home range size among tortoises even within the same population. For instance, Moskovitz & Kiester (1987) found a 200-fold difference in individual Geochelone home range sizes, with several individuals exceeding home range size of Testudo hermanni in the present study. Similar conclusions were also reached with radio-tracked tropical tortoises (Kinixys erosa and Kinixys homeana, see Lawson, 2006) as well as temperate tortoises (Smith et al., 1997), thus showing that local bioclimate is not the determinant factor for the magnitude of tortoise movements. We suppose that the movements of tortoise individuals may instead be potentially constrained by the local availability of suitable retreats (e.g., burrows, large bushes) or suitable hibernacula. ACKNOWLEDGEMENTS We thank the Agenzia Regionale per i Parchi, Regione Lazio (Rep. 139-RACC.3 to LL), and the Parco Naturale Regionale Marturanum (G.M. no. 108, August 03, 2007, to LL) for funding the present project. We are indebted to Prof. Brenda Bolton (University College, London) for having corrected the English style, and to Prof. Christian Erard (Muséum National d Histoire Naturelle, Brunoy) and to the anonymous reviewers for helpful comments on the submitted draft. REFERENCES Abell, A. (1999). Male-female spacing patterns in the lizard, Sceloporus virgatus. Amphibia-Reptilia, 20: 185-194. Blasi, C. (1993). Carta del fitoclima del Lazio (scala 1:250.000). Reg. Lazio, Dip. Biologia Vegetale Univ. La Sapienza. Tip. Borgia, Roma. Bonnet, X., Bradshaw, D. & Shine, R. (1998). Capital versus income breeding: an ectothermic perspective. Oikos, 83: 333-342. Brow n, R.M., Gist, D.H. & Taylor, D.H. (1995). Home range and ecology of an introduced population of the European wall lizard Podarcis muralis (Lacertilia, Lacertidae) in Cincinnati, Ohio. Am. Midl. Nat., 133: 344-359. Bruno, S. & Maugeri, S. (1977). Rettili d Italia. Martello, Florence. Calzolai, R. & Chelazzi, G. (1991). Habitat use in a central Italy population of Testudo hermanni (Reptilia Testudinidae). Ethol. Ecol. Evol., 3: 1-14. Corbett, K. (1989). Conservation of the European Reptiles and Amphibians. Christopher Helm, London. Ferner, J.W. (1974). Home-range size and overlap in Sceloporus undulatus erythrocheilus (Reptilia: Iguanidae). Copeia, 1974 (3): 332-337. Gotelli, N.J. & Gr av e s, G.C. (1996). Null models in ecology. Smithsonian Institution Press, Washington D.C. Hailey, A. (1989). Routine movements in a tortoise. Can. J. Zool., 67: 208-215. Hailey, A. (2000). The effects of fire and mechanical habitat destruction on survival of the tortoise Testudo hermanni in Northern Greece. Biol. Cons., 92: 321-333. Hayne, D.W. (1949). Calculation of size of home range. J. Mammal., 30: 1-18. Kie, J.G., Baldwin, J.A. & Evans, C.J. (1996). CALHOME: a program for estimating animal home ranges. Wildl. Soc. Bull., 24: 342-344. Lagarde, F., Bonnet, X., Henen, B., Legrand, A., Corbin, J., Nagy, K. & Naulleau, G. (2003). Sex divergence in space utilisation in the steppe tortoise (Testudo horsfieldi). Can. J. Zool., 81: 380 387. Law s o n, D.P. (2006). Habitat use, home range, and activity patterns of hingeback tortoises, Kinixys erosa and K. homeana, in Soutwestern Cameroon. Chelon. Cons. Biol., 5: 48-56. Longepierre, S., Hailey, A. & Grenot, C. (2001). Home range area in the tortoise Testudo hermanni in relation to habitat complexity: implications for conservation of biodiversity. Biodiv. Cons., 10: 1131-1140. Luiselli, L. (1992). Reproductive success in melanistic adders: A new hypothesis and some considerations on Andrén and Nilson s (1981) suggestions. Oikos, 64: 601-604. 77
Luiselli, L. (1995). The mating strategy of the European adder, Vipera berus. Acta Oecol., 16: 375-388. Madsen, T. (1984). Movements, home range size and habitat use of radio-tracked grass snakes (Natrix natrix) in southern Sweden. Copeia, 1984 (4): 707-713. Mazzotti, S. (2004). Hermann s tortoise (Testudo hermanni): current distribution in Italy and ecological data on a population from the north Adriatic coast (Reptilia, Testudinidae). Ital. J. Zool., 71 (suppt. 1): 97-102. Mazzotti, S, Pisapia, A. & Fasola, M. (2002). Activity and home range of Testudo hermanni in Northern Italy. Amphibia-Reptilia, 23: 305-312. Mazzotti, S. & Vallini, C. (2000). Seasonal activity and thermal relations of Testudo hermanni Gmelin in bare patches of the Bosco della Mesola (Po Delta, Northern Italy). Atti del I Congresso Nazionale della Societas Herpetologica Italica (Torino, 1996), Mus. reg. Sci. nat. Torino: 133-137. Moskovitz, D.K. & Kiester, A.R. (1987). Activity levels and ranging behaviour of the two Amazonian tortoises, Geochelone carbonaria and Geochelone denticulata, in north-western Brazil. Funct. Ecol., 1: 203-214. Pe r ry, G. & Garland, T. (2002). Lizard home range revisited: effect of sex, body size, diet, habitat and phylogeny. Ecology, 83: 1870-1885. Powell, R.A. (2000). Animal home ranges and territories and home range estimators. Pp. 243-269 In: L. Boitani & T.K. Fuller (eds.), Research techniques in animal ecology: controversies and consequences. Columbia University Press, New York. Rose, B. (1982). Lizard home ranges: methodology and functions. J. Herp., 16: 253-269. Row, J.R. & Blouin-Demers, G. (2006). Kernels are not accurate estimators of home range for Herpetofauna. Copeia, 2006 (4): 797-802. Smith, R.B., Breininger, D.R. & Larson, V.L. (1997). Home range characteristics of radiotagged gopher tortoises on Kennedy Space Center, Florida. Chelon. Cons. Biol., 2: 358-362. Swingland, I.R. & Stubbs, D. (1985). The ecology of the Mediterranean tortoise (Testudo hermanni): reproduction. J. Zool. (London), 205: 595-610. Vetter, H. (2006). Testudo hermanni. Reptilia Ediciones, Barcelona. Zuffi, M.A.L. & Odetti, F. (1998). Double egg-deposition in the European pond turtle, Emys orbicularis, from Central Italy. Ital. J. Zool., 65: 187-189. 78