Seasonal Abundance of Migrant Shorebirds in Baja California Peninsula, Mexico, and California, USA

Seasonal abundance of migrant shorebirds in Baja California Peninsula, Mexico, and California, USA

ROBERTO CARMONA1,3, GORGONIO RUIZ-CAMPOS2 & GEORGINA BRABATA1

•Universidad Autdnoma de Baja CaliforniaSur, Departamentode BiologfaMarina, Apdo. Postal 19-B, La Paz, Baja CaliforniaSur, C.P. 23080, Mexico. [email protected] 2Facultadde Ciencias;3Facultad de CienciasMarinas, Universidad Autdnoma de Baja California,Apdo, Postal 1653, Ensenada, Baja California, C.P. 22800, Mexico. U.S. mailingaddress: PMB 064, PO Box 189003-064, Coronado, California,92178, USA.

Carmona,R., Ruiz-Campos,G. & Brabata,G. 2004. Seasonalabundance of migrant shorebirdsin Baja CaliforniaPeninsula, Mexico, and California,USA. WaderStudy Group Bull. 105: 65-70.

Nearctic shorebirdsuse three main flyways (Pacific, Centraland Atlantic) and the largestnumbers are found in the Pacific Flyway, which includesthe Baja California Peninsula.There, at variousdifferent sites,the followingpatterns of abundancehave been observed: very low numbersin summer;large numbers in autumn; low and stablenumbers in winter; and moderatenumbers in spring.In contrast,further north in California, numbersduring spring migration are ashigh as or evenhigher than those in autumn.We contrastthe differential utilizationby shorebirdsof Baja Californiaduring spring and autumn. In autumn,southbound migrants arrive at sitesin southernCalifornia and the upperGulf of California,and then either continue along the mainland coastof Mexico or alongthe Baja CaliforniaPeninsula. Some continue migrating further south while others stayfor the winter.During the returnmigration in spring,modest numbers occur throughout the peninsula, with rathergreater numbers in the north-west,probably associated with the northwardmigration through the extensive wetlands of Sonora and Sinaloa on the mainland coast.

INTRODUCTION STUDY AREA AND METHODS

Of the 49 shorebirdspecies that breed in the Nearctic, 40 We have analysedshorebird abundance data for five sites migrateto temperateand tropicalareas of Mexico, Central alongthe south-westcoast of the United Statesand in north- and South America for the northern winter (Myers et al. westMexico thathave been previously identified as support- 1987). ing importantpopulations. These are HumboldtBay (north At typicalshorebird stopover and wintering sites through- California),the CentralValley of Californiaand three sites outthe PacificFlyway, there are distinct patterns of seasonal in the Baja California peninsula:Punta Banda, Guerrero abundance.In summer,numbers are very low becausemost Negro and Ensenadade La Paz (Fig. 1). are breedingin the north. In autumn,most sitesrecord the Although there are many siteswith extensiveintertidal largest numbersas first adults and later juveniles migrate mudflatsalong the PacificFlyway in California,we selected south.Lower numbersare present in winter,but theyremain just two (Humboldt Bay and Central Valley) to represent relatively stable becauselittle movement takes place. In thosethat are typical - in termsof bothhabitat and shorebird spring,northward migration leads to increasednumbers as abundanceof the onesused by migrantshorebirds in that birdsreturn to their breedingareas (Page et al. 1979). area (Colwell 1994, Shuford et al. 1998). Of the three major shorebirdflyways in North America All five sitesare influencedby a fairly stable,tropical, (Pacific, Atlantic and Central), the mostimportant in terms high-pressurebelt that bringsdry winds (Coda 1997). Cali- of numbersis the Pacific Flyway (Myers et al. 1987) and fornia andnorth-west Baja Californiahave a Mediterranean- midway along the Pacific Flyway lies the 1,100-kin Baja type climate,while the southernpart of the peninsula(around California peninsula(Fig. 1). There, duringthe pastdecade, 30øN) has an arid climate, except for its southerntip (the systematicdata on shorebirdabundance have beencollected Cape region) which is subtropical(Garcia 1981). for several different coastallocalities (e.g. Palacioset al. The largest areasof intertidal mudflats in California are 1991, Gonzfilez 1996, Carmona& Danemann, 1998). These found in Humboldt, San Francisco and San Diego bays studieshave shownthat the peninsulais particularlyimpor- (Peinadoet al. 1995), while on the west coastof Baja Cali- tant duringautumn migration and of lessimportance during fornia there are mudflats in the bays of Todos Santos,San spring migration (Carmona 1997, Fernfindezet al. 1998, Quint/n, Ojo de Liebre-GuerreroNegro, San Ignacio and Carmona& Danemann1998). In this paper, we presenta Magdalena. On the east coastthe largest mudflats are at generalhypothesis to explainthe patternsof seasonalabun- Ensenada de La Paz in Bahia de La Paz (Carmona et al. danceof migrantshorebirds along the Baja Californiapenin- 1994). sula.

65 Buffetin 105 December 2004 66 Wader Study Group Bulletin

I 110 ø N

1 35 ø C•© UnitedStates

--25 ø Mexico

Pacific Ocean

Fig, 1. Study area and shorebirdcensus sites. (HB = HumboldtBay (north California);CV = Central Valley of California;PB = Punta Banda(north-west Baja California);GN = GuerreroNegro (mid-west Baja California); and LP = Ensenada de La Paz (south Baja California.))

Our analysisrelates to datafrom six studiesduring 1991- becauseno data are availablefor Central Valley and Hum- 1998. Two were carried out at Estero Punta Banda (north- boldtBay), and(2) seasonalabundance of the speciesamong westBaja California)by Palacioset al. (1991) andGonzfilez the threesites having data for all four seasons.In bothcases, (1996) and coveredsampling periods of sevenmonths and whenthe independencehypothesis was rejected in the first one year, respectively. test, we determined which site contributed the most to this The study sites(Fig. 1) were: (1) Humboldt Bay, north rejection,and then removed it from the analysis.This proc- California (38øN, Colwell 1994); (2) CentralValley of Cali- ess was repeateduntil the hypothesiswas accepted(and fornia (35øN, Shufordet al. 1998); (3) Punta Banda, north- thereforea homogeneousgroup was formed). In addition,the west Baja California (31øN, Palacioset al. 1991, Gonzfilez excluded sites from the initial analysis were compared 1996); (4) GuerreroNegro (28øN, Carmona & Danemann mutually,restarting the procedure(ct = 0.05 in all the cases; 1998); and (5) Ensenadade La Paz (24øN, Carmona 1997). Zar 1999). The frequencyof samplingat eachsite was rather variable anddepended on the particularobjectives of eachstudy (see RESULTS Appendix).Therefore, to facilitatecomparisons on bothspa- tial andtemporal scales (i.e. sitesand seasons)we calculated In our monthlyanalysis, Humboldt Bay, California,had the the monthly average abundanceof each speciesand then greatestabundance of shorebirdsin autumn(November) and convertedto percentageson the basisthat the aggregateof spring(April), with lowernumbers in winter(February) (Fig. the monthlyaverages for eachsite was 100%. 2a). In the CentralValley of California, abundanceincreased The datafor HumboldtBay includea numberof separate graduallyfrom autumn(November) to winter (January)to countsfor siteswithin the bay (Colwell 1994), so for each spring (April) (Fig. 2b). At Punta Banda, Baja California, monththese were aggregated.For two sites,Humboldt Bay therewas greater evenness in seasonalabundances, with only and the CentralValley of California, therewere no datafor smallincreases in autumn(October & November)and spring severalmonths. Therefore, for these,the datawere analysed (March) and a notabledecrease in summer(May & June;Fig. on a seasonalrather than monthlybasis as follows: winter 2c). At Guerrero Negro, highest abundanceoccurred in (December-February), spring (March-April), summer autumn(August & September)with much lower but more (May-June) and autumn (July-November). Even on this stablenumbers in winter (December-February)and a small basis,there were no datafor HumboldtBay or CentralVal- increasein spring(April; Fig. 2d). At Ensenadade La Paz, ley for summer(see Appendix). the highestabundance occurred in autumn(September & An independenceZ2 test (ct = 0.05;Zar, 1999)was used October) and winter (January),with no increasein spring to comparethe abundancesof shorebirdsat PuntaBanda as (March & April; Fig. 2e). reportedby Palacioset al. (1991) andGonz•lez (1996). Like- In our analysisby season(Fig. 3), Humboldt Bay, Cali- wise, two types of comparisonsof shorebird abundance fornia, had the largestnumbers during springand autumn amongsites were made using Z 2 tests:(1) seasonalabun- migration,with only a smalldecrease in winter (Fig. 3a). In danceof the speciesamong the five sites(except for summer contrast,numbers in the CentralValley of California were

Buffetin 105 December 2004 Carmonaet al.: Seasonalabundance of migrantshorebirds in Baja CaliforniaPeninsula, Mexico, and California,USA 67

Fig.2. Meanmonthly relative abundance (%) forthe five study sites. Fig.3. Meanseasonal relative abundance (%) for the five study sites.

a. Humboldt Bay a. Humboldt Bay 30 4

0 Aug. Sept, Oct, Nov. Dec, Jan. Feb, Mar. Apr. May June July SPRING SUMMER AUTUMN WINTER

40 b, CentralValley b. Central Valley 35

3O • 25

5 ND ND ND ND o Aug. SepL Oct. Nov. Dec. Jan. Feb. Mar, Apr. May June July SPRING SUMMER AUTUMN WINTER

14 1 40 c. Punta Bande Punta Banda 35

30 • 25

0 Aug. Sept. Oct, Nov. Dec. Jan. Feb. Mar. Apr. May June July SPRING SUMMER AUTUMN W INTER

60 •

d. Guerrero Negro d. Guen•emNegro

2oI

o Aug. Sept. Oct. Nov. Dec. Jan, Feb. Mar. Apr. May June July SPRING SUMMER AUTUMN WINTER

25 e. La Par e. La Paz

0 Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July SPRING SUMMER AUTUMN WINTER

Bulletin 105 December 2004 68 Wader Study Group Bulletin

Table 1. Resultsof •,2tests on the relativeseasonal abundance of shorebirdsat five sites alongthe Pacific shorebirdflyway (excluding summer) (HB = HumboldtBay, CV = CentralValley, PB = Punta Banda, GN = Guerrero Negro, LP = Ensenadade La Paz).

Group compared Z2 P d.f. Site contributing most to the % of the site more rejection of the null hypothesis heterogeneous

HB-CV-PB-GN-LP 61.9 <0.001 8 LP 48 HB-CV-PB-GN 29.7 <0.001 6 GN 57 HB-CV-PB 8.2 0.085 4 - - GN-LP 5.0 0.083 2 - -

Table 2. Resultsof •,2 tests on the relativeseasonal relative abundance of shorebirdsat three sites in the peninsulaof Baja California,Mexico (including summer) (PB = Punta Banda,GN = GuerreroNegro, LP = Ensenada de La Paz).

Group compared Z2 P d.f. Site contributing most to the % of the site more rejection of the null hypothesis heterogeneous

PB-GN-LP 42.8 <0.001 6 PB 58 GN-LP 4.9 0.18 3 - -

Table 3. Seasonal peak counts of shorebirdsat the five study sites (ND = no data available).

Humbolt Bay Central Valley Punta Banda Guerrero Negro Ensenada De La Paz

Winter (Dec.-Feb.) 66,000 374,000 1,100 33,000 12,000 Spring (March-Apr.) 83,000 393,000 800 10,000 1,700 Summer(May-June) ND ND 600 1,900 470 Autumn (July-Nov.) 36,000 177,000 1,300 76,000 13,000

greatestin spring, and there were more in winter than in The sameresult emerges when the datafor the threesites autumn(Fig. 3b). In Baja California, PuntaPanda held the in Baja California(including summer) are analysed together: largestnumbers in winter and spring,fewer in autumnand the patternat PuntaBanda is shownto be significantlydif- only smallnumbers in summer(Fig. 3c). At GuerreroNegro, ferentfrom the othertwo, GuerreroNegro andEnsenada de highestnumbers occurred in autumn,relatively small num- La Paz, which are similar (Table 2). bersin spring,but the winteringpopulation was fairly high; only tiny numbersoccurred in summer(Fig. 3d). The pattern DISCUSSION was similar at Ensenadade La Paz with higher numbersin autumnthan in springand only smallnumbers in summer. Our studyfocuses on patternsof spatialand temporal abun- However, winteringnumbers were greaterthan at any other dance,not absolutenumbers. It is importantto rememberthis time of year (Fig. 3e). becauseabsolute numbers using a site may be muchgreater Comparingthe numbersreported for Punta Banda by than those counted at any one time becauseof turnover, Palacioset al. (1991) and Gonzglez (1996), we found that especiallyduring spring and autumnmigration. However, to therewere no significantdifferences (Z 2 = 3.87,d.f. = 5, give an indicationof the scaleof shorebirdmigration through p = 0.43). However,we decidedto usethe data of Gonzglez the five studysites we presentdata on the peak countsfor (1996) only becauseit covereda full year ratherthan just each season(Table 3). seven months. As expected,the lowestnumbers at the threePeninsula Comparingseasonal numbers across all five studysites siteswere recordedin summerwhen mostnearctic migrants exceptsummer (Table 1), it is evidentthat Ensenadade La arebreeding in the far north(Page et al. 1979,Rappole et al. Paz andGuerrero Negro contributed the mostheterogeneity. 1993). However, relatively more were recordedat Punta Once thesetwo sitesare removedfrom the analysis,those Bandain the north of the peninsulathan at the two sitesin remaining(Humboldt Bay, CentralValley andPunta Banda) the south.This might reflect a tendencyfor non-breeding can be consideredhomogeneous. Similarly, comparison birds (mainly one-year-olds)to over-summerclose to the betweenGuerrero Negro and Ensenadade La Paz indicates nestingareas (McNeil et al. 1994). thesesites are similar (Table 1). In summary,the statistical It is noteworthythat at all sites(except Guerrero Negro) testsindicate that there are two homogeneousgroups: the thewintering population was very similarto thepeak popu- three northern sites (Humboldt Bay, Central Valley and lation in the rest of the year. This may meanthat in winter PuntaBanda) and the two peninsularsites (Guerrero Negro eachsite is usedto carryingcapacity (though detailed feed- and Ensenadade La Paz). ing studieswould be neededto confirmthis). • Bulletin105December2004 Carmonaet al.: Seasonalabundance of migrantshorebirds in Baja CaliforniaPeninsula, Mexico, and California,USA 69

... 110i ø N

• United United • States States

Mexico •25 ø Mexico

Pacific Ocean Pacific Ocean

Autumn Spring

Fig. 4. Modelof migratorymovements of shorebirdsin autumnand springalong the coastsof the south-west of the United States of America and north-west Mexico.

In spring,relatively fewer birdsused the two sitesin the efforts in the southernparts of the peninsulashould be south of Baja California than Punta Banda or the sites in focussedon the seasonsof autumn and winter; while in the California, but in autumnthe patternwas almostreversed northmore effort shouldbe put into winterand spring. with largenumbers passing through Guerrero Negro and La The Baja Californiapeninsula is an importantpart of the Paz (a patternconfirmed by our statisticalanalysis). Our Pacificshorebird flyway, especiallyin autumn.Moreover hypothesisis thatthis indicates differential use of the Baja numericallylarge groupsof shorebirdswinter there. How- Californiapeninsula in springand autumnwith manypas- ever,it remainsrather isolated during the springmigration. sagebirds moving southwards down the peninsula in autumn (andthen onwards along the southernMexican coasttowards ACKNOWLEDGMENTS SouthAmerica) but mostlyfollowing the mainlandcoast of Mexico in spring(Fig. 4). We thankthe personnelof the aquaticbirds program of the Thuswe suggestthat in autumnmigrants coming from the UniversidadAut6noma de Baja California Sur, for theirsup- northtravel alongthe Pacificcoast until they arriveat the portand help in the field. The first authorthanks CONACYT base- the northernend - of the Baja Californiapeninsula. for the doctoralgrant that permitted the accomplishmentof There they have two options:to follow the mainlandcoast the presentwork. We alsothank Nils Warnockfor comments of Mexico through Sonora and Sinaloa or to follow the on an earlierdraft that substantiallyimproved our paper. peninsulaof Baja California.Some take oneroute, some the other.Those that follow the peninsulapass through to win- REFERENCES teringareas further south while othersstay for the winter.In spring,the northwardmovement of birdsthat have wintered American Ornithologists'Union (A.O.U.). 1998. Checklist of North in theBaja California peninsula leads to onlya modestpas- American birds'.Allen Press,Lawrence, Kansas. sagethrough La Paz and GuerreroNegro andfew birds,if Carmona,R. 1997.Riqueza especifica, distribuci6n y abundanciade aves any, passthrough from wintering sitesto the south.At this acufiticasen la Ensenadade La Paz, B.C.S., Mdxico. Unpublished Final Report CONABIO. time, mostpassage migrants follow the mainlandcoast of Carmona, R. & G. Danemann.1998. Distribuci6n espaciotemporal de Mexico.There are probably two mainreasons why theydo avesen la salinade GuerreroNegro, B.C.S., Mdxico. CienciasMari- this.First, following the coast may be saferthan making the nas 24: 389-408. 250 km sea-crossingto the southerntip of the peninsula. Carmona, R., J. Guzmfin, S. Ramirez & G. Fernfindez. 1994. Breed- Second,there are somelarge wetlandsalong the coastsof ing waterbirdsof La Paz Bay, Baja California Sur, Mdxico. Western Birds 25: 151-157. Sinaloaand southern Sonora (Engilis et al. 1998)that prob- Colwell,M.A. 1994.Shorebirds of HumboldtBay, California: abundance ably affordgood feeding opportunities for shorebirdsusing estimatesand conservationimplications. Western Birds 25:137-145. this route. Coria,R.B. 1997.Climatologia. pp. 27-34. In: L. Arriaga& R. Rodriguez We hopeour resultswill providea focusfor futurestud- (eds). Los oasis de la Peninsula de Baja California. Centro de iesof thephenology of shorebirdspassing through the Baja InvestigacionesBio16gicas del Noroeste,M6xico. Californiapeninsula as well asfor thetargeting of conserva- Cox, G.W. 1985.The evolutionof avianmigration systems between temperateand tropical regions of the new world. AmericanNat. 126:451- tionaction. Future studies should test our hypothesis for sev- 474. eral of the commonerspecies, using a variety of methods Engilis,A., L.W. Oring,J. Cartera, W. Nelson& A. Martlnez-L6pez. including systematiccounts, banding etc. Conservation 1998. Shorebird surveysin EnsenadaPabellones and Bahia Santa Bulletin105December2004• 70 Wader Study Group Bulletin

Maria, Sinaloa, Mdxico: critical winter habitats for Pacific flyway Myers, J.P., R.I.G. Morrison, P.Z. Antas, B.A. Harrington, T.E. shorebirds. Wilson Bull. lie: 332-341. Lovejoy, M. Sallaberry, S.E. Senner & A. Tarak. 1987. Conserva- Fernfindez, G., R. Carmona & H. de la Cueva. 1998. Abundance and tion strategyfor migratory species.American Scientist 75: 19-26. seasonalvariation of westernsandpipers (Calidris mauri) in B.C.S., Page, G., L. Stenzel & C. Wolfe. 1979. Aspectsof the occurrenceof Mexico. Southwest. Nat. 43: 57-61. shorebirdson a centralCalifornia estuary. pp. 15-32. In: F. Pitelka(ed.). Garcia, E. 1981. Modificacidn al sistemade clasificacidnclimdtica de Shorebirdsin marine environments.Studies in Avian Biology. USA. Koppen(para adaptarloalas condicionesde la ReptiblicaMexicana. Palacios, E., A. Escofet & D.H. Loya-Salinas. 1991. E1 Esterode Punta Mdxico D.F. Banda, B.C., M•xico como eslab6n del "Corredor del Pacffico": Gonzfilez, G.S. 1996. Aves playeras migratorias del Estero de Punta abundanciade avesplayeras. Ciencias Marinas 17: 109-130. Banda,Baja California,a travdsde un ciclo anual. Bachelor'sThesis. Peinado,M., F. Alcarfiz,J.L. Aguirre,J. Delgadillo& J. J•lvarez. UniversidadAut6noma de Baja California, Mdxico. 1995. Similarity of zonation within California-Baja California and Hayman, P., J. Marchant & T. Prater. 1988. Shorebirds:an identifi- mediterranean salt marshes. Southwest. Nat. 40: 388-405. cation guide. Houghton Mifflin, USA. Rappole, J.H., E.S. Morton., T.E. Levejoy & J.L. Ruos. 1993. Aves Massey, W.B. & E. Palacios. 1994. Avifauna of the wetlandsof Baja migratoriasnetirticas en los neotr6picos.Conservation and Research California, Mdxico: current status.Studies in Avian Biology 15: 45- Center, National Zoological Park, Virginia. 57. Shuford, W.D., G.W. Page & J.E. Kjelmyr. 1998. Patternsand dynamics McNeil, R., M.T. Diaz & A. Villeneuve. 1994. The mysteryof shorebird of shorebirdsuse of California'sCentral Valley. Condorlee: 227-244. over-summering:a new hypothesis.Ardea 82: 143-152. Zar, H.J. 1999. Biostatisticalanalysis. Prentice Hall, New Jersey.

APPENDIX

Number of monthlycounts carried out in the shorebirdstudies along the Pacific Flyway on which our analyses are based. The year is givenin parentheses:(), followedin the caseof Colwellb by the numberof sitesvisited in braces:{ }. (BC = Baja California,CA -- California, ND = No data available)

Punta Bandaa HumboldtBay b PuntaBanda c Ensenadade La Pazd GuerreroNegro e CentralValley f BC CA BC BC BC CA

January 2(89) ND 3(90) 2(97) 1(96) 1(93); 1(94); 1(95) February 2(89) 1(91){37};1(92){32};1(93){30} 3(90) 2(97) 1(96) ND March 2(89) ND 3(90) 2(97) 1(96) ND April 3(89) 1(91){30};1(92){26};1(93){27} 4(90) 2(97) 1(96) 1(92);1(93);1(94) May ND ND 3(90) 2(97) 1(96) ND June ND ND 3(90) 2(97) 1(96) July ND ND 3(90) 2(97) 1(96) August ND 1(91) { 22 }; 1(92) { 29 } 3(90) 2(97) 1(96) 1(92); 1(93); 1(94) September ND 1(90) { 32 } 3(89) 2(97) 1(96) ND October ND ND 4(89) 2(97) ND ND November 3(88) 1(90){33};1(91){31 };1(92){27} 4(89) 2(96) 1(96) 1(93);1(94) December 2(88) ND 3(89) 2(96) 1(95); 1(96) ND a Palacioset al. 1991;b Colwell1994; c Gonzglez1996; d Carmona1997; e Carmona& Danemann1998; f Shufordet al. 1998.

Bulletin 105 December 2004