Chelonoidis carbonaria (Testudines: Testudinidae) Activity Patterns and Burrow use in the Bolivian Chaco

Chelonoidis carbonaria (Testudines: Testudinidae) Activity Patterns and Burrow use in the Bolivian Chaco Author(s): A.J. Noss, R.R. Montaño F., F. Soria, S.L. Deem, C.V. Fiorello, L.A. Fitzgerald Source: South American Journal of Herpetology, 8(1):19-28. 2013. Published By: Brazilian Society of Herpetology DOI: http://dx.doi.org/10.2994/sajh-d-12-00028.1 URL: http://www.bioone.org/doi/full/10.2994/sajh-d-12-00028.1 BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder. BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

2013 Brazilian Society of Herpetology Chelonoidis carbonaria (Testudines: Testudinidae) Activity Patterns and Burrow Use in the Bolivian Chaco A.J. Noss 1, *, R.R. Montaño F. 2, F. Soria 3, S.L. Deem 4, C.V. Fiorello 5, L.A. Fitzgerald 6 1 University of Florida, Department of Geography, P.O. Box 117315, Gainesville, FL 32611, USA. 2 Wildlife Conservation Society-Bolivia, Santa Cruz, Casilla 6272, Bolivia. Email: rovina7@yahoo.com 3 Capitanía de Alto y Bajo Isoso, Casilla 3108, Santa Cruz, Bolivia. 4 Institute for Conservation Medicine, Saint Louis Zoo, One Government Drive, Saint Louis, Missouri, USA. Email: deem@stlzoo.org 5 Wildlife Health Center, School of Veterinary Medicine, University of California, Davis, California, USA. Email: drfiorello@gmail.com 6 Department of Wildlife and Fisheries Sciences, Texas A&M University, College Station, Texas, USA. Email: lfitzgerald@tamu.edu * Corresponding author. Email: anoss@ufl.edu Abstract. In the Bolivian Chaco, the tortoise Chelonoidis carbonaria is an important reptile for indigenous people for subsistence purposes and in traditional medicine. This article describes research on seasonal activity, daily activity, and burrow use for the species at two longterm research camps in the Kaa-Iya del Gran Chaco National Park. The principal research method is the use of internal and external temperature dataloggers in 15 individuals over a two-year period. Tortoises reduce their activity in the dry season, and are not active when air temperatures are below 20 C or above 37 C, though they can be active outside burrows at any time of year. Body temperature varies from 4 C in winter (monthly average of minimum daily temperatures) to 38 C in summer (monthly average of maximum daily temperatures). The instantaneous difference between body and environmental temperature could be as great as 23 C or +12 C, but the monthly average was between 2 C and +4 C. Tortoises rely principally on shelters including fallen trees, dense bromeliad ground cover, and leaf mulch; but also use armadillo burrows and rock crevices. They use multiple shelters / burrows over time, occasionally sharing shelters with other individuals. Burrow use is important both in cold periods as well as in hot and dry periods. The dataloggers provided extremely detailed data on body and environmental temperature, but the implantation caused problems in several individuals and the procedure may need modifications. Keywords. Bolivia; Daily Activity; Dataloggers; Kaa-Iya; Seasonal Activity. Resumen. En la región del Chaco boliviano la tortuga negra Chelonoidis carbonaria es un reptil importante para los pueblos indígenas por motivos de subsistencia y medicina tradicional. El presente artículo detalla estudios de actividad estacional, actividad diaria, y termo-regulación para la especie en dos campamentos de investigación a largo plazo del Parque Nacional Kaa-Iya del Gran Chaco. El método principal empleado ha sido el uso de termómetros externos y corporales (implantados mediante cirugía) en quince individuos durante dos años. Las tortugas disminuyen su actividad en la época seca, pero pueden estar activas y se encuentran fuera de cuevas todo el año. La temperatura corporal de las tortugas varía de 4 C en invierno (promedio mensual del mínimo diario) a 38 en verano (promedio mensual del máximo diario). La diferencia instantánea entre temperatura ambiental y temperatura corporal varía de 23 C a 12 C, aunque el promedio mensual es entre 2 C y 4 C. Los dataloggers proveen información extremadamente detallada sobre temperatura corporal y ambiental, pero la implantación causó problemas en varios individuales por lo cuál se recomiendo modificar la metodología. INTRODUCTION Tortoises worldwide are under threat from a variety of human pressures: hunting for meat or eggs by humans, or by their domestic and feral animals; the pet trade; habitat loss and habitat fragmentation; pollution; disease; etc. (Buhlmann et al., 2009; Rhodin et al., 2011). The study was part of a larger project to promote indigenous community management of indigenous lands and protected areas in the Bolivian Chaco. The red-footed tortoise Chelonoidis carbonaria (Spix, 1824) (previously Geochelone carbonaria; see Le et al., 2006) is reported as a game species for subsistence and commercial hunters in Venezuela (Aponte et al., 2003) and Brazil (Peres, 2000; Pezzuti et al., 2010). In the Bolivian Chaco, it is an important subsistence resource for the Ayoreo indigenous people (Ayala, 1997; Etacore et al., 2000), and its fat is used in traditional medicine by the Isoseño-Guaraní indigenous people (Cuéllar, 2000; Soria and Noss, 2000). Preliminary observations and interviews in the Kaa-Iya landscape suggested the species to be much more abundant in protected areas free of hunting and livestock pressure, despite assertions by local Isoseño-Guaraní residents that they did not consume its meat. The species is considered to be near threatened in Bolivia (Cortéz and Rey-Ortíz, 2009). It is also important as a prey item for jaguars Panthera onca in this region (L. Maffei, pers. comm.), and as a seed disperser (Moskovits and Bjorndal, 1990; Strong and Fragoso, 2006; Wang et al., 2011). The few field studies on this species have focused on reproduction, diet, seed dispersal, ranging behavior and abundance. Most research has concentrated in humid Amazonian forests (Medem et al., 1979; Moskovits and Kiester, 1987; Moskovits, 1985, 1988; Bjorndal, 1989; Moskovits and Bjorndal, 1990; Strong and Fragoso, 2006), with only one study in seasonally flooded Pantanal savanna (Wang et al., 2011) and one in dry forest (Aponte Submitted on: 08 November 2012 Accepted on: 31 January 2013 Handling Editor: Márcio Martins doi: 10.2994/SAJH-D-12-00028.1

et al., 2003). The Gran Chaco represents the southern distribution limit for Chelonoidis carbonaria, where it faces the most extreme climatic conditions in terms of cold and heat (<0 40 C+) and seasonal drought (six months or more without surface water). This paper provides information on how an ectothermic species adapts to these conditions, using for the first time internal and external temperature dataloggers to make simultaneous comparisons of body and ambient temperature across seasons and across individuals (male and female) at two widely separated and protected sites. Ectothermic animals thermoregulate through a series of physiological as well as behavioral mechanisms (Brattstrom, 1965; Zimmermann et al., 1994). This study focuses on the latter, in particular activity patterns combined with the use of burrows or shelters, in radio-tracked individuals. A complementary paper reports on ranging patterns (Montaño et al. in press). Study area The Kaa-Iya del Gran Chaco National Park covers 34,400 km 2 in eastern Santa Cruz Department. This is the most diverse and best-conserved portion of the Gran Chaco ecoregion (Taber et al., 1996), with over 50 reptile species recorded including four chelonians: Chelonoidis carbonaria, C. chilensis, Kinosternon scorpioides and Acanthochelys macrochephala (Gonzáles, 1998, 2001). Of these four, C. carbonaria is by far the most abundant and widely distributed across the Kaa-Iya National Park (Montaño, 2003; Gonzáles and Montaño, 2007; Montaño and Gonzáles, 2007, 2008). We established the Cerro Cortado field camp (19 31 35.9 S, 62 18 34.5 W) in 1997 on the boundary between the Kaa-Iya del Gran Chaco National Park and the neighboring Isoso TCO (indigenous territory). We established Ravelo field camp (19 17 43.2 S, 60 31 10.6 W) in 2000 in the southeastern portion of the Kaa-Iya National Park, 15 km from the Paraguay border, and over 170 km from the Cerro Cortado camp. The Chaco alluvial plain lies at an altitude of approximately 300 msl, with isolated hills (Cerro Cortado, Cerro Colorado, Cerro San Miguel) at the study sites that rise 100 200 m above the surrounding plain. Average annual rainfall is 550 mm at Cerro Cortado and 700 mm at Ravelo, with a dry season from May to November. The vegetation at Cerro Cortado is a xeric Chaco forest, with low forests and thorn scrub distributed according to soil types (sand versus clay). The canopy is 3 8 m high with a predominance of Acacia praecox (Leguminosae-Mimosoidae), Ruprechtia triflora (Polygonaceae) and Castela coccinea (Simaroubaceae). Emergents up to 20 m high are Cereus validus, Stetsonia coryne (Cactaceae), Aspidosperma quebracho-blanco, A. pirifolium, Caesalpinia paraguariensis (Leguminosae-Caesalpiniaceae) and Ziziphus mistol (Rhamnaceae). The vegetation at Ravelo is a transitional Chaco-Cerrado dry forest, with a canopy 10 15 m high, rich in vines but poor en epiphytes. The most abundant tree species include Astronium urundeuva (Anacardiaceae), Cordia glabrata (Boraginaceae), Aspidosperma cf. tomentosum, Caesalpinia pluviosa (Leguminosae-Caesalpiniaceae), Anadenanthera macrocarpa (Leguminosae-Mimosoideae) and Amburana cearensis (Leguminosae-Papilionoideae). Ground cover at both sites includes numerous species of cactus and bromeliads (Navarro, 2004). MATERIALS AND METHODS We recorded activity data and captured tortoises encountered opportunistically during line transect surveys and other field activities and obtained a body weight using a spring scale. In many tortoise studies, telemetry transmitters are attached using epoxy only (Boarman et al., 2008). We decided to combine this method with bolts in order to avoid detachment by tortoises pushing their way through the dense and thorny Chaco vegetation at our study sties, during tracking periods extending over a year. We anesthetized the tortoises in order to minimize pain from drilling the holes through the scutes, which occasionally induces minor bleeding, and to surgically implant internal thermometers. These protocols were approved by the Texas A&M University Institutional Animal Care and Use Committee / Animal Welfare Assurance Program (2000 230). In a first round of surgeries we anesthetized the animals using an intramuscular (I.M.) injection of Domitor (Medetomidine hydrochloride, 150 mcg/kg) combined with Ketamine (7.5 mg/kg). In all subsequent surgeries we anesthetized the animals using an I.M. injection of either Telazol (Tiletamine-Zolazepam, 20 mg/kg) or Ketamine (7.5 mg/kg) with Medetomidine (75 mcg/kg), combined with Isoflurane administered using a cotton ball coated with 0.25 ml of liquid Isoflurane and placed within a syringe case and positioned near the nares as previously described by Deem and Fiorello (2002). Additionally, approximately 0.8 mg Lidocaine was injected subcutaneously along the surgical incision in a few animals. When the tortoises were fully anesthetized, personnel used glue and bolts to attach a Holohil or ATS radio transmitter and an Onset Tidbit temperature datalogger to the dorsal aspect of the carapace. A second, sterilized datalogger was implanted subcutaneously anterior to the left hind limb in the first two rounds of surgeries, and intracoelomically in the subsequent rounds, using sterile techniques. Following the first round of surgeries when the units were gas-sterilized, the internal dataloggers were coated with Elvax an ethylene-vinyl acetate copolymer resin that is very smooth and extremely repellent of water. The external radio-transmitter allowed us to recover 20

the dataloggers and to conduct a parallel ranging behavior study. Those tortoises that were anesthetized with the Domitor and Ketamine/Medetomidine injectables were given an I.M. Antisedan /Atipamazole (375 mcg/kg) to reverse the Medetomidine. When the tortoises were fully anesthetized, glue and bolts were used to attach a Holohil or ATS radio transmitter to the dorsal aspect of the carapace. Tortoises were kept in captivity until full recovery from anesthesia, which ranged between 24 72 hours, and released at their capture location. We then obtained daily locations for the tortoises by tracking the radio signals with a Telonics receiver and a Yagi two-element directional antenna. After locating the animal visually, we took the coordinates of the precise location with a GPS receiver, and recorded the date, time and the tortoise s activity. If the tortoise was inside a burrow, we described and measured (entrance width and height) the burrow and marked it with flagging and a unique number. Onset Tidbit temperature dataloggers had a range of -20 50 C, accuracy of ±0.4 at 20 C, a resolution of 0.3 at 20 C. They measured 3.0 4.1 1.6 cm, and weighed 19.5 g. We programmed both temperature dataloggers to record the temperature every 12 (or 15) minutes, with a capacity for 35,000 records during eight months of monitoring. This methodology was initially developed in alligator snapping turtle Macroclemys temminckii (Fitzgerald and Nelson, 2011), and parallels a pioneering study of deep body versus air and ground temperatures in Galápagos tortoises Chelonoidis elephantopus (MacKay, 1964). Before the datalogger memory filled up, we located the animals using a radio-telemetry receiver. We removed the external datalogger and performed a second surgery to remove the internal datalogger. Using an infrared reader we downloaded the following data from the dataloggers: temperature, time and date. We analyzed these data in order to describe daily and seasonal activity patterns. We also compared simultaneous internal with external/ambient temperatures in order to evaluate thermoregulation, particularly in relation to tortoise activity, use of burrows or shelters, and use of micro-habitats. For purposes of simplification, throughout this paper we use internal temperatures for tortoise body temperatures, and external temperatures as ambient temperatures. burrows utilized by the nine radio-tracked tortoises at Cerro Cortado, we were able to analyze burrow use by individuals and by seasons. The six animals that we tracked for a year or more at Cerro Cortado each used from 4 18 different burrows. With the exception of male C1 who did not appear to utilize a fixed home range, the other tortoises returned to use previously occupied burrows, and also shared burrows with other individuals of the same and opposite sex. Figure 1 presents the seasonal pattern of burrow use for the nine tortoises radio-tracked at Cerro Cortado. The months of June-August coincide with the Chaco winter, when we would expect higher burrow use as a protection against temperatures that occasionally drop below freezing. We were surprised by the importance of burrows in September and in January, possibly because of the high temperatures and lack of rain. In contrast, burrow use declined during the rainy months (December April). Not all individuals followed the general pattern, however (Fig. 2). Even when winter temperatures dropped below freezing Figure 1. Seasonal burrow use by Chelonoidis carbonaria: average number of observations in burrows per month (N = 9 individuals). RESULTS Burrow use We measured 27 burrows, with a width from 20 100 cm (x = 39.7 cm, standard deviation 19.8) and a height from 17 200 cm (37.5 cm, 34.0). The burrows occurred in soil, in hollow or decayed logs, or in rock. Several burrows in Ravelo were originally dug by the giant armadillo Priodontes maximus. By marking and numbering Figure 2. Proportion of observations when individuals were found in burrows, Cerro Cortado. N = 40 or more observations per individual per season, excluding the wet season for tortoises 7C, 8C and 9C. 21

at Cerro Cortado, tortoises 1C and 3C did not use burrows. Instead, 1C in particular tended to use shelters against fallen or standing trees during the hotter summer months. Using a rock shelter or a hollow tree at Ravelo, tortoises 1R and 2R s external dataloggers recorded temperatures 5 10 C lower than air temperature in August and September 2001. At Cerro Cortado, tortoise 4C used earth burrows or hollow trees to register similar temperature spreads in October 2000 and March 2001, while tortoise 5C did so in April May and again in August-September 2001. Direct observations of the 15 radio-tracked tortoises across both sites suggested that tortoises are not active when air temperatures are below 20 or above 37 C. Tortoises are most likely to be active at air temperatures between 24 29 C. Body and ambient temperature From January 2000 to January 2002, we recovered external dataloggers from seven tortoises at Cerro Cortado (6 21 months per animal, 118 animal-months in total) and five tortoises at Ravelo (5 9 months per animal, 38 animal-months in total). At Cerro Cortado the tortoises faced temperaturas below 0 C (June July), and exceeding 40 C during the austral summer months (Fig. 3). Average monthly ambient temperatures ranged between 8 28 C. Cerro Cortado presented greater temperature extremes, both hotter and colder, than did Ravelo. During the same study period we recovered internal dataloggers from seven tortoises at Cerro Cortado (6 15 months per animal, 78 animal-months in total) and from four tortoises at Ravelo (5 8 months per animal, 29 animal-months in total). During the winter months, variation was greater both in body temperature across individuals, and in the range from minimum to maximum temperatures per individual. In general, average maximum monthly body temperatures ranged from 32 38 C throughout the year. Minimum recorded body temperatures were as low as 4 7 C in June July, but rose to 20 24 C in January February (Fig. 4). Tortoises maintained average monthly body temperatures between 17 29 C. In all cases when we recovered internal dataloggers, we also recovered the external dataloggers, therefore we could compare simultaneous body and ambient temperatures. Subtracting ambient from body temperature, the difference ranged from -23 12 C. However, the average monthly difference was between -2 and 4 C. Figure 5 presents observations by month of the 9 radio-tracked animals at Cerro Cortado, when we recorded whether the animal was active when it was encountered. According to these observations, Chelonoidis carbonaria activity is concentrated in the rainy season (November April), with the probability of encountering an animal Table 1. Chelonoidis carbonaria individuals tracked at Cerro Cortado (C) and Ravelo (R): duration and number of dataloggers recovered. No. Sex Weight (kg) Start date End date Months Datalogger internal/ external C1 M 6.0 10/1/00 24/3/01 14 1/1 C2 F 6.9 12/1/00 18/10/01 21 1/3 C3 F 8.5 10/1/00 13/10/01 21 3/2 C4 M 4.8 13/1/00 25/3/01 14 1/3 C5 F 6.2 13/1/00 14/10/01 21 1/3 C6 F 9.0 13/1/00 10/10/01 21 2/3 C7 M 7.0 6/4/01 22/6/01 2 C8 M 2.6 4/4/01 15/10/01 6 1/1 C9 M 16/4/01 14/10/01 6 R1 F 6.0 28/5/01 10/1/02 8 1/1 R2 M 6.5 27/5/01 10/1/02 8 1/1 R3 F 5.0 27/5/01 14/01/03 19 1/0 R4 M 5.5 28/5/01 10/1/02 8 1/1 R5 F 4.5 28/5/01 1/8/01 (lost) 2 R6 M 4.9 27/5/01 1/10/01 (died) 4 1/1 Total 15 167 active exceeding 80% for females in February and March, and reaching 100% for males in March. We observed copulation in January, February and April. However, tortoises were active during all months of the year, even during the coldest and driest periods. On occasion, we found tortoises in the open but inactive, particularly during cold snaps, thus inactivity did not coincide necessarily with burrow use. Overall throughout the year tortoises were active on 28% of occasions they were observed at Cerro, though activity in the wet season was much higher, 63 64% for males and females, and lower in the dry season, only 8% for males and 18% for females. Figure 6 presents daily activity observations of the 15 radio-tracked animals at both sites, when we recorded whether the animal was active when it was encountered. These records also suggest that Chelonoidis carbonaria is active throughout the day, with a slight increase in the afternoon, especially in the wet season from December April. We did not track tortoises at night to check for nocturnal activity. At any time of day, activity was higher in the wet season. Eight tortoises at both sites presented the dominant daily pattern in Fig. 7: body temperature higher than ambient temperature (up to nearly 4 C) during the night, and the opposite during the day (up to nearly 3 C). But three additional patterns also occurred: 1) the pattern for tortoise 6C from Cerro Cortado included two peaks, 2) the body temperature of tortoise 1R from Ravelo always exceeded ambient temperature, and 3) ambient temperature always exceeded body temperature of tortoise 6R from Ravelo. It is possible that the last animal, found dead after four months of tracking, was suffering during the study period from a disease. 22

Figure 3. Ambient temperatures ( C) at Cerro Cortado (A) and Ravelo (B) recorded by external dataloggers on Chelonoidis carbonaria. 23

Figure 4. Body temperatures ( C) of Chelonoidis carbonaria recorded by internal dataloggers at Cerro Cortado (A) and Ravelo (B). 24

DISCUSSION Radio-telemetry has been successfully used in several tortoise and turtle species to describe seasonal and daily activity patterns as well as burrow use: Gopherus polyphemus in the southeastern USA (Eubanks et al., 2003), G. agassizii in the southwestern USA (Bulova, 1994; Averill-Murray et al., 2002a, b), Testudo hermanni in Italy (Mazzotti et al., 2002), T. kleinmanni in Israel (Geffen and Mendelssohn, 1989), as well as Chelonoidis carbonaria and C. denticulata in Brazil (Moskovits and Kiester, 1987). We found the external use of glue and bolts to be very effective, though glue alone may be sufficient for most tortoise studies (Boarman et al., 1998). The methodology of fixing a combination of external radio-transmitters as well as external and internal dataloggers has been applied previously in other reptiles. Fitzgerald and Nelson (2011) developed the methodology with the aquatic turtle Macroclemys temminckii in the United States in order to evaluate micro-habitat selection and to understand this species thermoregulatory patterns. MacKay (1964), using ingested temperature telemetry devices in Chelonoidis elephantopus, similarly confirmed the tortoises ability to maintain relatively stable core body temperatures (28 32 C) with brief movements into sun and shade respectively, despite greater swings in ambient temperatures (23 42 C). The implantation of dataloggers requires specialized veterinarians, yet despite this assistance in our case it is probably that several animals suffered negative reactions to the surgery which may have led to the death of two individuals. The first anesthesia protocol, Domitor / Ketamine, though extremely successful in Macroclemys temminckii (Fitzgerald and Nelson, 2011), did not eliminate strong limb reflexes in Chelonoidis carbonaria despite doubling the dosage. The second anesthesia protocol was much more effective, though a side effect of Isoflurane was that the tortoises expelled mucous from the nares during anesthesia. However, in 7 of 13 (53%) tortoises that received subcutaneous implantations, the internal datalogger was dislodged from the subcutaneous space or not found at all. The use of Elvax did not relieve this problem. In 4 of 6 (66%) intracoelomic implantations, the internal datalogger was either dislodged from its original site and free in the coelomic cavity, walled off in a pseudocapsule, or expelled from the coelomic space and either in the subcutaneous space or not found at all. This was more common in females (4 of 5 individuals) than males (1 of 2 individuals), and might be related reproductive contractions and movement associated with egg production. However, based on the one male that experienced this problem, the complication was not limited to females. At least one tortoise (C6) developed a profound fibrinous peritonitis after instrument implantation, and a few animals had Figure 5. Proportion of observations by month when radio-tracked tortoises were active (N = 9 individuals, 1273 observations, Cerro Cortado only). Figure 6. Proportion of observations by time period when tortoises were active (radio telemetry observations at Cerro (Jan Dec, N = 1271) and Ravelo (Dec Apr = wet season only, N = 283). Figure 7. Daily temperature patterns of Chelonoidis carbonaria: difference between body and ambient temperatures ( C). abundant fibrous tissue associated with the incision and implant. This reaction may have been a result of infection or a reaction to the wax used to coat the datalogger. In one tortoise that was implanted with a silicone-coated, 25

gas-sterilized datalogger, a gross tissue response was not seen at the time of explantation. For these reasons, we recommend avoiding surgical implantation in future studies of Chelonoidis carbonaria. Implantation has also proved problematic in other turtle studies, for example radio transmitters in oviducts of tropical freshwater turtles Chelodina rugosa which resulted in oviducal adhesion in two cases, reduced reproductive output, and one death from surgery (Kennett et al., 1993). Researchers and animal care and use committees may wish to revisit their protocols regarding surgical implantation procedures. Daily and seasonal activity of Chelonoidis carbonaria in the Chaco is consistent with that reported for the same species in Brazil (Moskovits, 1985; Moskovits and Kiester, 1987) and for other tortoises elsewhere: considerable individual variation in ranging patterns and burrow use, increased activity in seasons with higher temperatures but especially higher rainfall, days of inactivity, and use of multiple refuges which often provide only concealment and shade rather than protection (Stickel, 1950; Bertram, 1979; Hailey, 1989; Hailey and Coulson, 1995, 1996, 1999; Eubanks et al., 2003; Luiselli 2003a, b, 2005). Activity patterns by sex do not follow the same pattern as in Testudo graeca or T. hermanni where females tend to be more active in the middle of the day and males in the morning and late afternoon (Lambert, 1981; Mazzotti et al., 2002). The tolerance for heat would appear to be higher in the larger Chelonoidis carbonaria, with reduced activity beyond 29 C but continued activity up to air temperatures of 37 C, as compared to most other species reported in the literature. In Kinixys spekii, activity drops off sharply when air temperatures exceed 29 C (Hailey and Coulson, 1996), whereas in Testudo graeca and T. kleinmanni individuals seek refuges when air temperatures exceed 28 C (Lambert, 1981; Geffen and Mendelssohn, 1989). The exception is the desert tortoise Gopherus assizii, reported to be active at air temperatures up to 45 C (Averill-Murray et al., 2002a). Red-footed and desert tortoises are equally sensitive to cold temperatures, with little activity below 20 C, in comparison with T. graeca and T. kleinmanni with activity starting at 17 18 C. Mean field body temperature is lower for C. carbonaria in the Chaco, 25.4 C (according to internal dataloggers) as compared to 27 C for Kinixys spekii, 28 28.4 C for K. homeana, 28.6 29.2 C for K. erosa, 29 C for Testudo kleinmanni and T. graeca, 31 C for Gopherus flavomarginatus, and 32.5 C for Stigmochelys pardalis, the latter five species according to thermometers in the cloaca (Meek and Jayes, 1982; Rose, 1983; Geffen and Mendelssohn, 1989; Hailey and Coulson, 1996; Luiselli 2005). However, using temperature transmitters ingested for periods of 5 19 days, Moskovits (1985) reports slightly higher mean body temperatures for C. carbonaria between 27 28.5 C. The dominant temperature pattern found among red-footed tortoises at the two Chaco sites, with body temperature exceeding air temperature early and late in the day, but air temperature exceeding body temperature in the middle of the day, is consistent with that reported by Moskovits (1985) at an Amazonian site. In contrast, hinge-back tortoises Kinixys homeana and K. erosa in a west African forest maintain body temperatures nearly always below ambient temperatures (Luiselli, 2005). Red-footed tortoises in the Chaco were more active than other tortoises for which such data has been collected. We observed tortoises resting on 72% observations overall, 76% in the dry season but only 37% in the wet season, versus 76.3 92.5% of observations overall for the same species at an Amazonian site (Moskovits and Kiester, 1987), 90% of observations for the steppe tortoise Testudo horsfieldi during its fleeting three-month active season (Lagarde et al., 2003), and a remarkable 98% of observations in the case of the desert tortoise G. agassizii which include a period of hibernation (Nagy and Medica, 1986; O Connor et al., 1994, Averill-Murray et al., 2002a). Resource availability in the Chaco is more seasonal than in Amazonia, and therefore may induce greater activity in red-footed tortoises as they seek more patchy and seasonally restricted resources including food and water. However, these resources and climate extremes are not so severe as to produce the extremely sedentary behavior of the desert tortoise in the southwestern USA. Red-footed tortoises do not hibernate, probably because cold snaps and heat waves generated by cold south wind or hot north wind fronts at these Chaco sites do not last more than 2 3 weeks before the wind direction reverses. The longest an individual remained at a single location was 20 days. The desert tortoise Gopherus agassizii uses a similar variety of burrows and shelters (shade, crevices, shrubs), individuals use multiple burrows during the year, and individuals (especially males and females) may share burrows/shelters as we observed with red-footed tortoises in the Chaco (Averill-Murray et al., 2002a, b). A combination of factors including habitat conservation, microhabitat diversity, and hunting protection seem to favor the abundance of African hinge-back tortoises in Nigeria (Luiselli, 2003a). In addition to food and water resources, measures to conserve the red-footed tortoise on community and private ranch lands outside protected areas should ensure the availability of varied burrows and shelters, including leaf mulch, dense undergrowth patches, and fallen trees. Lower abundance in the hunting and livestock areas near communities may be less a consequence of direct hunting pressure than of changes in soil compaction, understory cover and fallen trees resulting from livestock pressure and community collection of firewood. However, indigenous territories and ranches in dry forests where this species occurs will need to set aside extensive communal and private reserves in order to maintain tortoise populations on their lands. 26

ACKNOWLEDGMENTS This publication was made possible in part by financial support from the United States Agency for International Development (USAID, Cooperative Agreement No. 511-A-00-01-00005). The opinions expressed here are those of the authors and do not necessarily represent those of USAID. The Wildlife Conservation Society (WCS) and the Capitanía del Alto y Bajo Isoso provided further support. CABI s parabiologists and wildlife monitors tracked the tortoises with radios. REFERENCES Aponte C.G., Barreto R., Terborgh J. 2003. Consequences of habitat fragmentation on age structure and life history in a tortoise population. Biotropica 35:550 554. doi: 10.1111/j.1744-7429.2003. tb00612.x Averill-Murray R.C., Martin B.E., Bailey S.J., Wirt E.B. 2002a. Activity and behavior of the Sonoran desert tortoise in Arizona. Pp. 135 158, in Devender T. R. (Ed.). The Sonoran desert tortoise: natural history, biology, and conservation. University of Arizona Press, Tucson. Averill-Murray R.C., Woodman A.P., Howland J.M. 2002b. Population ecology of the Sonoran desert tortoise in Arizona. Pp. 109 134, in Devender T. R. van (Ed.). The Sonoran desert tortoise: natural history, biology, and conservation. University of Arizona Press, Tucson. Ayala C.J.M. 1997. Utilización de la fauna silvestre del grupo étnico Ayoréode en la comunidad Tobité, Santa Cruz, Bolivia. Undergraduate thesis, Universidad Autónoma Gabriel René Moreno, Bolivia. Bertram B.C.R. 1979. Home range of a hingeback tortoise in the Serengeti. African Journal of Ecology 17:241 44. doi: 10.1111/ j.1365-2028.1979.tb00260.x Bjorndal K.A. 1989. Flexibility of digestive responses in two generalist herbivores, the tortoises Geochelone carbonaria and Geochelone denticulata. Oecologia 78:317 321. doi: 10.1007/BF00379104 Boarman W.I., Goodlett T., Goodlett G., Hamilton P. 1998. Review of radio transmitter attachment techniques for turtle research and recommendations for improvement. Herpetological Review 29:26 33. Buhlmann K.A., Akre T.S.B., Iverson J.B., Karapatakis D., Mittermeier R.A., Georges A., Rhodin A.G.J., van Dijk P.P., Gibbons J.W. 2009. A global analysis of tortoise and freshwater turtle distributions with identification of priority conservation areas. Chelonian Conservation and Biology 8:116 149. doi: 10.2744/ CCB-0774.1 Bulova S.J. 1994. Patterns of burrow use by desert tortoises: gender differences and seasonal trends. Herpetological Monographs 8:133 143. http://www.jstor.org/stable/10.2307/1467077 Cortéz F. C., Rey-Ortíz G. 2009. Chelonoidis carbonaria (Spix, 1824): Testudines-Testudinidae. Pp. 611 612, in Aguirre L.F., Aguayo R., Balderrama J., Cortéz C., Tarifa T. (Eds.), Libro Rojo de la fauna silvestre de vertebrados de Bolivia. Ministerio de Ambiente y Agua, La Paz. Cuéllar R.L. 2000. Uso de los animales silvestres por pobladores Izoceños. Pp. 471 484, in Cabrera E., Mercolli C., Resquin R. (Eds.). Manejo de fauna silvestre en Amazonía y Latinoamérica. CITES Paraguay, Fundación Moises Bertoni, University of Florida, Asunción. Deem S.L., Fiorello C.V. 2002. Capture and immobilization of freeranging edentates. Document B0135.1202, in Heard D. (Ed.). Zoological Restraint and Anesthesia. International Veterinary Information Service, Ithaca. Accessible at http://www.ivis.org/ special_books/heard/deem/ivis.pdf Etacore J., Higazi A., Beneria-Surkin J., Townsend W.R. 2000. Yoca Iaá Utatai = La Peta Negra: Conocimiento Ayoreo de la Comunidad El Porvenir sobre Cómo Vive la Tortuga Negra (Geochelone carbonaria), Publicaciones Proyecto de Investigación No. 9. Confederación de Pueblos Indígenas de Bolivia, Santa Cruz. Eubanks J.O., Michener W.K., Guyer C. 2003. Patterns of movement and burrow use in a population of gopher tortoises (Gopherus polyphemus). Herpetologica, 59:311 321. doi: 10.1655/01-105.1 Fitzgerald L. A., Nelson R.E. 2011. Thermal biology and temperaturebased habitat selection in a large aquatic ectotherm, the alligator snapping turtle, Macroclemys temminckii. Journal of Thermal Biology 36:160 166. doi: 10.1016/j.jtherbio.2011.01.003 Geffen E., Mendelssohn H. 1989. Activity patterns and thermoregulatory behavior of the Egyptian tortoise Testudo kleinmanni in Israel. Journal of Herpetology 23: 404 409. http:// www.jstor.org/stable/10.2307/1564052 Gonzáles A. L. 1998. La herpetofauna del Izozog. Ecología en Bolivia 31:45 52. Gonzáles A. L. 2001. Los anfíbios y reptiles en una zona del Chaco boreal de Santa Cruz, Bolivia: riqueza, composición y biogeografía. Undergraduate thesis, Universidad Autónoma Gabriel René Moreno, Bolivia. Gonzáles L., Montaño R. 2007. La herpetofauna del Parque Nacional Kaa Iya del Gran Chaco y la Tierra Comunitaria de Origen Isoso, Santa Cruz-Bolivia. Technical Report #193. Wildlife Conservation Society, Museo de Historia Natural Noel Kempff Mercado, Santa Cruz. Hailey A. 1989. How far do animals move? Routine movements in a tortoise. Canadian Journal of Zoology 67:208 15. doi: 10.1139/ z89-028 Hailey A., Coulson I.M. 1995. Habitat association of the tortoises Geochelone pardalis and Kinixys spekii in the Sengwa Wildlife Research Area, Zimbabwe. Herpetological Journal 5:305 309. Hailey A., Coulson I.M. 1996. Temperature and the tropical tortoise Kinixys spekii: constraints on activity level and body temperature. Journal of Zoology240:523 536. doi: 10.1111/j.1469-7998.1996. tb05303.x Hailey A., Coulson I.M. 1999. Measurement of time budgets from continuous observation of thread-trailed tortoises (Kinixys spekii). Herpetological Journal 9:15 20. Kennett R., Christian K., Pritchard D. 1993. Underwater nesting by the tropical freshwater turtle, Chelodina rugosa (Testudinata: Chelidae). Australian Journal of Zoology 41:47 52. doi: 10.1071/ ZO9930047 Lagarde F., Bonnet X., Corbin J., Henen B., Nagy K., Mardonov B., Naulleau G. 2003. Foraging behaviour and diet of an ectothermic herbivore: Testudo horsfieldi. Ecography 26:236 242. doi: 10.1034/j.1600-0587.2003.03365.x Lambert M.R.K. 1981. Temperature, activity and field sighting in the Mediterranean spur-thighed or common garden tortoise Testudo graeca L. Biological Conservation 21:39 54. doi: 10.1016/0006-3207(81)90067-7 Le N., Raxworthy C.J., McCord W.P., Mertz L.A. 2006. A molecular phylogeny of tortoises (Testudines: Testudinidae) based on mitochondrial and nuclear genes. Molecular Phulogenetics and Evolution 40:517 531. doi: 10.1016/j.ympev.2006.03.003 Luiselli L. 2003a. Comparative abundance and population structure of sympatric Afrotropical tortoises in six rainforest areas: the differential effects of traditional veneration and of subsistence hunting by local people. Acta Oecologica 24: 157 163. doi: 10.1016/ S1146-609X(03)00072-9 Luiselli L. 2003b. Seasonal activity patterns and diet divergence of three sympatric Afrotropical tortoise species (genus Kinixys). Contributions to Zoology 72: 211 220. http://dpc.uba.uva.nl/ctz/ vol72/nr04/art02 27

Luiselli L. 2005. Aspects of comparative thermal ecology of sympatric hinge-back tortoises (Kinixys homeana and Kinixys erosa) in the Niger Delta, southern Nigeria. African Journal of Ecology 43:64 69. doi: 10.1111/j.1365-2028.2004.00546.x Mackay R.S. 1964. Galapagos tortoise and marine iguana deep core body temperature measured by radiotelemetry. Nature 204:355 358. doi: 10.1038/204355a0 Mazzotti S., Pisapia A., Fasola M. 2002. Activity and home range of Testudo hermanni in Northern Italy. Amphibia-Reptilia 23:305 312. doi: 10.1163/15685380260449180 Medem F.O., Castaño V., Lugo-R M. 1979. Contribución al conocimiento sobre la reproducción y el crecimiento de los morrocoyes (Geochelone carbonaria y G. denticulata; Testudines, Testudinidae). Caldasia 12:497 511. Meek R., Jayes A.S. 1982. Body temperature and activity patterns of Testudo graeca in north west Africa. British Journal of Herpetology 6:194 197. Montaño F. R. R. 2003. Caracterización de la herpetofauna de la Salina Ravelo, Departamento Santa Cruz, Bolivia. Technical Report #90. Wildlife Conservation Society, Santa Cruz. Montaño R., Gonzáles L. 2007. Diversidad de anfibios y reptiles en el humedal Palmar de las Islas. Technical Report #189. Wildlife Conservation Society, FHF Ramsar, Santa Cruz. Montaño R., Gonzáles L. 2008. El Humedal Palmar de las Islas: una guía ilustrativa. Wildlife Conservation Society, FHF Ramsar, Santa Cruz. Montaño R., Cuéllar E., Fitzgerald L.A., Soria F., Mendoza F., Peña R., Dosapey T., Deem S.L., Noss A.J. In press. Ranging patterns by the red-footed tortoise Geochelone carbonaria in the Bolivian Chaco. Ecología en Bolivia. Moskovits D.K. 1985. The behavior and ecology of the two Amazonian tortoises, Geochelone carbonaria and Geochelone denticulata, in northwestern Brazil. Ph.D. dissertation. University of Chicago, United States. Moskovits D.K. 1988. Sexual dimorphism and population estimates of the two Amazonian tortoises (Geochelone carbonaria and G. denticulata) in northwestern Brazil. Herpetologica 44:209 217. http://www.jstor.org/stable/10.2307/3892519 Moskovits 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. Functional Ecology 1:203 214. http://www.jstor.org/stable/10.2307/2389422 Moskovits D.K., Bjorndal K.A. 1990. Diet and food preferences of the tortoises Geochelone carbonaria and G. denticulata in northwestern Brazil. Herpetologica46:207 218. http://www.jstor. org/stable/10.2307/3892906 Navarro G. 2004. Mapa de Vegetación del Parque Nacional y ANMI KAA-IYA del Gran Chaco. Wildlife Conservation Society, Santa Cruz. Peres C.A. 2000. Evaluating the impact and sustainability of subsistence hunting at multiple Amazonian forest sites. Pp. 31 56, in Robinson J.G., Bennett E.L. (Eds.). Hunting for sustainability in tropical forests. Columbia University Press, New York. Pezzuti J.C.B., Lima J.P., Silva D.F., Begossi A. 2010. Uses and taboos of turtles and tortoises along Rio Negro, Amazon Basin. Journal of Ethnobiology 30:153 168. doi: 10.2993/0278-0771-30.1.153 Rhodin A.G.J., Walde A.D., Horne B.D., van Dijk P.P., Blanck T., Hudson R. 2011. Editorial introduction and executive summary. Pp. 3 16, in Rhodin A.G.J., Walde A.D., Horne B.D., van Dijk P.P., Blanck T., Hudson R. (Eds.), Turtle Conservation Coalition. Turtles in trouble: The world s 25+ most endangered tortoises and freshwater turtles 2011. IUCN/SSC Tortoise and Freshwater Turtle Specialist Group, Lunenburg. Rose F.L. 1983. Aspects of the thermal biology of the Bolson tortoise, Gopherus flavomarginatus. Occasional Papers of the Museum of Texas Tech University 89:1 8. Soria M.F., Noss A. 2000. Herpetofauna de Cerro Cortado con referencias específicas a Tupinambis spp. y Chelonoidis spp. Pp. 361 365, in Cabrera E., Mercolli C., Resquin R. (Eds.), Manejo de fauna silvestre en Amazonía y Latinoamérica. CITES Paraguay, Fundación Moises Bertoni, University of Florida, Asunción. Stickel L.F. 1950. Populations and home range relationships of the box turtle, Terrapene c. carolina (Linnaeus). Ecological Monographs 20:351 378. http://www.jstor.org/stable/10.2307/1943570 Strong J.N., Fragoso J.M.V. 2006. Seed dispersal by Geochelone carbonaria and Geochelone denticulata in northwestern Brazil. Biotropica 38:683 686. doi: 10.1111/j.1744-7429.2006.00185.x Taber A., Navarro G., Arribas M.A. 1997. A new park in the Bolivian Gran Chaco an advance in tropical dry forest conservation and community-based management. Oryx 31:189 198. doi: 10.1046/ j.1365-3008.1997.d01-11.x Wang E., Donatti C.I., Ferreira V.L., Raizer J., Himmelstein J. 2011. Food habits and notes on the biology of Chelonoidis carbonaria (Spix 1824) (Testudinidae, Chelonia) in the southern Pantanal, Brazil. South American Journal of Herpetology 6:11 19. doi: 10.2994/057.006.0102 28