Mate Retention in Caspian Terns

Home , Caspian tern, Fairy tern, White tern

The Condor87:74-18 0 The CooperOrnithological Society 1985

MATE RETENTION IN CASPIAN TERNS

FRANCESCA J. CUTHBERT

ABSTRACT. -Colonial seabirds that nest in stable, predictable environments tend to breed with the same mate for consecutive seasons.In some of these populations, mate retention has been shown to be correlated with previous reproductive success.Caspian Terns (Sterna cuspiu)were observed in northeastern Lake Michigan at several colony sites that vary in stability becauseof fluctuating water levels. Objectives of my study were to determine if (1) individuals tend to keep the same mate for consecutive breeding attempts, (2) mate retention is influenced by previous reproductive successor inter-year nest site stability, and (3) reproductive successis affected by mate change.Only 25% of the original pairs bred together for consecutive seasons,mate retention was independent of reproductive successthe previous year but not of inter-year nest site stability, and I found no significantadvantage in retaining the same mate for consecutiveseasons. These results indicate that factors other than previous reproductive success(e.g., habitat stability) influence mate selection and retention in this population of Caspian Terns. Caution should be exercised in assuming that the relationship between mate retention and previous reproductive successholds for other species of colonial seabirds.

In birds that exhibit bi-parental care, selection pendent of reproductive successin the pre- of a breeding partner has great consequences vious season?(3) is mate retention indepen- for individual reproductive success.Data from dent of inter-year nest site stability? (4) is long-term studies of marked individuals in- reproductive successaffected by mate change? dicate that many colonial seabirdstend to keep and (5) if the first nesting fails and terns re- the same mate for consecutive breeding sea- nest in the same season,do they keep the same sons. Examples include Sphenisciformes mate for the second nesting attempt? (Richdale 1947, 19 5 7) Procellariiformes (Richdale 1947, Fisher 1976, Macdonald 1977, STUDY AREA AND METHODS Brooke 1978, Ollason and Dunnet 1978, Morse The study area included four islands in north- and Buchheister 1979, Boersma et al. 1980), eastern Lake Michigan where 1,100 pairs of Pelecaniformes(Nelson 1972, Fleet 1974), and Caspian Terns (30% of the Great Lakes pop- Charadriiformes (Austin 1947, Coulson and ulation) nested in 1978 and 1979. The colony White 1958, Tuck 1960, Dorward 1963, Coul- sites were: (1) the northeastern point of High son 1966, Mills 1973). While studying colony Island (45”45’N, 85”4O’W), (2) Hat Island site tenacity in Caspian Terns (Sterna cuspiu) (45”47’N, 85”18’W), (3) Shoe Island, a gravel in the Great Lakes, I discovered that mate re- bar 1 km S of Hat Island, and (4) Ile aux Galets tention was low. Only a few other investigators (45”41’N, 85”ll’W). The islandsvaried in size of colonial seabirdshave found mate changes from less than 1 ha (Shoe Island) to 3 ha (Ile to be common, and most examples are from aux Galets), 7 ha (Hat Island) and 1,430 ha pelecaniform populations (Snow 1963, Kepler (High Island). Distances between the islands 1969, Harris 1979). Previous researchon mate ranged from l-39 km and all four colonies retention in colonial seabirds has been con- were separatedby > 90 km from the only other ducted on birds that nest in stable and pre- Caspian Tern colony in Lake Michigan. Dur- dictable environments. In contrast, Caspian ing the two breedingseasons, colony sizeranged Terns from the Great Lakes population nest from 125-520 pairs (X = 278). Three other in habitat that varies in stability because of speciesof larids, Herring Gull (Lurus urgen- fluctuations in water level. I report here the tutus), Ring-billed Gull (L. dehvurensis),and results of a two-year investigation of mate re- Common Tern (Sternahirundo) nested adja- tention in Caspian Terns breeding in north- cent to the Caspian Terns (Scharf 1978). Water eastern Lake Michigan. The purpose of my level on the Great Lakes fluctuates cyclically study was to determine if habitat instability is (Cohn and Robinson 1976) and breeding Cas- an important variable affecting pairing behav- pian Terns are affecteddirectly by thesechanges ior in this population of colonial nestingbirds. (Shugart et al. 1978, Cuthbert 1981). When Five questionswere considered:(1) do Cas- water levels are above average, Shoe Island is pian Terns keep the same mate for consecutive submerged and parts of the other islands are breeding seasons?(2) is mate retention inde- vulnerable to storm damage.

1741 CASPIAN TERN MATE RETENTION 75

To determine degree of mate retention be- TABLE 1. Pair combinations in consecutive breeding tween consecutive breeding seasons,in 1978 seasons. and 1979 I attempted to follow the nest his- Pairs tories of color-marked terns. In early June Pm combinations (n = 24) Percent 1978, 24 pairs were captured with a monofila- Same mate 25 ment line snare on their nestsat Ile aux Galets New mate (previous mate present) 2 25 and High Island, and were banded with unique New mate (previous mate not seen) 9 38 color combinations of plastic leg bands and a Non-breeders U.S. Fish and Wildlife Service (USFWS) leg Both individuals 1 4 band. One individual (previous mate not seen) 1 4 Data on reproductive successof each tern Both individuals not seen 1 4 were collectedin the following manner: all nests of color-banded birds were numbered with wooden stakes and chicks were banded with USFWS bands when they were 2-3 days old. MATE RETENTION AND PREVIOUS Nests were checked periodically from early in- REPRODUCTIVE SUCCESS cubation through banding of chicks. After the The discovery that only one-fourth of the pairs chicks were banded, I watched the color- in my sample remained together for consec- marked parentsand their offspringfrom blinds utive breeding seasonsraised the following located on the edge of the colonies. Most of question:is mate retention independent of pre- my efforts were concentrated at Ile aux Galets vious success?Data were analyzed in two ways. and High Island, but I visited the other islands First, I looked at pairs regardlessof whether every 8-10 days throughout the seasonto lo- or not the original mate was present in the cate and monitor birds that were no longer study area. Of 2 1 pairs where at least one mem- nestingat their 1978 colony. Adults were con- ber was found breeding in 1978, 17 had raised sidered to have failed to reproduce only if the one or two chicks to fledging in 1979. Five entire nest contents failed to hatch or were (29%) of these successfulpairs bred together destroyed. Terns that raised one or two chicks again in 1979, but 12 pairs (7 1%) were not re- to fledging(40 + days) were recorded as having establishedthe following year. Four pairs were a successfulreproductive season.All data were unsuccessfulin 1978. One of these remained analyzedusing the Fisher Exact Test (Zar 1974). together in 1979; three individuals from the other pairs were seen breeding with different RESULTS mates in 1979. When these data were tested, I found that the null hypothesis that mate re- MATE RETENTION IN CONSECUTIVE BREEDING SEASONS tention is independent of previous reproductive successheld true (P > 0.50). The identity of mates of color-marked terns For the second analysis, I considered only was recorded in 1978 and 1979 to determine the pair combinations where the original mate if Caspian Terns keep the same mate for con- was known to be present in the study area. secutive breeding seasons.Only 25% (6/24) of Birds that changedmates becausethe original the 1978 pairs were present in 1979 (Table 1). mate died or disappearedwere excluded from Of the remaining 18 pairs, 17 were no longer this analysis.The sample size was 12 pairs, of together; in one case, mate retention was un- which 10 were successfulin 1978. Of the suc- determined becauseneither member of the pair cessfulpairs, five were noted breeding together was seen at any of the active colony sites. Of again’in 1979. The other five pairs, however, the pairs that were no longer together in 1979, were no longer intact the following year and two (12%) involved situations where one or each member of the original pair bred with a both members were nonbreedersfor the entire new partner, even though its mate of the pre- 1979 season.In the remaining 15 pairs (88%), vious seasonwas observed in the study area. one or both individuals were seen breeding Two of the pairs were unsuccessfulin 1978. with a new mate in 1979. Birds seen breeding One of these remained together in 1979. The with new mates could be divided into two cat- other pair separatedand both individuals were egoriesbased on whether or not the mate from seen breeding with different mates in 1979. 1978 was present in 1979. Six pairs (40%) had These data also were tested and, again, the null separated, even though their 1978 mate was hypothesis held true (P > 0.50). present in the study area. An additional nine pairs (60%) were no longer together and one REPRODUCTIVE SUCCESS FOLLOWING member of each of these pairs was not seenin MATE CHANGE the study area for the entire 1979 breeding To determine how mate change affects sub- season. sequentproductivity of chicks, I examined the 76 FRANCESCA J. CUTHBERT

TABLE 2. Reproductive successand mate retention in years; five pairs remained together and four consecutivebreeding seasons. were found nestingwith new partners. The hy- pothesisthat mate retention is independent of 1919 inter-year nest site stability was tested and re- 1978 SUCCeSSfill Unsuccessful jected (P < 0.05). These results suggestthat Successfulterns habitat stability may influence mate retention Retained mate 6 4 in this population. Changed mate 4 8 Unsuccessfulterns DISCUSSION Retained mate 2 0 My resultsare not consistentwith similar stud- Changed mate 1 1 ies on mate retention in numerousother species n= 26 of seabirds. For example, my findings indi- cated that mate retention between breeding seasonsoccurs in about 25% of the pairs. This breeding successof 26 color-marked terns with is in contrast to 95% retention reported for the known mates and breeding history in 1978 and Northern Fulmar (Fulmarus glacialis; Ollason 1979 (Table 2). The sample of two terns that and Dunnet 1978); 80% for the Common Tern failed to reproduce in 1978, and of one tern (Austin 1947) White Tern (Gygis alba; Dor- that changed mates in 1979, was too small to ward 1963), Silver or Red-billed Gull (Larus analyze. Of 22 terns that were successfulin novaehollandiae; Mills 1973), and Manx 1978, 12 changedmates and 10 kept the same Shearwater (Pufinus pujinus; Brooke 1978); partners in 1979. Of those that changedmates, 75% for the Common Murre (Uris aalge; Tuck only four (33%) were successful,whereas six 1960); 7 1% for the Fork-tailed Storm-Petrel (60%) of the birds that bred with the samemate (Oceunodromafurcata; Boersma et al. 1980); raised at least one chick to fledging. Although 67 and 68%, respectively, for Leach’s Storm- these data suggestthat there may be an ad- Petrel (0. leucorhoa; Morse and Buchheister vantage to retaining the same mate for con- 1979) and the Red-tailed Tropicbird (Phae- secutive seasons,the hypothesis that repro- thon rubricauda; Fleet 1974); and 64% for the ductive success is independent of mate Black-legged Kittiwake (Rissa tridactyla; retention held true (P > 0.50). Coulson 1966). MATE RETENTION IN RE-NESTERS I also found that Caspian Terns often change As the study progressed,an additional ques- mates, even though offspring are raised to tion developed: if Caspian Terns fail to repro- fledgingin the previous breeding season.Coul- duce and then re-nest within the same season, son (1966) suggestedthat in kittiwakes, mate do they keep the same mate for the second retention is advantageous unless the pair is attempt? Although the sample was small, the incompatible (fails to hatch eggs).He proposed resultswere interesting.Six of the 24 pairs failed that changing mates is an adaptive behavior to reproduce. Of these, one pair separatedand for unsuccessfulpairs because they are more did not attempt to breed a second time. The likely to breed successfullywith a new mate. five other pairs remained together and re-nest- I found, however, that most pairs which failed ed, and four of them raised at least one off- to reproduce and then initiated a second nest spring to fledging. did not change mates; Caspian Terns that changed mates did not significantly improve MATE RETENTION AND INTER-YEAR their reproductive successin the subsequent NEST SITE STABILITY season.A number of investigatorshave found Becauselarge portions of the two primary col- that in certain populations of seabirds, indi- ony sites(High Island and Ile aux Galets) were viduals that fail to reproduce are more likely submerged or significantly altered by winter to change mates than birds that successfully storms between the 1978 and 1979 breeding produce offspring (Coulson 1966, Mills 1973, seasons,I was able to examine if mate reten- Fleet 1974, Brooke 1978, Ollason and Dunnet tion is influenced by whether or not birds can 1978, and Boersma et al. 1980). To my knowl- return to the same nest site. In 1979, I located edge,nothing has been reported about whether at least 1 breeding member from each of 2 1 of or not mate retention is influenced by previous the 1978 pairs. Twelve of the 1978 nest sites reproductive successin other speciesof terns. were no longer available becauseof high water Nevertheless, studies on the Masked Booby or changesin beach configuration. Of the 12 (Sula dactylatra; Kepler 1969) and the Flight- pairs associatedwith thesesites, 11 (92%) were less Cormorant (Nannopterum harrisi; Harris no longer together and only 1 (8%) was still 1979) also did not find a relationship between intact. In contrast, the nest sites of nine pairs mate retention and previous reproductive suc- were not noticeably different between the two cess. These studies and my data on Caspian CASPIAN TERN MATE RETENTION 77

Terns show that, as a single factor, previous terns breed in northwestern Lake Michigan reproductive successdid not have a strong in- (Gravelly Island) and in northern Lake Huron fluence on mate retention in at least three (North Channel and Georgian Bay). species.Therefore, caution shouldbe exercised in assumingthat the relationshipbetween mate ACKNOWLEDGMENTS retention and previous reproductive success holds for all speciesof colonial seabirds. I am grateful to H. B. Tordoff for his advice, support,and Becausemate retention is high in a number friendship throughoutthe courseof my study at the Uni- of speciesand appears to be related to repro- versity of Minneiota. I thank N. P. Be&tein,E. C. Bimey, ductive success,it is important to considerwhy D. J. BruEers. J. P. Ludwig. D. F. McKinnev. J. S. Ouinn. J. P. Ryd;;, 6. W. Shugaz,‘J. L. D. Smith, H. B. T&doff Caspian Terns and perhaps other seabird and M. W. Weller for reading and commenting on earlier speciesdo not fit this pattern. To date, most drafts of this paper. Field assistancewas provided by E. studieson mate retention in seabirdshave ex- Donnelly, A. Kraupa, and M. Whitmore. D. DeRuiter amined speciesthat nestin stablehabitats, often transportedme by floatplanebetween islands.I would also like to thank the Michigan Department of Natural Re- with individuals returning to occupy the same sourcesfor their cooperationand continued interest in my burrow or cliff ledge as the previous year. In research. This research was supported by a Bergstrom fact, several authors have pointed out the Award (NEBBA) and grantsfrom the Frank M. Chapman problem of measuring mate retention in pop- Memorial Fund (American Museum of Natural History), ulationsthat breed at stablesites. Nelson (1978) Dayton Natural History Fund (Bell Museum of Natural History), Department of Ecology and Behavioral Biology stated that Northern Gannets (Sula bassana) (Universitv of Minnesota), National Audubon Societv.So- nesting in the North Atlantic are faithful to ciety of Sigma Xi, and ‘Wilkie Fund for Behavioi’and both mate and site and that it is difficult to Evolution (Bell Museum of Natural History). study these two factors independently. Morse and Buchheister(1979) found that site fidelity LITERATURE CITED far exceeded mate fidelity in Leach’s Storm- ALI, S., AND S. D. RIPLEY. 1969. Handbook of the birds Petrel and suggestedthat re-mating is a func- of India and Pakistan. Vol. 3. Oxford Univ. Press. tion of burrow fidelity. Recently Morse and London. Kress(1984) demonstratedexperimentally that AUSTIN,0. L. 1947. A study of the mating of the Com- retention of the same mate by Leach’s Storm- mon Tern (Sterna hirundo). Bird-Banding 28: 1-16. Petrel during successivebreeding seasonsis BOERSMA,P. D.: N. T. WHEELF&GHT,M. K. I?ERINI, AND E. S. WHEELWRIGHT.1980. The breeding biology of largely or totally dependent on site tenacity. the Fork-tailed Storm-Petrel (Oceunodromafurcuta). Data for three speciesof terns which nest in a Auk 91~268-282. range of habitats suggestthat mate retention BROOKE.M. DEL. 1978. Some factorsaffectine the lavine is high in stablehabitats (White Tern, Dorward date, incubation and breeding successof-the hian; Shearwater(Pujinuspujinus). J. Anim. Ecol. 47:477- 1963), medium in habitats of mixed stability 495. (Common Tern, Austin 1947), and low in un- COHN,B. P., AND J. F. ROBINSON.1976. A forecastmodel predictable environments (Caspian Tern, this for Great Lakes water levels. J. Geol. 84:455-465. study). The relationship between mate reten- COULSON,J. C. 1966. Influence of the pair-bond and age tion and habitat stability may be similar to on the breedingbiolonv of the Kittiwake Gull. J. Anim. Ecol. 351269-279. -- that proposed by McNicholl (1975) between COULSON.J. C.. AND E. WHITE. 1958. The effect of ape habitat stability and nest and colony site te- on tde breeding biology of the Kittiwake (Rissa tk- nacity: the more predictable the environment, dactyla).Ibis 100:40-5 1. the higher the degree of mate retention. CUTHBERT,F. J. 1981. CaspianTern coloniesof the Great Lakes: responsesto an unpredictable environment. Additional studies on speciesthat use hab- Unpubl. Ph.D. diss. Univ. of Minnesota, Minneap- itats of different degreesof stability are needed olis. to clarify the relationship between site tenacity DORWARD,D. F. 1963. The Fairy Tern (GyRisulba) on and mate retention. I would expect mate re- Ascension Island. Ibis 103b:365-378. tention to be lower in the marsh- and riverbed- DORWARD.D. F.. AND N. P. ASHMOLE. 1963. Notes on the biblogy df the Brown Noddy (Anow stolidus)on nestingtern species(e.g., Forster’s Tern, [Ster- AscensionIsland. Ibis 103bz447-457. nu firsteri] , McNicholl 1982; Black-bellied FISHER.H. I. 1976. Some dvnamics of a breedingcolonv Tern [S. acuticuudu],Ali and Ripley 1969) than of’Laysan Albatrosses.Wilson Bull. 88: 12l-i42. - in populations that occupy stable sites such as FLEET,R. R. 1974. The Red-tailed Tropicbird on Kure Atoll. Omithol. Monogr. No. 16. rock crevices or ledges(Brown Noddy [Anous HARRIS,M. P. 1979. Population dynamics of the Flight- stolidus],Dorward and Ashmole 1963). It also lessCormorant (Nannopterumharrisz]. Ibis 121: 135- would be interesting to examine mate reten- 146. tion in the other sub-populations of Caspian KEPLER,C. B. 1969. Breedingbiology of the Blue-faced Terns breeding in the Great Lakes. The colony Booby on Green Island, Kure Atoll. Publ. Nuttall Omithol. Club No. 8. sites in northeastern Lake Michigan are on MACDONALD,M. A. 1977. Adult mortality and fidelity small islandsthat are lessprotected from storm to mate and nestsite in a group of marked Fulmars. damage than some of the sites where these Bird Studv 24~165-168. 78 FRANCESCA J. CUTHBERT

MCNICHOLL,M. K. 1975. Larid site tenacity and group RICHDALE,L. E. 1957. A population study of penguins. adherencein relation to habitat. Auk 92:98-104. Clarendon Press, Oxford. MCNICHOLL,M. K. 1982. Factors affectingreproductive SCHARF,W. C. 1978. Colonial birds nesting on man- successof Forster’s Terns at Delta Marsh, Manitoba. made and natural sites in the U.S. Great Lakes. U.S. Colonial Waterbirds 5:32-38. Army Engineers Waterways Experiment Station, MILLS. J. A. 1973. The influence of aae and pair-bond Vicksburg, MS, Rep. Wes-7R-0-78-FWSOBS-17. onthe breeding biology of the Red-billed Gull. J. SHUGART,G. W., W. C. SCHARF,AND F. J. CUTHBERT. Anim. Ecol. 42: 147-162. 1978. Statusand reproductive successof the Caspian MORSE,D. H., AND C. W. BUCHHEISTER.1979. Nesting Tern (Sterna caspiu)in the U.S. Great Lakes. Proc. patternsof Leach’s Storm-Petrelson Matinicus Rock, Colonial Waterbird Group 2: 146-156. Maine. Bird-Banding 50:145-158. SNOW,B. K. 1963. The behaviour of the Shag.Br. Birds MORSE,D. H., AND S. W. KRESS. 1984. The effect of 56:77-103. burrow loss on mate choice in the Leach’s Storm- TUCK, L. M. 1960. The Murres,their distribution,pop- Petrel. Auk 101:158-160. ulations and bioloav. Canadian Wildlife Series: 1. NELSON,J. B. 1972. Evolution of the pair bond in the Queen’s Printer, Ottawa. Sulidae. Proc. XV Int. Omithol. Congr. (1970):371- ZAR, J. H. 1974. Biostatistical analysis. Prentice-Hall, 388. Englewood Cliffs, NJ. NELSON,J. B. 1978. The Sulidae: gannetsand boobies. Oxford Univ. Press, Oxford. OLLASON,J. C., AND G. M. DUNNET. 1978. Age, expe- Bell Museum of Natural History, 10 ChurchStreet S.E., rience and other factors affectingthe breeding success Minneapolis,Minnesota 55455. Presentaddress: Depart- of Fulmars in Orkney. J. Anim. Ecol. 47:961-976. mentof Biology, 221 Life ScienceBuilding, Universityof RICHDALE,L. E. 1947. The pair bond in penguinsand Minnesota-Duluth, Duluth, Minnesota 55812. Received petrels; a banding study. Bird-Banding 18:107-l 17. 26 November 1983. Final acceptance6 August 1984.

The Condor87:78 0 The CooperOrnithological Society 1985 Southern African Ornithological Society

RECENT PUBLICATIONS ESTABLISHED1930 TELEPHONE011782-1547 P.O. Box 87234, HOUGHTON, JOHANNESBURG, Bird habitats in Britain.-R. J. Fuller. 1982. T&AD Poy- SOUTH AFRICA, 2041. ser Ltd., England. 320 p. $35.00. Source: Buteo Books, P.O. Box 481, Vermillion, SD 57069. The bird atlas proj- Objectives ect of the British Trust for Omitholoav (see Eniovinnor- To promote a greater interest in ornithology and nithology,noted at right) yielded da; that could be re- the welfare of birds in Africa. organized and used for a guide to bird habitats that have specialpotential for acquisitionfor conservationand man- To assistand advise local authorities on all matters agement purposes. The Register of Ornithological Sites relating to ornithology and in all its aspectsinclud- was examined for sites that deserved further study. De- ing the framing of protective legislation. scriptions of the avian population, and habitats for im- To publish a journal OSTRICHfour times a year. portant sitesthroughout Britain are provided. Lists of bird To publish a popular magazine BOKMAKIERIE four and plant speciesmentioned, descriptions of methodol- times a year. ogy, and sample descriptionsare containedin appendices. Bibliography. Tables. Index. Of value to the student of bird populations and their relationship to the structureof Subscriptionsfor the calendar year 1985 the habitat. -J. Tate. Membership includes both quarterly publications Institutional U.S. $30.00. Individual U.S. $22.00

Enjoyingornithology.-Ronald Hickling, ed. 1983. T&AD Publications Poyser Ltd., England. 296 p. $30.00. Source: as at left. Back numbers of OSTRICH, OSTRICHSUPPLEMENTS, The British Trust for Ornithology celebrated its fiftieth BOKMAKIERIEare available, prices on application. year (1933-l 983) by publishingthis history of birdwatch- OSTRICHIndex ing and ornithology in the British Isles. The book gives not only the history of the Trust and its predecessors,but 1951-1979 Volumes 21-50 U.S. $15. also a review of its major causesand projects. Sister or- BOKMAKIERIEIndex ganizations receive more than a friendly wave as well. 1948-1977 Volumes l-30 U.S. $10. Conservation and bird problems are dealt with in several S.A.O.S. CHECKLISTOF SOUTHERNAFRICAN BIRI~S, relatedchapters. Forty pagesof data may be of useto those Edited by P. A. Clancey U.S. $22. interested in: banding and recovery rates, most distant Proceedings of the Fourth Pan African Ornitho- recoveries, fastest movements as shown by banding lon- gevity, causesof mortality, woodland and farmland pop- logical Congress, Seychelles, 1976 U.S. $22. ulation densities,breeding statistics,winter temperatures, breedingseason rainfall, speciesubiquity, and bird weights. Tapes of South African Bird Calls are also avail- List of speciesnames mentioned. Index. - J. Tate. able.