GROWTH AND ENERGETICS OF ARCTIC TERN CHICKS (STERNA PARADISAEA ) MARCEL KLAASSEN?2 CLAUS BECH,1 DIRKJAN MASMAN, 2'3AND GURI SLAGSVOLD • XDepartmentof Zoology,University of Trondheim,N-7055 Dragvoll,Norway, 2ZoologicalLaboratory, University of Groningen,P.O. Box14, 9750AA Haren,The Netherlands, and aCenterfor IsotopeResearch (C.I.O.), Universityof Groningen, Westersingel34, 9718 CM Groningen,The Netherlands ABSTRACT.--Westudied energy requirementsof Arctic Tern chicks(Sterna paradisaea) in Ny ,idesund,Svalbard (79øN, 12øW) with special emphasis on thermoregulatory andactivity costs.We useddoubly labeled water to estimateenergy expenditurein the field (Ea•w)and madelaboratory measurements of thedifferent components of totalenergy requirement Comparisonof DLW-estimateswith oxygen consumptionmeasurements showed that underestimatedtotal energy expenditure4.5-16.0% depending on the duration of the ex- periment. This was probablydue to incorporationof isotopesin newly synthesizedtissue. Field estimatesof total energy expenditurefrom Ea•• were correctedaccordingly. A tern chick model was usedto measureoperative temperature (T,). At middayT, reachedvalues up to 20øCabove ambient temperature (Ta) which ranged from 3.4-9.0øC. Based on the field estimates of Ea•wand Teand the laboratorymeasurements of basalmetabolic rate and thermalconduc- tance,we concludethat there wasa considerableenergy saving (456 kJ = 26%of •eq) during the first 10-11 daysof life, due to parentalbrooding. After this period,when the parents stopbrooding, the energyrequired for thermoregulationaccounted for only 16%of F_•q.From 10 days after hatching onwards,the energy needed for activity increasedconsiderably, up to 50%of Er,qjust before fledging (Day 20). Comparisonof the energybudgets of ArcticTern chickswith the more southerlyoccurring closely related Common Tern (S. hirundo;Ricklefs and White 1981)revealed only a slightly higher energyexpenditure in the ArcticTern chicks. Received5 April 1988,accepted 23 November1988. ENERGYexpenditure of chicksliving in arctic ing, with specialemphasis on thermoregulatory and antarctic environments has been consid- costs.We measuredfield growth, operativetem- ered to be dominatedby the costsfor thermo- perature, and total energy expenditure using regulation.Many physiologistshave focused on doublylabeled water. In additionwe analyzed the ability of chicksto copewith low environ- carcassesand measuredoxygen consumption in mental temperatures(e.g. Maher 1964, Norton relation to ambient temperature in chicks of 1973, Aulie and Steen 1976, Boggset al. 1977, different ages.From this we estimatedthe costs Pedersen and Steen 1979, Bech et al. 1984, Jar- of basalmetabolism, growth, thermoregulation, gensen and Blix 1985, Taylor 1985, Boersma activity, and total energy requirements,which 1986). However, the abiotic environment of were comparedwith the Common Tern (S. hi- chicksat high latitudesand the contribution of rundo)and the Sooty Tern (S. fuscata;Ricklefs thermoregulatorycosts to their total energy ex- and White 1981). penditure have not been quantified precisely. In addition to estimatingthermoregulatory costs in relation to other energy requiring processes, METHODS interspecific comparison of closely related speciesfrom different latitudes should provide Westudied Arctic Tern chicks in Ny,•lesund, Sval- bard (79øN,12øW) from 12 July to 6 August 1986.The a better understanding of the influence of the breedingbiology of theArctic Terns in Ny ,•lesund polar environment on chick energy require- has been describedby Bengtson(1971) and Lemme- ments. tyinen (1972). We studied energy requirementsof Arctic The colonywas visited regularly and individually Tern (Sterna paradisaea)chicks in the Arctic marked chicksof known age were weighed with a throughout the period from hatching to fledg- Pesolaspring balance.Wing length was measured 240 The Auk 106:240-248. April 1989 April 1989] ChickGrowth and Energetics 241 with a ruler to the nearest ram, and tarsus and head Eth= h(T• - Te) - BMR kJ.g-•.day-• (3) length with a flexible ruler to 0.1 min. We determinedbody compositionin I1 chicksthat rangedin agefrom 0 to 20 days,and in 2 adult birds. if h(Tb- Te) > BMR, where Tb is body temperature If the ageof the chickswas not knownfrom hatching (39'C; Klaassenet al. 1989). If h(Tb - Te) _<BMR, records,body massand total headlength were used was assumedto be 0 kJ.g-•.day-L to estimate the age from the field measurementsob- In the field CO2 production was measuredwith tained in the colony. After collection,the carcasses doubly labeled water (DLW; Lifson and McClintock were weighedimmediately and deep-frozen.The car- 1966, Nagy 1980). We injected 10 chicksthat varied casseswere analyzedfor water, lipid and nonlipid in age between0 and 19 dayswith 0.14 to 0.19 ml of dry matter contentat the Departmentof Zoology of a mixture containing44% of 99.84 atoms%D20 and the University of Trondheim.We dried the carcasses 56%of 90.23atoms% H2•80. The amount dependedon at 70øCto constantweight and lipids were extracted the size of the bird. A 15-gl blood samplewas taken in petroleumether. The energydensity of individual before injection, 3 h after injection (when equilibra- birdswas calculatedusing energyequivalents of 38 tion was assumedto be complete) and every subse- kJ/g lipid and 20 kJ/g nonlipid dry matter (Ricklefs quent 12 h during the following 36 h. In two cases 1974). the injectedchicks were not recapturedafter 3 h, but Weatherdata were obtainedfrom the meteorolog- were sampled12 h after injection.Blood samples were icalstation in Ny ,idesund.To estimate the combined taken from a neck vein in chicksup to 4-days-old, effect of ambienttemperature, wind exposure,and and from a wing vein in older chicks.Samples were solarradiation, we measuredthe operativetempera- sealedin a glassmicrocapillary. It took ca. 20 rain to ture (Te)which is the temperaturea chickwould attain catch,weigh, measure, bleed, and releasechicks. Blood if it lackedmetabolic heat productionand water loss sampleswere stored at 5øCuntil analysesby Isotope (Bakken1976). We measuredoperative temperature RatioMass Spectrometry (Masman and Klaassen1987) as the core temperature(copper-constantan thermo- at the Centerof IsotopeResearch (C.I.O.) in Groning- couple)of a tin castcovered with the pelt of a I-day- ell. oldArctic Tern chick.This mannequin was positioned Validation of the DLW method was obtained on in the typicalmicrohabitat for a restingchick, which three chicksof different ages,which we transported wasa gravelfield with patchilydistributed vegetation from Svalbard to Trondheim. We followed the same not exceeding10 cm in height. Operativetemperature procedureas in the field, but between the blood sam- was recordedhourly with a Leeds and Northrup plingswe measuredoxygen consumption in an open Speedomaxrecorder during 423 out of 576 hours of flow system.Air was sucked through a 9 or 17 1 investigation,of which 384hours formed 16 complete metabolicchamber with a flow rate of approximately days.The operativetemperature measurements were I l/min. The air was dried over silica-gel and oxygen used to extrapolatefrom metabolismchamber con- concentration was determined with a Servomex 1100A ditionsto the naturalhabitat for all age classes,be- oxygenanalyzer. Ambient temperaturein the cham- causesize apparently has a negligible effect on Te ber varied between 3 and 300C. The chicks were fed (Chappellet al. 1984,Walsberg and Weathers1986), ad ht•iturnwith fresh Torsk (Brosrnebrosrne). Oxygen andthe colorof the chickschanges only little during consumptionwas calculatedaccording to Hill (1972) developmentuntil fledging. and convertedto CO2production with a RQ of 0.73 In the laboratorywe measuredoxygen consump- as the diet consistedmainly of protein and fat. tion of postabsorptivechicks (n = 48) at temperatures We assumedlinear masschanges in the chicks,and ranging from 0 to 37øC(for methods, see Klaassenet calculatedCO2 production from •sO and D enrich- al. 1989).Oxygen consumption was converted to en- mentsin the bloodsamples with eq. 21 of Lifsonand ergy expenditure using 20 kJ/1 O2 assumingfat me- McClintock (1966), which was adaptedfor physical tabolism.Basal metabolic rate (BMR) and thermal con- fractionation effects(Lifson and Lee 1961, Lifson and ductance (h) of the chicks are described as functions McClintock 1966). Body water was calculatedon the of body mass(M, g): basis of the carcassanalyses results. To obtain two independent estimatesof CO2 production over ca. 24 h per chick,we usedeach blood sample analysis only BMR = 0.60 + 0.029M once.We calculatedCO• productionfor the interval - 0.00023M2 kJ.g •.day-• (1) between the first and third and between the second h = 0.58M 0.486kj.g-•.day •.oC-• (2) and fourth sampleafter injection.In 4 casesthe fourth blood sample was too low in •sO to allow accurate calculationof the CO2 production.This gave 16 in- dependent CO2 production estimatesin 10 chicksof (Klaassenet al. 1988). We calculateddaily thermo- differentages. Carbon dioxide production was con- regulatorycosts (Eth) in the field from operativetem- verted to total energy expenditure (Ea•w)using an perature and body massusing Eqs. I and 2: equivalent of 25 kJ/l COy 242 KLAASSENET AL. [Auk, Vol. 106 5331321413795543454654431231 5 120- 2(•13622151063223 2 3 2142113 1214 • 90- 14 30- 0 54352716127105 4 2 5 2 6 4 5 6 3 3 31 1 3 1 111 1 2 32122 33 4 70- 60- 50- •20 40- 30- 1ø)30 • 1•) 16 20 ADULT age, days age, days Fig. 1. Developmentof bodymass (A), tarsuslength (B), head length (C),