Long-Distance Movements by American Avocets and Black-Necked Stilts

Home , American avocet, Avocet, Recurvirostridae

J. Field Ornithol., 67(2):307-320

LONG-DISTANCE MOVEMENTS BY AMERICAN AVOCETS AND BLACK-NECKED STILTS

JULIEA. ROBINSON1 AND LEWISW. ORING Ecology,Evolution & ConservationBiology Program /186 and Departmentof Environmentaland ResourceSciences Universityof Nevada, Reno 1000 ValleyRd. Reno, NV 89512 USA

Abstract.wColor-banded American Avocets(Recurvirostra americana) and Black-neckedStilts (Himantopusmexicanus) were resighted>50 km from their banding locationsin the Great Basin.Records from 90 migrationand winterresightings of birdsbanded in the Honey Lake Valley (northeasternCalifornia) and 3 birds banded at Great Salt Lake (Utah) provided unique data on long-distancemovements by individualsof thesespecies. Avocets and stilts left Honey Lake breeding areasand went to stopoverareas at Mono Lake and in the Tulare Basin as early asJuly. Some individualsspent 48 daysor more in the Tulare Basin.Both avocetsand stiltswere observedin siblinggroups during migration.Although one pair of bandedstilts was observed during migration,differences in sexratio at a migrationstopover sitesuggested that male and femalestilts might havedifferent wintering ranges or migration behavior.Avocets from Honey Lake were found wintering along the California coastfrom Arcata southwardand along the westcoast of Sinaloa,Mexico. First-yearavocets were more often seen in coastal areas of California (relative to inland areas) than were adults. Stilts were found wintering in central California. Avocetsand stiltsbanded in Utah were seen in the Tulare Basin (one stilt) and at winteringgrounds in Mexico (one avocetand one stilt), indicatingthe potential for populationmixing. We proposethat avocetsmigrate in short hops,retain partial family group associationsduring migration,and may have age-specific differencesin migratorybehavior or winteringrange. We proposethat stiltsretain partial family group associationsduring migration, maintain pair bonds beyond the breeding attempt, and have sexualdifferences in migratory behavior or wintering range. These results have important ramificationsfor understandingimpacts of drainwaterevaporation ponds in the Tulare Basin on nonbreeding avocetsand stilts.

MOVIMIENTOS DE LARGA DISTANCIA EN RECURVIROSTRA AMERICANA Y EN HIMANTOPUS MEXICANUS Sinopsis.--Individuosde Recurvirostraamericana y de Himantopusmexicanus anillados con colores fueron redetectados visuaimente sobre 50 km de los sitios donde se anillaron en la Gran Cuenca. Registrosde 90 casosde redetecci6n visual de avesmarcadas con bandas de coloresen el Valle del Lago Honey (California nororiental) y tres avesanilladas en Gran Lago Salado (Utah) son datos 6nicos para estudiar movimientosa larga distanciade estas especiesdurante la migraci6ny el invierno.Las aves dejaron las 5teas reproductivas del Lago Honey y fueron a •eas de descansoen el Lago Mono yen la Cuencade Tulare t•n temprano como en julio. Algunasaves estuvieron 48 diaso mssen la Cuencadel Tulare. Arebasespecies se observaronen gruposde hermanosdurante la migraci6n.Aunque durante la migraci6n se observ6un par de Himantopusmexicanus con anillasde colores,las diferenciasen raz6n de sexo detectadasen un lugar de parada migratoria sugierenque existendiferencias en •reas de invernaci6no en comportamientomigratorio entre los sexosde esta especie.In- dividuosde Recurvirostraamericana del Iago Honey se hallaron invernandoa lo largo de la costade California desdeArcata hacia el sur y a lo largo de la costaoeste de Sinaloa,Mexico. Las avesde un afio de Recurvirostraamericana se vieton m•s comfmmenteque los adultos en las costas de California (en relaci6n a zonas mss internas). Se hailaron individuos de Recurvirostra americana invernando en el centro de California. Individuos de Recurvirostra

Currentaddress: Department of Biology,University of Houston,Houston, Texas 77204 USA.

307 308] j. A. Robinsonand L. W. O•ing J. Field Ornithol. Spring 1996 americanay de Himantopus mexicanusanillados en Utah se observaron en la Cuenca del Tulare (un ave) yen terrenos de invernar en Mexico (una de cada especie),indicando el potencialpara la mezclainterpoblacional. Proponemos que Recurvirostraamericana migra en pequefiossaltos, retienen asociacionesparciales de gruposfamiliares durante la migra- ci6n, y pueden tener diferenciasentre edadesespecificas con respectoal comportamiento migratorio o en su distribuci6ninvernal. ProponemosademJs que Himantopusmexicanus retienen parcialmenteasociaciones de gruposfamiliares durante la migraci6n,mantienen unionesde parejasm•s alia del esfuerzoen aparearse,y tienen diferenciassexuales en comportamientomigratorio o en la distribuci6ninvernal. Estos resultados tienen ramificaciones importantespara entenderel impactode las charcasde evaporaci6nde aguasde escorrentia en la Cuencadel Tulare en avesno anidantesde ambasespecies.

Wetlandsin the westernUnited Stateshave declined dramaticallysince 1780 (Dahl 1990). For example,inland wetlandsin the CaliforniaCentral Valleyhave declined 91%; wetlandsin Colorado,Idaho, and Nevadahave declined more than 50%; and wetlandsof Utah, Oregon, and Arizona havedeclined by 35% (Dahl 1990). AmericanAvocets (Recurvirostra americana) and Black-neckedStilts (Himantopus mexicanus)are two of the most conspicuousbreeding birds in wetlandsof the Great Basin. Because avocetsand stilts rely on these inland habitatsfor breeding, migration, and wintering, their populationshave declined in responseto this habitat loss(Page and Gill 1994). Dramaticannual changesin wetlandavailability make information on large-scalemovements of shorebirdscritical to con- servingremaining wetlands (Alberico 1993, Skagenand Knopf 1993, Rob- inson and Warnock 1996). In addition to wetland losses,many remaining wetlandsin the western United Stateshave been contaminatedas a result of irrigation (U.S. Fish and Wildlife Service 1992). The discoveryof selenium-inducedteratoge- nesisin avocet and stilt embryosin central California (Hoffman et al. 1988; Ohlendorf et al. 1986a, 1989), has made an understandingof re- curvirostridlife historieseven more important. The impactsof complete reproductive failure at KestersonReservoir (Ohlendorf et al. 1990, Wil- liams et al. 1989), and adverseeffects at other California breeding sites (Grasslands,Ohlendorf et al. 1987;Tulare Basin,Skorupa and Ohlendorf 1991) can be better interpreted if the roles of contaminated wetlands in the species'annual cyclesare known. In particular,it is important to know the role that these wetlandsplay in the migration and wintering of re- curvirostridsbreeding at noncontaminated sitesin western North Amer- ica. Although recurvirostridbreeding and wintering behaviorhas been well- studied (e.g., Boettcher et al. 1994, Evansand Harris 1994, Gibson 1971, Hamilton 1975), and individualswere markedin severalpublished studies (Gibson1971, James 1995, Sordahl1984), relativelylittle informationex- istson long-distancemovements of knownindividuals. As part of breeding population studies,we have conductedlarge-scale banding of avocetsand stiltsat noncontaminatedsites in the Great Basin.The magnitude of this effort, coupledwith the assistanceof numerouscolleagues and volunteers, has provided considerableinformation on long-distancemovements of individual American Avocetsand Black-neckedStilts. Here, we present Vol.67, •o. 2 Avocetand StiltMovements [309

1500- 1371 American .•,, 1992 Avocets 1 1993 1 1994 '• 1000-

632 500 296 295

Females Males Chicks Chicks

Honey Lake Valley Great Salt Lake

622

4001 355

200øø1 100 73 81

o Females Males Chicks Chicks

Honey Lake Valley Great Salt Lake hou• 1. Numbers of American Avocetsand Black-neckedSdlts banded at Honey Lake Valley, California, and Great Salt Lake, Utah, 1992-1994. resightingsof migratory or wintering individualsand evaluatemigration routes and wintering areasfor each species.

STUDY AREAS AND METHODS Totals of 591 adult avocets,2003 juvenile avocets,154 adult stilts,and 977 juvenile stiltswere banded at two sites:the Honey Lake Valley,Cali- fornia and Great Salt Lake, Utah (Fig. 1). Birdswere bandedwith unique combinationsof UV-resistantcolored bands (A. C. Hughes, Ltd., Middle- 310] J. A. Robinsonand L. W. Oring j. FieldOrnithol. Spring 1996 sex,U.K.). Avocetswere sexedby bill curvature(Palmer 1967), and stilts were sexedby plumage (Palmer 1967:151,Prater et al. 1977). Similar proportions of adult femalesand maleswere banded (386:391 for avocets, 73:84 for stilts). In California, most banding was conducted in the Honey Lake Valley from 1992-1994 at three managedwetlands: the Jay Dow, Sr. Wetlands (40ø10'N,120ø13'W), a researchfacility of the Universityof Nevada,Reno, and the two units of the Honey Lake StateWildlife Area, the Dakin Unit (40ø18'N,120ø23'W), and the Fleming Unit (40ø19'N,120ø17'W). Limited banding wasalso conducted at other sitesin the valleyincluding the Wild Goose Hunting Club. (40ø07'N, 120ø13'W), Amedee Hot Springs (40ø18'N,120ø12'W), and alongthe shoreof Honey Lake itself (40ø09'N, 120ø15'W). Juvenile avocetsand stiltswere marked with Utah-specificcolor band combinationsat Great Salt Lake in 1992 and 1993. Banding in Utah was conductedprimarily at Layton Marsh (41ø02'N, 112ø00'W), Ogden Bay Wildlife ManagementArea (41ø12'N, 112ø15'W),Harold Crane Wildlife ManagementArea (41ø20'N,112ø10'W), Bear RiverNational Wildlife Ref- uge (41ø25'N, 112ø15'W),and northeasternFarmington Bay (40ø53'N, 112ø03'W). After trying many different trapping protocols in 1992, the following protocol was adopted in 1993 to minimize disturbanceof breeding birds. Adultswere trapped on the nest after 14 d of incubation.Set trapswere visuallymonitored at a distanceusing a portable blind or field vehicle, and birds were removedimmediately after they discoveredthey were con- fined. Birds were trapped during the heat of the day when they were highly motivated to incubate and were not kept off the nest for more than 20 min. To prevent damageto eggsby overheatingor by the movements of the trapped bird, eggswere temporarilyreplaced with painted wooden replicasduring trapping. The real eggswere concealedin shade near the nest when possible,or stored in an egg carton in the blind, and were replaced immediately after trapping. Members of a breeding pair were not trapped on consecutivedays, and the second bird usuallywas trapped only when eggswere pipped. Chicks were banded within 6 h of hatch, or occasionallywere run down after leavingthe nest. We publicized our banding effort using advertisementsin birding mag- azines, the OrnithologicalNewslett• and by direct contact with refuge managers,local Audubon Societychapters, and colleagues.In addition to the resightingsreported to us, severalmetal bandswere returned to the Bird Banding Laboratory, National Biological Service. Approximately 10,000 wintering avocetswere examined by LWO on each of two census- ing trips in Sinaloa, Mexico. Prior to inclusion in the data set, Tulare Basin resightingswere screenedto remove birds banded locally by C. Marn. Resightingswere classifiedas migratory or wintering, based on chronology of departure from the study site, arrivals of marked birds in the Tulare Basin, and wintering population trendsat Humboldt Bay,California (Evansand Har- Vol.67, No. 2 Avocetand Stilt Movements [311 ris 1994). Sightings any time after departure from the breeding site through October were classifiedas migratory observations.Sightings in November, December,January, or February were classifiedas wintering observations.We did not receive resightingsthat were clearly associated with spring migration. Breeding seasonresightings (April through July) will be presentedelsewhere. Contingencytables were analyzedusing the ConditionalBinomial Exact Test (Rice 1988). Analysesof contingencytables assume that each observation was independent.Although this assumptioncould have been vio- lated by these data, avocetsdid not depart in cohesiveflocks, and were not subsequentlyobserved in flocksof cohesivecomposition. In addition, we attempted to protect the assumptionof independencein constructing contingencytables. In no caseswere membersof a familygroup or former matesincluded in the samecontingency table.

RESULTS AND DISCUSSION Prior to initiation of this study only 13 recoveriesof banded avocets and 10 recoveriesof stiltswere on file with the Bird Banding Laboratory, National BiologicalSurvey. Only nine records,all of avocets,documented movement •50 kin. Eight of these documented a movement from the banding site to a migration or wintering area, and can be summarizedas follows: (1) A chick banded in northcentral California in 1940 wintered near San FranciscoBay. (2) Four avocetsbanded in northwesternNevada 1969-1970 were recoveredat the followingwintering sites: San Francisco Bay, Salton Sea, Sinaloa, Mexico, and Michoacan, Mexico (Mexican recoverieswere recordedonly by statein the 1970s). (3) An avocetchick banded in Kansas in 1962 was recovered that winter in Florida. (4) An avocet chick banded in Nebraska in 1938 was recovered in the winter of 1943 in northern Arizona. (5) A chick banded in central Montana in 1983 was recovered in southeastIdaho that September. Of 110 resightingsreceived during this study,90 were identifiable as individuals banded at Honey Lake (Table 1), 3 were banded in Utah, 5 were banded by other researchers,and 12 could not be identified to banding origin. Sixty-ninepercent of resightings(n = 62, avocetsand stiltscombined) were from evaporationponds in the Tulare Basinwhere a team of researchersconducted daily field observations.Some individuals were seen at more than one site or in more than one year (footnotes to Table 1). A total of 65 individualswas seen •50 km from the banding location in Honey Lake. Migration routesfor birds leaving Honey Lake.--Resighting of migrant avocetsat Mono Lake, San FranciscoBay, and inland California, and the range of wintering sites(see below) suggestthat avocetstake a varietyof migratoryroutes on leavingHoney Lake in summer (Fig. 2). Migrant stilts from Honey Lake were only seen on evaporationponds in the Tulare Basin.Details of resightingsof adultsare presentedin Table 2. The successof the previousbreeding attempt did not appear to influ- ence where a bird was seen as a migrant. Of four adult avocetsseen at 312] J. A. Robinsonand L. W. Oring J.Field Ornithol. Spring 1996

TABLE1. Numbers of individual avocetsand stiltsmarked in the Honey Lake Valley,Cali- fornia, and resightedfarther than 50 km from the bandingsite.

Avocet Stilt

First First Location Adult Year Adult Year

Coastal California Arcata Marshes I a 2 b BolinasLagoon 2 2b San FranciscoBay 1 6 Monterey 1 1 Inland California Sacramento NWR 1 Mono County 4 Grassland Water District 1 Tulare Basin 18c 7 9 d 5 e Edwards Air Force Base 1

Sinaloa, Mexico 2 TOTAL: Migratory/winteringsightings f 30 19 10 6

• This bird wasalso resighted at the Tulare Basin. • One individual was seen twice. c33 totalresightings, 23 resightingsof individualsat differentlocations in the TulareBasin, 17 individuals,but one wasseen in two different years. d 14 totalresightings, 12 resightingsof individualsat differentlocations within the Tulare Basin, 9 individuals. e 8 total resightings,5 individuals. f 90 totalresightings, 72 resightingsof individualsat differentlocations within each region, 65 individuals counted as noted above.

Mono Lake, two left Honey Lake after being banded, and two left Honey Lake after their chickshad fledged.Of 19 adult avocetsseen in the Tulare Basinafter breeding at Honey Lake (18 individualsfrom Table I plus the individualseen again a secondyear), three left Honey Lake immediately after being banded, four left after their chickshatched, two after their chicksfledged, and nine after their nestsfailed. One female seenin fall in the Tulare Basin,was only seenas a springmigrant at Honey Lake. Of nine Honey Lake stiltsseen at the Tulare Basin,three left Honey Lake after banding, one left after chickshatched, three after chicksfledged, and two after nest failure. Sex ratios of avocetsdid not differ among migration sites.The numbers of female to male avocetswere 2:2 at Mono Lake, and 10:7 at the Tulare Basin(p = 0.43). In contrast,more female than male stiltswere seenin the Tulare Basin(7:2, p -- 0.067), suggestingpossible differential migration routes or timing. First-yearavocets were more likely than adultsto be seenat coastalsites. We comparedresightings in Californiabecause there were multiple inland and coastalobservations (Fig. 2). Sevenfirst-year avocets were seen Vol.67, h•o. • Avocetand StiltMovements [313 314] j. A. Robinsonand L. W. O•'ng j. Field Ornithol. Spring 1996 Vol.67, No. 2 Avocetand StiltMovements [315 316] J. A. Robinsonand L. W. (2ring J.Field Ornithol. Spring 1996 inland and 12 were seen at coastal sites. In contrast, 23 adult avocets were seen at inland sitesand 5 at coastalsites (p = 0.001). Birds seenas migrantswithin the Tulare Basinwere often seenon more than one date and sometimesmoved between evaporationponds (6-25 km). Adult avocetswere seen in the Tulare Basin an averageof 1.88 times (median = 2, range = 1-5), and adult stiltswere seen an averageof 1.5 times (median = 1, range 1-3). Adult avocetsseen at least once were later seen at another site (on a different evaporation pond within the Tulare Basin) in 5 of 18 cases (28%), stilts in 3 of 9 cases (33%). For thosebirds seenmore than once in the Tulare Basin,avocets were present a mean minimum of 21.33 days (SD = 10.38, range = 2-26, n = 9), and stiltswere presenta mean minimum of 28.5 days(SD = 14.75, range = 13-48, n = 4, counting a brood of three siblingsas one observation). Thus, at least some individualsspent long periods of time in the Tulare Basin. The complexity of migration movements is demonstrated by avocet male #149 (Table 2). After breeding at Honey Lake, he was seen on migrationin the Tulare Basin (480 km to the south) from 18 Aug.-7 Sep. 1993, and again wintering at Arcata, California (700 km to the north) on 2 Dec. 1993. The bird did not return to Honey Lake to breed in 1994. Members of family groups were resighted together in three cases.(1) Two avocet siblingsin their hatch year were sighted on the same pond in the Tulare Basin on 3 Sep. 1992. One of these siblingswas later seen at the mouth of the SalinasRiver on 17 Sep. 1992, and finally returned and bred in the Honey Lake Valley. (2) Three stilt siblingsin their hatch yearwere sightedtwice on an evaporationpond in the Tulare Basin.Some groupsof siblingsin both speciesmay migrate together as hypothesized by Alberico et al. (1992) for Spotted Sandpipers(Actitis macularia). (3) A pair of stilts that had bred together in the Honey Lake Valley were sightedtogether in the Tulare Basinon 5 Aug. 1992. The male waslater seenalone on 17 Aug. 1992. This observationsuggests the possibilitythat pair bonds extend beyond nesting in this species. Winteringareas.--Avocets wintered along the California coastfrom Ar- cata southwardand along the west coastof Mexico (Fig. 2). Avocet use of Arcata Marshesand Humboldt Bay has been increasingover the last 50 yr (Evans and Harris 1994). The Sinaloa coast is a major wintering area for avocets(Harrington 1992, and unpublishedreports for 1993 and 1994, Morrison et al. 1992). Resightingsof adult avocetswere distributed asfollows: one at Arcata Marshes,two at BolinasLagoon, one at Monterey, one at San FranciscoBay, and two in Sinaloa (at Bahia Santa Maria and Ensenadade Pabellon). Resightingsof first-yearavocets were distributed as follows:two at Arcata Marshes,one at Bolinas Lagoon, two at San FranciscoBay, and two in the Tulare Basin.Based on consecutiveresight- ings, one first-yearavocet spent a minimum of 51 d wintering in Arcata, and another spent 121 d wintering at BolinasLagoon. Stiltsfrom Honey Lake have been sightedin the SacramentoNational Wildlife Refuge com- Vol.67, No. 2 Avocetand StiltMovements [317 plex, and at the ponds of the GrasslandsWater District aswintering birds (Fig. 2). Connectednessof the easternand westernGreat Basin.--To date, none of the birds banded at Great Salt Lake have been seen breeding in the western Great Basin. However, three birds from Great Salt Lake have been resighted at other times of the year. A hatch-yearavocet was seen wintering near Bahia Santa Maria, Sinaloa, Mexico in December 1993. A hatch- year stilt wasrecovered near Navojoa, Sonora, Mexico in November 1992, and a second-yearstilt was seen in the Tulare Basin in September 1993. Thus, it appears that at least some birds from the eastern Great Basin winter in the same locations as do birds from the western Great Basin.

CONCLUSIONS Avocetsand stiltsfrom the western Great Basin rely on the wetlandsof California for stopoversduring migration. Avocetsand stiltsbreeding in California and the western Great Basin share wintering areas, and migrants from northern California mix with breeders at sitesin central Cal- ifornia. Although it is impossibleto distinguishheavy bird use from heavy observationeffort, it is clear that the controversialevaporation ponds of the Tulare Basin (Williams 1994) are usedby many migratory avocetsand stilts that breed in noncontaminated areas. No single wetland complex can currently be identified as most important for migratory avocets.The broad geographicdispersion of wintering American Avocetsfrom Honey Lake wassimilar to that reported for Pied Avocets (R. avosetta)from England (Cadbury and Olney 1978) and France (Watier and Fournier 1980). In contrast,habitat use by migratory Black-neckedStilts was restricted to inland sites.Stilts might depend more stronglythan avocetson wetlands of central California as stopoversites on their way to wintering sitesin western Mexico. We have documented that some avocets and stilts from the eastern Great Basinjoin birds from the western Great Basin at wintering sites along the west coastof Mexico. Thus, the potential for some population mixing exists,as is confirmed by the observation of a Great Salt Lake stilt during September in the Tulare Basin. Since birds from the eastern and western Great Basin share wintering sites, can the Great Salt Lake be consideredpart of the PacificFlyway for recurvirostrids(cf. Page and Gill 1994)? The Nebraskato Arizona migration documented by the Bird Band- ing Laboratory suggeststhat the heuristic constructsof the Pacific and Central Flyways(cf. Morrison 1984) cannot unambiguouslybe applied to avocets.At first glance, our evidence supportsthe concept of Great Salt Lake being part of the Pacific Flyway.However, considering the heavy observer presence in the Tulare Basin, we saw many fewer Utah birds than we would have expected if mixis occurred between the eastern and western Great Basin.At California sitesused during migration, there appears to be nearly complete separationof the eastern and western Great Basin birds. Further observation of the degree of mixing at wintering areas will be necessaryto understand the degree of isolation between 318] J. A. Robinsonand L. W..Oring j. FieldOrnithol. Spring 1996 eastern and western Great Basin. With the exception of birds breeding in the westernGreat Basinand central Galifornia,the continent-widepat- terns of avocet and stilt migration remain unknown. These results are relevant to understanding the effects of irrigation drainwater on avocetand stilt populations.We now know that selenium- contaminatedevaporation ponds in the Tulare Basinare used by avocets and stilts migrating from noncontaminatedsites in northern Galifornia (this study), and the rest of the western Great Basin (Bird Banding Lab- oratory recoveries).Evaporation ponds could have positive impacts (by providing food-rich stopoverareas) as well as negativeimpacts (through selenium bioaccumulation) on migrants. Some birds reside in these ponds for as long as 48 d, long enough for significantselenium bioaccumulation to occur (Heinz et al. 1990). As a corollary, avocetsor stilts collectedon evaporationponds during July (e.g., Ohlendorf et al. 1990) were not necessarilybreeding residentsin the Tulare Basin.The unknown historiesof collected birds could partially explain variabilityin selenium bioaccumulation in livers of collected adults (Ohlendorf et al. 1986b). Our observationslead to severalgeneral hypothesesabout migratory movementsof avocetsand stilts.For avocets,we hypothesizethat (1) av- ocetsmake small "hops" betweenwetland sitesas they move from breeding to wintering areas, (2) avocetsiblings sometimes migrate together to stopoversites, and (3) first-yearavocets have migratorybehavior that dif- fers from adults such that they more often use coastalsites. For stilts,we hypothesizethat (1) stilt siblingssometimes migrate together to stopover sites,(2) stilt pair bonds sometimesextend beyond the breeding attempt, but (3) that male and female stiltsmay differ in their migratory behavior or wintering ranges.These hypotheseswill be testableas our studiesof marked avocets and stilts continue.

ACKNOWLEDGMENTS

Field assistancewas provided by R. Cole, C. Collier, C. S. Elphick, L. Hanaw, Lengyel S., D. Lewis, D. Lutz, L. Powers,E. Larsen, and H. Wilbanks. D. Lutz and S. Warnock helped with data processing.Logistic support in California was provided by D. Brimm, the late J. Dow, Sr., C. Holmes,J. Pruett, and P. Sullivan.Logistic support in Nevada was provided by B. and L. Ayers, A. Janik, G. Lott, L. Neel, and N. Sakke. Logistic support in Utah was provided by V. Bockman,J. Dolling, D. Lee, P. Paton, V. Roy, and P. White, plus many additional volunteers.Access, personnel and/or airboatswere provided by the California Department of Fish and Game, CanvasbackGun Club, Greenhead Duck Club, Mahala Gun Club, Nevada Department of Wildlife, Stillwater National Wildlife Refuge, Bear River Na- tional Wildlife Refuge, Utah Department of Wildlife Resources,and Wild Goose Hunting Club. Resightingsand recoverieswere provided by P. Ashfield, J. de D. B. Mercado, M. Colwell,S. Detwiler, C. S. Elphick, L. Feeney,S. Fellows,IL Garrett, R. Gerstenberg,T. Good- ier, IL Gualtieri, T. Hurd, C. Marn, P.J. Metropulos,J. Morlan, C. Otahal, J. Pearce,v. Roy, F. K. Schleicher,D. Ward, L. Wells,O. Williams,M. Wolder and S. Zador.Without the friendly collaborationand coordinationwith C. Marn and J. Skorupa,band combinationscould not have been assignedaccurately. We also benefited from C. Marn's daily field observationsin the Tulare Basin,and from the recordsprovided by the Bird Banding Laboratory,National BiologicalService. Valuable comments on early draftsof this manuscriptwere providedby C. S. Elphick, R. A. James,Jr., c. Marn, T. A. Sordahl, and N. Warnock. This researchwas funded by the Jay Dow, Sr. Wetlands(formerly B&B Wetlands),a subcontractto LWO from Vol.67, No. ') Avocetand StiltMovements [319

a National Science Foundation EPSCoR grant to North Dakota, the Nevada Agricultural Experiment Station, the Nevada Water ResourcesCenter, the Lahontan Audubon Society George Whittell Nevada Environmental Fund, National Science Foundation Grant DEB- 9196050 to LWO, National Science Foundation Dissertation Improvement Grant DEB- 9422526,and a National ScienceFoundation Graduate Fellowship to JAR.

LITERATURE CITED

ALBERICO,J. A. R. 1993. Drought and predation causeavocet and stilt breedingfailure in Nevada. Western Birds 24:43-51. , J. M. REED,AND L. W. OraNG. 1992. Non-random philopatry of sibling Spotted Sandpipers (Actitis macularia). Ornis Scand. 23:504-508. BOETTCHER,R., S. M. HAIG, AND W. C. BRIDGES.1994. Behavioral patterns and nearest neighbordistances among nonbreeding American Avocets. Condor 96:973-986. CA,BURY, C.J., ANDP.J.S. OI•NEV.1978. Avocetpopulation dynamics in England.Brit. Birds 71:102-121. DAHL,T. E. 1990. Wetlands:Losses in the United States1780's to 1980's.U.S. Department of Interior, Fish and Wildlife Service,Washington, D.C. EVANS,T. J., ;d,4DS. W. tL•RP,IS. 1994. Statusand habitat use by American Avocetswintering at Humboldt Bay,California. Condor 96:178-189. GIBSON,F. 1971. The breeding biologyof the AmericanAvocet (Recurvirostraamericana) in central Oregon. Condor 73:444-454. HAMILTON,R. B. 1975. Comparativebehavior of the AmericanAvocet and the Black-necked Stilt (Recurvirostridae).Ornithol. Monogr. 17. HARRINGTON,B.A. 1992. A coastalaerial winter shorebird surveyon the Sonora and Sinaloa coastsof Mexico,January 1992. Wader Study Group Bull. 67:44-49. HEINZ, G. H., G. W. PENDLETON,A. J. KRYNITSKY,AND L. G. GOLD. 1990. Selenium accu- mulation and elimination in mallards. Arch. Environ. Contam. Toxicol. 19:374-379. HOFFM•'•, D. J., H. M. OHLENDORF,AND T. W. ALDRICH.1988. Selenium teratogenesisin naturalpopulations of aquaticbirds in centralCalifornia. Arch. Environ.Contam. Tox- icol. 17:519-525. JAMES,R. A., JR. 1995. Natal philopatry,site tenacity,and age of first breedingof the Black- necked Stilt.J. Field Ornithol. 66:17-111. MORPOSON,R. I. G. 1984. Migration systemsof somenew world shorebirds.Pp. 125-202, in J. Burger, and B. L. Olla, eds. Shorebirds:migration and foraging behavior.Plenum Press, New York, New York. , R. K. Ross,AND S. TORRES.1992. Aerial surveysof Nearctic shorebirdswintering in Mexico: some preliminary results.Can. Wildl. Serv. ProgressNotes. OHLENDORF,H. M., D. J. HOFFMAN,M. K. SAIVd,AND T. W. ALDPd(:H.1986a. Embryonic mortalityand abnormalitiesof aquaticbirds: apparent impacts of seleniumfrom irrigation drainwater.Sci. Total Environ. 52:49-63. •, R. L. HOTHEM,C. M. BUNCK,T. W. ALDPOC:H,J. F. MOORE. 1986b. Relationships betweenselenium concentrations and avianreproduction. Trans. 51st N. Am. Wildl. Nat. Res. Conf. 51:330-342. --, AND K. C. MAROIS. 1990. Bioaccumulation of selenium in birds at Kesterson Reservoir, California. Arch. Environ. Contam. Toxicok 19:495-507. , --, AND D. WELSH. 1989. Nest success,cause-specific nest failure, and hatch- ability of aquaticbirds at selenium-contaminatedKesterson Reservoir and a reference site. Condor 91:787-796. , --, T. W. ALDm(:H,AND A. J. IrmVNITS•:V.1987. Seleniumcontamination of the Grasslands,a major Californiawaterfowl area. Sci. Total Environ.66:169-183. PAGE,G. W., AN•)R. E. GILL,JR. 1994. Shorebirdsin westernNorth America:late 1800sto late 1900s. Stud. Avian Biol. 15:147-160. PALMER,R. S. 1967. Speciesaccounts. Pp. 139-142 in G. D. Stout, ed.. The shorebirdsof North America. Viking Press,New York, New York. PRATER,T., J. MARCHANT,ANDJ. VUORINEN. 1977. Guide to the identificationand ageing of Holarctic waders.Field Guide 17. British Trust for Ornithology,Tring, Herts. 168 pp. 320] J. A. Robinsonand L. W. Oring J.Field Ornithol. Spring 1996 lhCE,W. R. 1988. A new probabilitymodel for determining exact P-valuesfor 2 X 2 contingencytables when comparingbinomial proportions.Biometrics 44:1-22. RoBI•SO•,J. A., •D S. E. W•ocI•. 1996. The stagingparadigm and wetland conservation in arid environments: shorebirds and wetlands of the North American Great Basin. In- ternationalWader Studies,in press. SK^CE•, S. K. •D E L. I•O?F. 1993. Toward conservation of midcontinental shorebird migrations.Conserv. Biol. 7:533-541. SI•ORUp^,J.P., ^•D H. M. OI•E•DO•. 1991. Contaminantsin drainage water and avian risk thresholds.Pp. 345-368, in A. Dinar and D. Zilbermann,eds. The economicsand man- agement of water and drainage in agriculture.Kluwer AcademicPublishers, Norwell, Massachusetts. SOP•D•-•I•,T. A. 1984. Observationson breedingsite fidelity and pair formation in American Avocets and Black-necked Stilts. N. Am. Bird Bander 9:8-11. U.S. FISH AND WILDLIFESERVICE. 1992. An overviewof irrigation drainwater techniques, impactson fish and wildlife resources,and managementoptions. Division of Environ- mental Contaminants,U.S. Fish and Wildlife Service,Washington, D.C. W^TIER,J.-M., ANDO. FOt:•IE•. 1980. Elements de d•mographie de la population d'Avo- cettes (Recurvirostraavosetta) de la c•te atlantique fran•aise. Oiseau Rev. Fr. Ornithol. 50:307-321. WILLIAMS, M. L., R. L. HOTHEM, AND H. M. OHLENDORF. 1989. Recruitment failure in Amer- ican Avocetsand Black-neckedStilts nesting at KestersonReservoir, California, 1984- 1985. Condor 91:797-802. WtI•I•I•MS, T. 1994. Death in a black desert. Audubon 96(1):24-31. Received26 Apr. 1995; accepted25 Sep. 1995.