sign in sign up
<<
Home , Penguin

376 SHORT COMMUNICATIONS

BIRKHEAD,T R., AND A. I? MILLER. 1992. Numbers stimulate late summer breeding in the Pition Jay. and size of sperm storagetubules and the duration Nature 250:80-82. of sperm storage in : a comparative study. MOORE,E R., AND P KERLINGER.1987. Stopover and Biol. J. Linn. Sot. 45:363-372. fat deposition by North American wood-warblers BLANCHER,P, A. CYR, S. DROEGE,D. HUSSELL,AND L. (Parulinae) following spring migration over the THOMAS.1993. Results of a U.S./Canada work- Gulf of Mexico. Oecologia 74:47-54. shop on monitoring of landbirds during migration MOORE,E R., AND M. V MCDONALD. 1993. On the and recommendationstowards a North American possibility that intercontinentallandbird migrants Migration Monitoring Program (MMP). Natl. copulate en route. 110:157-160. Biol. Survey, Washington,DC. PROCTOR,N. S., AND P. J. LYNCH. 1993. Manual of BRISKIE,J. V 1994. Seasonalpatterns of sperm stor- : avian structure and function. Yale age in the Yellow-headed Blackbird Xanthoceph- Univ. Press, New Haven, CT. alus xanthocephalus. Ibis 136:323-330. PYLE, P, S. N. G. HOWELL,R. F! YUNICK,AND D. E BRISKIE,J. V. 1996. Lack of sperm storageby female DESANTE. 1987. Identification guide to North migrants and the significance of copulations en American passerines.Slate Creek Press, Bolinas, route. Condor 98:414-417. CA. BROOKS,E. W. 1994. Braddock Bay Observato- QUAY, W. B. 1984. Cloaca1lavage of sperm: a tech- ;;;e;pFyg 1994 passerine banding report. Roch- nique for evaluation of reproductive activity. N. Am. Bird Bander 9:2-7. 3 . DELIUS,J. D. 1965. A population study of Skylarks QUAY,W. B. 1985a. Cloaca1sperm in spring migrants: occurrence and interpretation. Condor 87:273- Alauda arvensis. Ibis 107:466492. 280. DRENT,R. H., AND S. DAAN. 1980. The prudent par- QUAY, W. B. 1985b. Sperm release in migrating ent: energetic adjustmentsin avian breeding. AI- wood-warblers (Parulinae) nesting at higher lati- dea 68:225-252. tudes. Wilson Bull. 97:283-295. GUDMUNDSSON,G. A., A. LINDSTROM,AND T. ALER- QUAY, W. B. 1986. Timing and location of spring STAM. 1991. Optimal fat loads and long-distance soerm release in northern thrushes.Wilson Bull. by migrating Knots Calidris canutus, Sand- 98:526-534. erlings C. alba, and TurnstonesArenaria inter- QUAY, W. B. 1989. Insemination of TennesseeWar- pres. Ibis 133:140-152. blers during spring migration. Condor 91:660- HATCH,S. A. 1983. Mechanism and ecological sig- h7n nificance of sperm storagein the Northern Fulmar RICKLEFS,- - ’ R. E. 1974. Energetics of reproductionin with reference to its occurrence in other birds. birds, p. 152-292. In R. A. Paynter [ed.], Avian Auk 100:593-600. energetics.Publ. Nuttall Omithol. Sot. 15., Cam- KUENZI,A. J., E R. MOORE,AND T. R. SIMONS. 1991. bridge, MA. Stopoverof Neotropical landbird migrantson East WALSBERG,G. E. 1983. Avian ecological energetics, Ship Island following trans-gulf migration. Con- p. 161-220. In D. S. Farner,J. R. King, and K. C. dor 93:869-883. Parks [eds.], Avian biology. Vol. 7. Academic LIGON, J. D. 1974. Green cones of the pifion pine Press, New York.

The Condor 100:376-381 0 The Cooper Omlthological Society 1998

SATELLITE TRACKING OF MAGELLANIC MIGRATION’

DAVID L. STOKESAND F! DEE BOERSMA Department of Zoology, Box 351800, University of Washington, Seattle, WA 9819.5, e-mail: [email protected] LLOYD S. DAVIS Department of Zoology, University of Otago, P.O. Box 56, Dunedin,

Abstract. We tracked four male Magellanic Pen- where they had bred and molted, all four guins (Spheniscus magellanicus) from the beginningof birds traveled northeastalong the coast at speedscon- their winter migration for periods ranging from three sistent with sighting and band-return data which sug- to six weeks. Upon leaving the colony at PuntaTombo, gest a round-trip winter migration in this speciesof up to several thousandkilometers. Three of the birds fol- lowed a similar route, moved at approximately the ’ ’ Received 15 May 1997. Accepted 6 January 1998. same rate, and were still moving northeastwardat a SHORT COMMUNICATIONS 377

distance of over 800 km from the colony when signals devices (attached to males #l-3) were 0.25-watt mi- ceased. The fourth bird departed from the northeast croprocessor-controlled satellite-linked time-depth re- route at approximately 306 km into its trip and re- corder/transmitters manufactured by Wildlife Comput- mained in a large bav for at least 29 davs. when the ers Inc. (Redmond. WA) using Telonics ST-10 trans- last signal was”rece&ed. Initially, the birds traveled mitters. These devices, including waterproof epoxy rapidly and made few deep dives. Later, their rate of housing, each weighed approximately 155 g, and had travel slowed, their paths became more erratic, and a cross-sectional area of 13.5 cm2, approximately 7- they made more deep dives, all indications of probable 8% of the cross-sectional area of a male Magellanic foraging. This suggests that location of the penguin Penguin. The fourth device (attached to male #4) was migration route and rate of travel along it may be de- manufactured by Sirtrack Ltd. (Havelock North, New termined by seasonal movement of prey. As all birds Zealand) and also used a Telonics ST-10 transmitter were within 250 km of shore for the entire period in packed in an epoxy housing. It weighed 96 g and had which signals were received, our results underscore the a cross-sectional area of 6.5 cm2. We used fast-setting importance of controlling pollution associated with pe- epoxy to attach the devices to the in the center troleum transport along the coast of Argentina, cur- of the penguin’s lower back. All devices had sloped rently a major source of mortality. leading and trailing faces and were positioned poste- rior to the line of maximum girth to minimize drag Key words: Magellanic Penguin, migration, pen- (Bannasch et al. 1994). guins, satellite telemetry, Spheniscus magellanicus. We programmed the Wildlife Computer devices to transmit at 45-set intervals every other day, primarily As with other , little is known of Magellanic during the night-time hours most likely to coincide Penguins’ (Spheniscus magellanicus) activities during with satellite passes. At night, Magellanic Penguins the nonbreeding season. After chicks fledge, breeding tend to rest at the surface and hence the trans- adults undergo a synchronous molt in March and April mitter signal is less likely to be broken due to sub- and then leave the breeding colony, not to return until merging of the antenna (P D. Boersma, unpubl. data). the following September or October. Their where- The Sirtrack device was pre-set to transmit at 60.set abouts during this 5-to-6 month period have been in- intervals for 6 hr every 72 hr. In addition to location, ferred from sightings of unmarked birds as well as the Wildlife Computer devices recorded frequencies of band returns, mostly from dead birds that have washed dives in six depth classes (3-5, 6-10, 11-20, 21-40, ashore. These methods indicate that Magellanic Pen- 41-60. and 61-80 m) for 6-hr oeriods (03:00-09:OO. guins migrate northward in the winter (Daciuk 1977), 09:00~15:00, etc. local time) ea’ch day. ~ in some cases apparently more than 2,000 km (I? D. We received transmissions through the Service AR- Boersma, unpubl. data). However, sightings of un- GOS satellite system. Sixty-one percent (n = 86) of marked birds give no indication of breeding status or locations received (n = 140) were based upon at least colony of origin, and band returns yield imprecise lo- four messages per satellite pass and were accurate to cations of live birds and give little indication of pattern within 1 km (location 1 or better; ARGOS 1996). of movement or distance from shore of migrating The accuracy of the remainder of locations could not birds. Such information is important for the protection be determined with certainty (ARGOS 1996); however of this , which is currently declining (Boersma based upon locations received for birds at known po- 1997) primarily due to human activities at sea, chiefly sitions (Stokes and Boersma, in press), locations cal- commercial fishing and oil pollution (Gandini et al. culated from three messages per pass (class 0 and class 1994, Boersma and Stokes 1995). A, n = 45) are sufficiently accurate for the purposes The recent development of a practical technique for of this study (mean error 2 SD = 2.9 2 3.6 km, n = tracking penguins at sea using satellite telemetry has 12). We did not use locations based upon fewer than the potential to improve our understanding of penguin three messages (class B and Z, n = 9). For the pur- migration by providing precise day-to-day monitoring poses of calculating travel speed, we used only loca- of live individuals of known breeding site, sex, age, tions of class 1 or better. Maps of travel routes (Figs. and breeding status. Studies using this-technique (Ker- 1 and 2) were made, and distances and swimming ry et al. 1995, Davis et al. 1996) have documented speeds calculated, using Arc/Info geographic infor- long-distance post-breeding migration in Adelie Pen- mation software (ESRI 1992). Values presented are guins ( adeliae) in the and have means ? SD. described the routes followed bv individuals in the ear- ly part of the migration. Here-we report on the first RESULTS use of satellite telemetry to track the migration of a temperate penguin species, the Magellanic Penguin. Upon leaving the colony, all birds initially traveled northeast (Fig. 1). After passing Peninsula Valdts, METHODS male #l veered northwest into the Gulf of San Matias From lo-15 April 1996, we attached satellite trans- and stayed within the Gulf at least until signals StODDcd.a ueriod of 29 davs (. 2). Males #2-4 con- mitters to four male Magellanic Penguins (mean wt. of II I ~ _ birds = 4.32 kg, range = 4.13-4.53) after they had tinued in a northeastward*direction; all moving along finished molting at the large penguin colony at Punta a similar path at approximately the same rate. None of Tombo, Argentina (44”02’S, 65”ll’W). All four males the birds returned to the colony after leaving, and none were experienced breeders, having been banded as were detected on land at any time while they were breeding adults in the colony prior to the study, and tracked. all had bred in the 1995-1996 season. Three of the The date of last received signal ranged from 1 to 20 378 SHORT COMMUNICATIONS

*:**@011$-#3 N ,*rn@ ce#4

A_ 0 100 200 300_. SO 500 Km

FIGURE 1. Migration routes of four Magellanic Penguins from Punta Tombo, Argentina, April and May 1996. All were males that had bred at Punta Tombo in the 1995-1996 season. Routes of males #l-3 were each based upon an average of 43 locations (range = 30-48). Six locations were received for male #4, two each on 24 and 27 April and 1 May. Dashed line indicates the shortest swimming route between the 18 and 22 April locations for male #l. Black line parallel to coast of in inset map indicates the section of coast where most penguins banded during the breeding season at Punta Tomb0 are found during the nonbreeding season.

May (Table 1). As all birds were seen alive and with- from shore was similar for males #2-4 (mean maxi- out transmitters the following season, the devices ap- mum = 234 % 11.3 km). After passing Peninsula Val- parently fell off or failed within 3-to-6 weeks of ap- des, these three birds were between 100 and 200 km plication. At the time of the last signal from each trans- from shore for most of their routes. mitter, the birds’ swimming distance from Punta Tom- The device with the smaller cross-sectional area (on bo averaged 756 t 157 km. Last locations for males male #4) undoubtedly caused less drag than the larger #2-4 were all similar distances from the colony (Table devices. Nonetheless, the similar location and rate of 1). Minimum distance traveled by the tracked birds movement northward of males #2-4 suggest that al- ranged from 844 to 1,216 km. These are likely to be though the devices probably affected swimming effi- underestimates of actual distance traveled because ciency, they probably did not affect the pattern of birds may have departed from the shortest path be- movement. tween known locations when devices were not trans- Travel speed differed by stage of trip. Males #l-3 mitting. The location farthest from shore for all birds averaged 69.4 km day-’ in the first week of travel, and was 246 km (male #3, 25 April). Maximum distance 29.6 km day-i thereafter. Initial travel speed of males SHORT COMMUNICATIONS 379

-x- #1 22April - 20 May

FIGURE 2. Route of male #l, within the Gulf of San Matias, 22 April-20 May 1996. Because of the finer spatial scale, locations of class A (6 of the 39 locations in this time period; see text) were excluded in making this figure. An “A” indicates last location received (20 May).

#l-3 differed very little (Table 1). Males #2 and 3 also = 46 6-hr periods), and only one of these was deeper had very similar subsequent speeds (37.9 and 37.8 km than 5 m. day-i, respectively), whereas male #1 averaged 13.1 km day-r during the time it spent in the Gulf of San DISCUSSION Matias. Because we received no signals for male #4 Returns of bands found during the austral winter sug- before 24 April and were unable to determine its date gest that adult Magellanic Penguins breeding at Punta of departure, its initial and subsequent speeds could Tombo commonly migrate more than 2,000 miles not be compared. However, its rate of travel also northeastward along the coast of South America to the slowed over the course of its trip, from 72.9 km day-’ waters off the coast of southern Brazil (I? D. Boersma, between 24-27 April to 53.1 km day-r between 27 unpubl. data). The results of the present study are con- April and 1 May. sistent with this conclusion. The birds we tracked Dive patterns of the birds also differed markedly moved northeastward along the coast and although sig- between the first days of their trips and the remainder nals ceased when the birds were 486-873 km from the of the tracking period, with a smaller proportion of colony, the speeds at which they traveled to that point deep dives in the initial period. This difference was (> 800 km in three weeks for the three birds that con- evident for both male #l (which went to the Gulf of tinued northward) indicate that the 5-to-6 month mi- San Matias) and male #2 (which stayed off the coast), gratory period is sufficient for the round-trip migration the two birds for which we have complete dive records of up to 5,000 km suggested by band returns and sight- (Fig. 3). For both birds. the distribution of dive denths ings. differed significantly between the first week and thkre- The difference between the paths of male #1 and the after, with more dives to greater depth after the first other three birds also is consistent with winter sight- week (Fig. 4). Dive depth in the initial days of travel ings and band recoveries. Although most bands have was not limited by ocean depth, which is greater than been recovered along the coast of southern Brazil (Fig. 50 m for most of the area traversed in the initial travel 1), some have been found along the Argentine coast period. much closer to Punta Tomb0 (I? D. Boersma, unpubl. Birds generally did not make deep dives at night. data). Jehl (1974) reported seeing Magellanic Penguins Of the total of 8,510 dives we recorded during 149 6- of unknown age and breeding status in the Gulf of San hr periods, only 45 were made during periods that in- Matias during a ship-transect winter bird survey of the cluded no daylight hours (21:00-03:OO local time; n Argentine coastal shelf, and Jehl et al. (1973) sighted 380 SHORT COMMUNICATIONS

TABLE 1. Summary of movements of four male Magellanic Penguinsdeparting from Punta Tombo, Argentina in April and May 1996, as determinedby satellite telemetry. Distance from colony at last location is the minimum swimming distance; actual distance traveled was considerablygreater (see text). Initial travel speed covers the first week following departure.Subsequent travel speed covers the remainder of the trip to the date of last signal. Initial and subsequenttravel speed could not be determined for male #4 because date of departure from the colony was unknown.

Distance Greatest No. of from colony distance Travel speed (km day-‘) Last date Date of last usable at last from shore Male at colony location locations location (km) (km) Initial Subsequent #l 15 April 20 May 48 487 48 69.0 13.1 #2 14 April 4 May 30 806 219 70.2 37.9 #3 15 April 7 May 47 859 246 69.1 37.8 #4 215 April 1 May 6 873 238 -

Magellanic Penguins in the Gulf of San Jose (on the These individual differences and the variable speed south margin of the Gulf of San Matias) in the winters of movement by stage of trip suggestthat migration of 1971 and 1972, estimatingthat 400 Magellanic Pen- pattern may depend upon the rate at which prey is guins wintered there in 1972. Thus, althoughit appears encountered. Penguins may travel until they locate that most birds migrate thousandsof kilometers, per- concentrationsof anchovy (Engraulis nnchoita), hake haps in a fairly well defined route, some may migrate (Merluccius h&xi), or other prey species, and only much shorterdistances. Differences among individuals move when those concentrationsmove or are depleted. also are observed in foraging location during the The location and dive data are consistentwith this in- breeding season(Stokes and Boersma, in press). terpretation.In the initial days of their trips, although they probably had not eaten since they began molting several weeks earlier, the tracked penguinstraveled in 100 _ A. a direct route, at a fast rate, and made few deep dives.

60 A. MALE #1

50 UI n FIRST WEEK n=1,204 2 40 q SUBSEQUENT n=3,496

E 20

E MALE #l

P 0 i:

x100 8 i I I B.

260 .#. 0

t;5 6040 =I

MALE #2

DAYSAFTERDEPARTURE 3-5 6-10 11-20 21-40 >40 FIGURE 3. Percent of dives deeper than 5 m for (A) DIVE DEPTH (m) male #l (n = 2,535 dives) and (B) male #2 (n = 859 dives) between 09:OOand 15:00 local time from date FIGURE 4. Dive depths of males #l and #2 during of departure from the colony to date of last signal. 6-hr periods with some daylight (03:00-21:00 local Only dives in the midday period are included in time) in the first week of migration (l&22 April) ver- to exclude periods of darknesswhen birds do little or sus remainder of period tracked. Dive depth frequen- no foraging. Male #3’s dive record was incompletebe- cies differ significantly by stage of trip. (A) male #l cause of device malfunction and therefore was not in- (xz4 = 705.0, P < 0.001). (B) male #2 hzq = 129.6, cluded. P < 0.001). SHORT COMMUNICATIONS 381

Later, their paths became more erratic, their travel BOERSMA,I? D. 1997. The decline of Magellanic Pen- speeds slower, and their dives deeper, all indications guins. Penguin Conserv. 10:2-5. of probable foraging (Wilson 1995). Anchovy, an im- BOERSMA,F? D., AND D. L. STOKES.1995. Conserva- portant componentof the Magellanic Penguin-dietdur- tion: threats to penguin populations,p. 127-139. ing the breeding seasonat a colony approximately 100 In T D. Williams [ed.], The penguins. Oxford km south of Punta Tombo (Frere et al. 1996), are dis- Univ. Press, Oxford. tributed over the continental shelf along the South DACIUK,J. 1977. Notas faunisticasy bioecologicasde American coast from approximately23 ”s to 47’S, and Peninsula ValdCs y IV. Observaciones spawn in high concentrationsin the summer between sobre areas de nidificacion de la avifauna de1 li- 37% and 41% (Lima and Caste1101995). Spawning toral maritimo Patagonico (Provincias de Chubut activity moves northward in fall, and in winter occurs y Santa Cruz, Rep. Argentina). El Homer0 11: off the coast of southern Brazil, the same section of 361-376. coast where most banded penguinsfrom Punta Tombo DAVIS, L. S., P. D. BOERSMA,AND G. S. COURT.1996. are found between breeding seasons.Thus, the route Satellite telemetry of the winter migration of Ade- of Magellanic Penguin migration may follow seasonal lie Penguins (Pygoscelis a&he). Polar Biol. 16: movements of anchovy. 221-225. The results of this study have implications for con- ENVIRONMENTALSYSTEMS RESEARCH INSTITUTE (ESRI). servation. Migrating Magellanic Penguins appear to 1992. ARC/INFO. ESRI, Redlands, CA. remain relatively close to shore (within 300 km), and FRERE,E., P GANDINI,AND V. LICHTSHEIN.1996. Var- consequentlyare in the coastal shipping lanes present- iacion latitudinal en la dieta de1 pinguino de Ma- ly used by tankerstransporting petroleum from oil ter- gallanes(Spheniscus magellunicus) en la costaPa- minals south of Punta Tombo to refineries in Buenos tagonica, Argentina. Ornithol. Neotropica 7:35- Aires. Oil spills and oily ballast discharge associated 41. with this oil transnortare estimatedto kill 4 1,000 Maa- GANDINI,I?, I? D. BOERSMA,E. FRERE,M. GANDINI,T ellanic Penguinsannually along the coast of southe; HOLIK, AND V. LICHTSCHEIN.1994. Magellanic Argentina (Gandini et al. 1994). Nearly all the lethally Penguins are affected by chronic petroleum pol- oiled penguinsobserved each at Punta Tombo and lution along the coast of Chubut, Argentina. Auk other breeding colonies are seen at the beginning of 111:20-27. the breeding season, when birds are returning from JEHL,J. R., JR. 1974. Distribution and ecology of ma- winter migration. Taken together, these observations rine birds over the continental shelf of Argentina suggestthat effective conservationof this speciesde- in winter. San Diego Sot. Nat. Hist. Trans. 17: pends upon large-scaleapproaches that addressits ex- 217-234. tensive marine requirements,and in particular, JEHL,J. R., JR., M. A. E. RUMBOLL,AND J. P WINTER. the problem of chronic oil pollution along its migration 1973. Winter bird populationsof Golfo San Jose, route in the coastal waters of Argentina. Argentina. Bull. Br. Ornithologists’ Club 93:56- 63. This researchwas funded by the Exxon Corporation, KERRY,K. R., J. R. CLARKE,AND G. D. ELSE. 1995. Esso Argentina, and the Wildlife ConservationSociety The foraging range of Adelie Penguins at Be- (PDB and DLS), and a Fulbright award and an Otago chervaise Island, Mac.RobertsonLand, Research Grant (LSD). The work was made possible as determined by satellite telemetry, p. 216243. by a joint agreement between the Wildlife Conserva- In I? Dann, I. Norman, and l? Reilly [eds.], The tion Society and The Province of Chubut, Argentina. penguins:ecology and management.Surrey Beat- We thank Graham Harris and Albert0 Neira for logis- ty, Sydney. tical assistance.The comments of two anonymousre- LIMA, I. D., AND J. E?CASTELLO. 1995. Distribution viewers improved the manuscript.Roger and Sue Hill and abundance of Southwest Atlantic anchovy at Wildlife Computersprovided advice on device set- spawners (Engraulis anchoita) in relation to tings and deployment. oceanographicprocesses in the southernBrazilian shelf. l&h. Oceanogr.4: 1-16. LITERATURE CITED STOKES.D. L.. AND F? D. BOERSMA.In mess. Where ARGOS. 1996. User’s manual. CLS Service Argos, breeding Magellanic Penguinsforage: satellite te- Toulouse,. lemetry results and their implications for penguin BANNASCH,R., R. I? WILSON, AND B. CULIK. 1994. conservation.Mar. Omithol. Hydrodynamic aspectsof design and attachment WILSON,R. I? 1995. Foraging ecology, p. 81-106. In of a back-mounted device in penguins. J. Exp. T. D. Williams [ed.], The penguins. Oxford Univ. Biol. 194:83-96. Press, Oxford.