WESTEt BIRDS Volume 32, Number 3, 2001 DISTRIBUTION AND ABUNDANCE OF WINTER SHOREBIRDS ON TOMALES BAY, CALIFORNIA: IMPLICATIONS FOR CONSERVATION JOHN P. KELLY,Cypress Grove Research Center, Audubon Canyon Ranch, Marshall, California 94940 ABSTRACT: I analyzedthe distributionand abundanceof winteringshorebirds (Scolopacidae,Charadriidae, and Recurvirostridae) in Tomales Bay, California, on the basisof 57 baywidecounts conducted over 10 years,from 1989-90 to 1998-99. TomalesBay supportsup to 20,689 shorebirdsin earlywinter, thus qualifying as a wetlandof "regional"importance in the Western HemisphereShorebird Reserve Network. Minimum overall shorebird abundance fell as low as 1291 in late winter. TomalesBay supported approximately a thirdof the winteringshorebirds in the Point Reyes/Bodegaarea in early winter.Observations of tidallystructured flock move- mentsof severalspecies suggested that the northernand southernends of Tomales Bay are occupiedby differentwintering groups. In associationwith cumulative seasonalrainfall, most species declined in abundancesignificantly in midwinter.The Sanderlingand MarbledGodwit increased with cumulativerainfall in the north and southbay, respectively, suggesting weather-related influxes from outercoastal beaches. After accountingfor the effectsof cumulativeseasonal rainfall and a 10-yeartrend in annualrainfall, I detectedno trendsin species'abundances. Foraging and roosting shorebirdsat the northernend of the baywere vulnerable to directdisturbance from concentratedrecreational use. Long water-residence times in southernTomales Bay suggestthat shorebirdsthere may be particularlyvulnerable to toxicspills or anthro- pogeniceutrophication. The closenessof San FranciscoBay impliesa highpotential for invasionof nonnativeorganisms established there, which could alter the availabil- ity of benthicprey to shorebirdsin TomalesBay. Shorebirdfeeding habitat at the deltasof Walker and Lagunitascreeks may be adverselyaffected by heavyrainfall leadingto the depositionof sediment.Daily influxesof roostinggulls from a local landfill were associated with reduced shorebird use of tide fiats. Shorebirds' use of open tide fiatsdeveloped for maricultureis reduced,although floating oyster bags provide roosting areas during high tides. Breachinglevees that isolate historic wetlandsmay increaseshorebird use in some areas.The likelihoodof regularor episodicintraseasonal movements among Point Reyes/Bodegaarea wetlandssug- gestsTomales Bay and other nearbywetlands are worthyof broadprotection. WesternBirds 32:145-166, 2001 145 DISTRIBUTION AND ABUNDANCE OF WINTER SHOREBIRDS ON TOMALES BAY TomalesBay is one of California'slargest and leastdisturbed estuaries. It hasbeen recognized as a wetlandof "regionalimportance" on the basisof its supportingup to 20,000 shorebirds,a criterionfor inclusionin the Western Hemisphere Shorebird Reserve Network (WHSRN; Harrington and Perry 1995, Page and Shuford2000). An analysisof shorebird abundance and distribution on the Pacific coast of the United States identifiedTomales Bay as a keywetland for shorebirdconservation because it supportsin at leastone seasonat least1% of the totalpopulation of 8 of 13 shorebirdspecies that concentratein estuariesand brackishwetlands (Pageet al. 1999). In spiteof thisrecognition, little has been published on the spatialor temporalvariations in shorebirdabundance on TomalesBay, or the birds'vulnerability to environmentalthreats. Kelly and Tappen (1998) reportedpreviously on the valueof TomalesBay to otherwinter waterbirds. Shufordet al. (1989) analyzedthe resultsof up to 10 consecutiveyears of shorebirdcounts in the adjacentPoint Reyesarea between1965 and 1982 but providedlittle information on shorebirduse of TomalesBay. As part of a statewidesurvey, Jurek (1974) reportedon fiveyears of monthlyshorebird countsat the Walker Creek delta, in the northernpart of the TomalesBay, but did not surveyabundances baywide. Over recentdecades, Tomales Bay has been suggestedfor protectionas habitatfor shorebirds(Smail 1972, National Oceanicand AtmosphericAdministration 1987, Neubacheret al. 1995, Pageand Shuford2000). However,with one exceptiondealing with the effectsof aquacultureon shorebirds(Kelly et al. 1996), evaluationsof conservationissues in TomalesBay havenot directlyaddressed patterns of shorebird use. In this paper, I (1) presentresults from 10 years of baywidewinter shorebirdcensuses on TomalesBay; (2) comparethese results with other studiesto providea historicaland geographicalperspective on the impor- tanceof TomalesBay to shorebirds;(3) evaluatespecies' distributions within the baywith regardto the importanceof particularareas used by shorebirds; (4) examineprocesses that influenced abundance trends over the 10 yearsof study,and (5) identifyneeds for conservationof winteringshorebirds on TomalesBay. I addressthese objectiveswith regardto all Scolopacidae, Charadriidae,and Recurvirostridaeassociated with TomalesBay and imme- diatelyadjacent seasonal wetlands (Figure 1). STUDY AREA TomalesBay floodsthe lower20 km of the fault-riftOlema Valley on the centralCalifornia coast, about 45 km northwestof San Francisco(Figure 1; Galloway1977). Approximately18% of the bay's28.5-km z areais inter- tidal,providing primarily sand or mudfiats suitable for foragingshorebirds and cobblestonebeaches along the eastshore. In general,sediments grade from primarilyfine to coarsesand in the northernreaches of the bay to muddiersubstrates in southernportions of the bay (Daetwyler1966). Two primarypoints of freshwaterinflow, LagunitasCreek at the south end of the bay and Walker Creek near the north end of the bay, are associatedwith largetidal deltas suitable for foragingshorebirds (Figure 1). Numeroussmaller delta marshes and tidal fiatsoccur where smallperennial 146 DISTRIBUTION AND ABUNDANCE OF WINTER SHOREBIRDS ON TOMALES BAY Dillon Beach 8.7 35.4 Creek 110.3 BlakesLanding 31.9 Cypress Point Pacific Marshall Ocean 55.7 37.5 215.4 Inverness 10.0 0 2 4 km Lagunita• Creek Figure 1. Shorebirdcount areas on Tomales Bay, California. 1, White Gulch; 2, Lawsoh'sMeadow; 3, SandPoint; 4, Tom's Point;5, WalkerCreek delta; 6, North Marshall;7, South Marshall;8, LagunitasCreek delta; 9, GiacominiPasture; 10, Invernessshoreline. Arrows indicatelength of shorelinein each count area; labels indicateextent (ha) of exposedtidal flat at MLLW. and ephemeralstreams enter the bay. Adjacentseasonal wetlands suitable for shorebirdsare normallylimited to about15 ha of a 200-ha dikedpasture at the south end of the bay and approximately20 ha of wet meadow surroundedby sanddunes at the north end of the bay. Additionalseasonal wetlands,used by shorebirdsduring periods of heavy flooding,occur in agriculturalareas northeast of the bay. Most(95%) of the annualrain fallswhen wintering shorebirds are present, from October throughApril, with 55% falling from Decemberthrough February(Audubon Canyon Ranch, unpublished data). Constraints on tidal exchangewith the ocean,imposed by the linearshape of the bay, interact with winter runoffto createcontrasting habitats at the two endsof the bay. In the southernhalf of the bay, variablyhigh levelsof freshwaterinflow duringwinter, low flowsin summer,and long water-residence times result in highlyvariable salinities (Hollibaugh et al. 1988). Salinitiesrange from nearly 147 DISTRIBUTION AND ABUNDANCE OF WINTER SHOREBIRDS ON TOMALES BAY freshafter heavywinter runoff to slightlyhypersaline in late fall (Smithand Hollibaugh1997). In northernTomales Bay, regulartidal mixing maintains salinitiesthat consistentlyreflect those of nearshorewaters along the outer coast.The differencebetween mean high and mean low tides is about 1.1 m, with an averageannual maximum tide swingof about2.5 m (U.S. National Oceanicand AtmosphericAdministration harmonics and correctiontables; Tides and Currents,Nautical Software, Inc.). METHODS ! dividedTomales Bay into 10 subareaswithin which all shorebirdscould be countedin 60 to 90 minutes(Figure 1). Theseareas include almost all of the intertidalflats in the bay,with the exceptionof a few smallareas at creek mouths along the west shore. Teams of qualifiedobservers counted all shorebirdsby speciesin all subareassimultaneously. Observerscounted during risingtides, at tide levelsbetween 0.76 and 1.22 m abovemean lower low water(MLLW) at Blake'sLanding (Figure 1). Count-areaprotocols were coordinated so that adjacent areas were surveyed simultaneously.Abundances generally represented counts of foragingindi- viduals,with estimatesof largemobile flocks made only rarely. During three workshopsheld to examineand reduce observer bias in estimatingflock size, biaseswere inconsistentand withouta clear centraltendency. Therefore observer bias was assumed not to affect overall estimates of abundance. The time and directionof all flock movements,departures, and arrivalsduring count periodswere recordedand examinedlater to minimize chancesof birds being doublecounted. Counts were conductedonly on days with weathersuitable for usingtelescopes to identifyshorebirds. Each year from 1989-90 to 1998-99, we completedapproximately three countsin each of two intraseasonalperiods (n -- 57): early winter (mean 2.8, range 1-4; 1 November-19 December)and late winter (mean 2.9, range 2-3; 15 January-4 March). I summarizedresults within intraseasonalgroups to compareearly- and late-winterpopulation levels. Mostindividual shorebirds were identifiedto species.Occasionally, individu- alswere not identified to speciesand were recorded in pooledspecies groups suchLeast/Western