A Novel Incubator To Simulate The Natural Thermal .

ARTICLE IN PRESSJournal of Thermal Biology 35 (2010) 138–142Contents lists available at ScienceDirectJournal of Thermal Biologyjournal homepage: www.elsevier.com/locate/jtherbioA novel incubator to simulate the natural thermal environment ofsea turtle eggsJessica López-Correa a, Miguel Ángel Porta-Gándara b, Joaquı́n Gutiérrez b, Victor M. Gómez–Muñoz a,nabCentro Interdisciplinario de Ciencias Marinas, I.P.N., PO Box 492, La Paz, B.C.S., MéxicoEngineering Group, Centro de Investigaciones Biológicas del Noroeste, PO Box 128, La Paz, B.C.S., Méxicoa r t i c l e in f oa b s t r a c tArticle history:Received 10 November 2009Accepted 21 January 2010A novel sea turtle egg incubator was developed in which the heating element is placed above the clutch,which more closely simulates solar heating in nature. An electronic thermometer in conjunction with athermostat located in sand beneath a heater plate was used to obtain the desired temperature in theplaced eggs, as compared to previous methods of controlling global temperature within the interior of achamber. To test the new incubator, Lepidochelys olivacea eggs were incubated under different thermalconditions in order to identify the temperature-dependent sex determination (TSD) period moreprecisely. Four incubation experiments were designed to test the performance of the incubator wherethe temperature was lowered from 32 to 28 1C during 60 h and then reestablished at 32 1C untilhatching occurred. A significant mean hatching success rate of 89.6% was obtained for all theexperiments. The main result from these preliminary findings was that the sex determination period toproduce males was reduced from 15 (days 15–30) to eight days (days 19–27). Overall, the incubatorprovides precise control and simulates a natural thermal environment that may improve control of TSDin sea turtles.& 2010 Elsevier Ltd. All rights reserved.Keywords:Turtle eggs incubatorSex determinationLepidochelys olivacea1. IntroductionThe temperature of the sand during incubation of sea turtleeggs varies daily and seasonally and determines hatchling sex andinfluences embryonic survival and duration of incubation.Temperature-dependent sex determination (TSD) in sea turtleshas been documented for all extant species (Yntema andMrosovsky, 1980; Miller and Limpus, 1981; Rimblot et al., 1985;Morreale et al., 1982; Shaver et al., 1988; Dalrymple et al., 1985).For Lepidochelys olivacea, temperatures below 28 1C result inmales and above 32 1C in females, while temperatures between28 and 32 1C result in hatchlings of both sexes (McCoy et al.,1983). Due to physiological mechanisms of TSD in the sea turtleclutches, the temperatures should be measured as a routine partof the research work during the nesting period (Miller, 1999).Temperature data are extremely important to understand thenatural incubation processes. This is especially true in conservation programs, since nests are typically moved to protectedlocations where different thermal conditions prevail. In somecases hatching success in hatcheries is lower than in natural nestseven when hatcheries are constructed and supervised byconscientious staff. Also, hatchling sex ratios are often skewed,nCorresponding author. Tel.: 52 612 1225344, fax: 52 612 1225322.E-mail address: vgomez@ipn.mx (V.M. Gómez–Muñoz).0306-4565/ - see front matter & 2010 Elsevier Ltd. All rights ng on conditions in the hatchery; for instance, ValadezGonzález et al. (2000) sexed 320 hatchlings of the marine turtleLepidochelys olivacea and found a 7:3 sex ratio (230 females and90 males) at the nesting beach La Gloria, Jalisco, Mexico. Inaddition, Godfrey et al. (1999) estimated the overall seasonal sexratios of hatchling hawksbill turtles produced in Bahia, Brazil,during 6 nesting seasons, based on incubation durations. Theyfound a sex ratio greater than 90% female, which is significantlymore female-biased than estimated sex ratios of hatchlingloggerhead turtles from Bahia and Florida, USA. In contrast, ithas been reported that there is a danger of masculinizing turtleembryos by incubating eggs in artificial conditions, and attentionhas been drawn to the need for monitoring temperature inhatcheries to ensure that both sexes are produced (Mrosovsky andYntema, 1980).These problems have indicated a need to conduct research onthe effects of varying incubation temperatures with differenttypes of artificial incubators, both rustic and commercial. Forexample, Wallace et al. (2006) studied relationships between eggsize, egg components, and hatchling size in leatherback turtles(Dermochelys coriacea). To assess changes in egg mass duringincubation, they incubated 28 leatherback eggs from a singleclutch in four styrofoam Hova-Bators incubators; their resultsindicate that albumen might play a particularly significant role inleatherback embryonic development. Other incubators are commonly made of materials such as aluminum or plastic, with the

ARTICLE IN PRESSJ. López-Correa et al. / Journal of Thermal Biology 35 (2010) 138–142eggs placed in racks or buried in sterile vermiculite. In these typesof incubators a constant temperature is set for the whole chamberbut they differ from the natural environment where the sunlighthits the sand surface and produces nonhomogeneous nesttemperatures that vary as a function of depth (Drake and Spotila,2002; Hewavisenthi and Parmenter, 2002).Incubator designs in the past were markedly different. Forexample, Lang et al. (1989) built an incubator for Crocodryluspalustris using commonly available components. The basic unitconsisted of a box of molded plastic foam insulation with anaquarium heater housed in an air conditioned insulated room, andused passive techniques to stabilize daily thermal fluctuations.Another incubator for Pacific leatherback turtle Dermochelyscoriacea, was constructed with a 5.0 cm thick styrofoam box withan aquarium heater, controlled by a thermostat made with amercury switch contact (Binckley et al., 1998).The aim of this project was to design, construct and evaluate aprogrammable electronic incubator that would control theincubation temperature more precisely and duplicate the naturalthermal environment encountered by sea turtle eggs. Theperformance of the incubator was tested using olive ridley(Lepidochelys olivacea) eggs, under different temperature scenarios.2. Electronic incubatorThe principal design feature of this programmable electronicincubator for sea turtle eggs was that measurement and controlwere decoupled, unlike previous designs where control andmeasurement occur within the same chamber. In this design theautomated temperature control was decoupled by burying anupper heater plate element in the sand above the temperaturesensor. This arrangement emulates heat transfer to the clutchfrom the sun. The incubation chambers consisted of 20-lpolystyrene boxes with several holes to permit oxygen interchange (Fig. 1).The programmable electronic incubator system (Fig. 2) wasimplemented with an 8-Bit CMOS PIC16F84A (U1) microcontroller,with processing speed of 20 MHz, 68 Bytes of RAM, 64 of dataEEPROM, 11 input/output pins and 2 dedicated serial RS-232driver/receivers MAX233A (U5). The PIC16F84A requires 5 Vdc at20 mA. The microcontroller is programmed using a personalcomputer connected via a serial RS-232 port to set the desiredclutch temperature. The microcontroller PIC16F84A measures and139controls the heater temperature with a thermostat incorporating asolid state temperature sensor DS1822 (U4). This sensor is a digitalthermometer with 72 1C accuracy over a 10 to 85 1C range.Data is read out over a 1-Wires serial bus in 12 bit, 2 s complementformat, with a resolution of one bit which corresponds to0.0625 1C. The DS1822 is powered by a 5 Vdc supply. Anothertemperature sensor DS1822 (U3) is included for measuring thetarget temperature in the sand, particularly where the eggs will beplaced. The temperature is displayed by the microcontroller on a2 16 character LCD (LCD1). To maintain the programmedtemperature, the microcontroller cycles the 0.1 m nanocarbonheater on and off. The 10 W heater plate (127 Vac, 60 Hz) iscontrolled by an MOC3011 optoisolator (U2) that is switched offwhen the desired temperature is reached at the thermostatbeneath the heating plate.3. Temperature calibration procedureFor the apparatus set up and the experimental design twoDS122 temperature electronic sensors and a 12 channel scanningthermocouple thermometer (Cole-Parmer model 92000-00) wereemployed with J type thermocouples with resolution of 0.1 1C.The 14 sensors were calibrated by means of an insulated vacuumflask filled with corn oil. The reference thermometer was a CT40mercury thermometer, with 0.1 1C divisions and a range 1 to51 1C. All 14 temperature sensors and the reference thermometerwere immersed in oil at two temperatures (20 and 40 1C) and thedifferences were adjusted using the calibration procedure recommended for the scanner and for the DS122 using the microcontroller software.4. ApparatusThe appropriate heater plate temperature was determinedexperimentally with the incubator box filled only with sand, toachieve the desired target temperature at a depth of 0.15 mmeasured with the DS122 temperature sensor. To obtain amasculinizing incubation temperature of 28 1C at the clutch zone,33 1C was programmed at the thermostat. Similarly to obtain afeminizing temperature of 32 1C the thermostat was set to 37 1C.The thermostat and the temperature sensor values were measured and recorded and transferred via the RS232 communicationport to the computer.This procedure allowed for control of the desired clutchtemperature by the setup temperature of the heater plate, whichwas programmed in the microcontroller. With this system, avariation around 70.3 1C beneath the heater plate (Fig. 3)produced a fluctuation about 70.1 1C in the target zone, due tothe attenuation effect of the sand.5. Incubator test experimentFig. 1. Diagram of the incubation system, including incubator, microcontroller,computer and temperature scanner.In order to evaluate the performance of the incubator, fourperiods of thermal incubation were studied to assess the sensitiveperiod of TSD. This experiment allowed for the evaluation of thesystem versatility and operative characteristics. The experimentconsisted of four incubators equipped to permit changing theclutch temperature from 32 to 28 1C during days 15–30 of theincubation period. After 60 hours the temperature was set back to32 1C. These values correspond to the feminizing and masculinizing temperatures for the olive ridley sea turtle (McCoyet al.,1983).