Seasonal Variation in Diurnal and Nocturnal Distributions of Nonbreeding Shorebirds at North Humboldt Bay, California’

The Condor98: 196-207 0 The CooperOrnithological Society 1996

SEASONAL VARIATION IN DIURNAL AND NOCTURNAL DISTRIBUTIONS OF NONBREEDING SHOREBIRDS AT NORTH HUMBOLDT BAY, CALIFORNIA’

SARAH L. DODD*ANDMARK A. COLWELL Wildlife Department,Humboldt State University,Arcata, CA 95521

Abstract. Recent studiesof nocturnalforaging by shorebirds(Charadriiformes: Charadrii) suggestthat many speciesfeed at night. Much of this researchhas been qualitative and/or restricted to a small portion of the annual cycle (e.g., a few nights or one season)making it difficult to evaluate the extent to which nocturnal foragingvaries seasonally.Consequently, we examined seasonalvariation in abundance and distribution of diurnal and nocturnal foraging shorebirdsfrom 10 Jan 1992-10 Jan 1993 at North Humboldt Bay, California. Shorebirdsforaged primarily during the day. Overall, day/night frequency of occurrence (percentof censuseswith birds) was 87%/48%. In fall, frequencyof occurrenceof shorebirds differed less between day and night (day/night: 82%/64%) than during spring (day/night: 79%/14%) or winter (day/ninht: 100%/42%). Moreover. nocturnal abundance of Marbled Godwit (iimosa fedoh), -Wili& (Catoptrophorussemipaimatus), dowitchers (Limnodromus scolopaceusand L. griseus),and Black-bellied Plover (Pluvialis squatarola)peaked in fall, whereasdiurnal abundanceof thesespecies peaked in winter (Marbled Godwit, Black-bellied Plover, and dowitchers) or spring (Willet). Taxa varied in day/night patterns. For Scolopacids,diurnal abundancesignificantly ex- ceedednocturnal abundance. However, abundanceofAmerican Avocet (Recurvirostraamer- icana), Black-bellied Plover, and Semipalmated Plover (Charadriussemipalmatus) did not differ significantly between day and night. Our results suggestthat a researcher’schoice of seasonor taxa may influence observed patterns of diurnal and nocturnal distributions of shorebirdsconsiderably. Key words: shorebirds;nocturnal foraging; seasonalvariation: nonbreeding distributions; Humboldt Bay.

INTRODUCTION Goss-Custard et al. 1977, Baker 198 I, Barnard The extent to which birds forage at night has and Thompson 1985, Zwarts et al. 1990) or to received considerable attention (McNeil 199 I, a limited extent (Heppleston 197 1, Evans 1976, McNeil et al. 1993). Increasing evidence suggests Tree 1979, Pienkowski 1982, Hockey 1984, that many shorebird species (Charadriiformes: Pienkowski et al. 1984, Maron and Myers 1985, Charadrii) of several families forage at least oc- Zwarts et al. 1990, Manseau and Ferron 1991) casionally at night throughout the year and across are frequent in the literature. Moreover, evidence a broad range of latitudes (Dodd 1995). How- suggeststhat nocturnal foraging by shorebirds at ever, the relative frequency with which shore- northern latitudes is confined principally to win- birds forage nocturnally is largely unknown (sen- ter (Goss-Custard 1969, Heppleston 197 1, Pien- su Robert et al. 1989, Mouritsen 1994). kowski 1982, Stenzel et al. 1976, Evans and Har- Robert et al. (1989) suggestedthat shorebirds ris 1994). The ecologicalexplanation for this pat- forage regularly at night in response to the pe- tern has been that shorebirds at northern lati- riodicity and availability of tidally influenced tudes are forced to feed at night in winter because feeding habitat. However, research conducted energy requirements cannot be met during pe- mainly at northern latitudes suggeststhat shore- riods of short day length (Goss-Custard 1969, birds are predominately diurnal feeders. Ac- Heppleston 197 1, Goss-Custard et al. 1977, Put- counts of shorebirds not foraging at night (Feare tick 1984, P. R. Evans 1988). 1966, Hartwick and Blaylock 1979, Puttick 1979, However, shorebirdsforage extensively at night in the tropics (McNeil and Robert 1988, Robert and McNeil 1989, Robert et al. 1989, Swennen 1990, Zwarts et al. 1990, RomprC and McNeil I Received 26 June 1995. Accepted 9 February 1996. * Present address:Woodland Farm, Rt. 2 Box 60-C 1994) where for most of the year day length is Yemassee, SC 29945. longer and temperatures are warmer than at

11961 SHOREBIRD FORAGING ECOLOGY 197 northern latitudes (Robert et al. 1989). Studies ing rising tides becauseshorebirds foraged along conducted at southern and northern latitudes in- the advancing tide edge eventually congregating dicate that shorebirds also forage at night during near the shoreline, which allowed close obser- spring (Burger 1984, Swennen 1990, Zwarts 1990, vation. We censusedonly one plot within a 24- Zwarts et al. 1990, Romprt and McNeil 1994) hour period. Within each week, we randomly and fall (Manseau and Ferron 1991; Mouritsen chosea 24-hour period in which to censusa plot 1992, 1994; RomprC and McNeil 1994). from 24-hour periods that had tides of sufficient Protection and management of shorebird pop- magnitude so the water reached the plot both in ulations requires complete knowledge of their darkness and in daylight. However, these tidal activity patterns during both day and night. conditions did not occur for four to seven days However, it is difficult to evaluate the extent to on 13 occasions in mid-winter and summer. which distributions of foraging shorebirds vary Consequently, we conducted additional censuses between day and night. Few studies (Zwarts et during other weeks. This resulted in a fairly even al. 1990, Robert et al. 1989, Manseau and Ferron sampling effort acrossthe year and from all hours 199 1, Mouritsen 1994, RomprC and McNeil ofa 24-hour period (Dodd 1995). We categorized 1994) quantified abundance and/or frequency of (a priori) censusesby seasonbased on migratory occurrenceof shorebirdson feeding groundsdur- patterns of shorebirds in the Humboldt Bay area ing both day and night. In addition, few studies (Gerstenberg 1979, Harris 1991) as follows: 1) observed shorebirds for more than a few nights fall: 1 July-30 November (n = 85); 2) winter: 1 (Dodd 1995). Only one study (Rompre and December-l 7 March (n = 50); 3) spring: 18 McNeil 1994) quantified both diurnal and noc- March-8 May (n = 29); and 4) summer: 9 May- turnal abundance of a foragingshorebird for more 30 June (n = 22). than one season. In this paper, we compare di- We conducted day censusesbetween nautical urnal and nocturnal foraging patterns of several sunrise and nautical sunset and night censuses speciesof shorebirdsin a northern latitude, coastal between nautical sunsetand nautical sunrise. We bay for a year. Our objectives are to: 1) examine considered nautical sunrise and nautical sunset day/night variation in abundanceof foragingbirds to occur when we could and could not see, re- and frequency of occurrence of birds; and 2) as- spectively, channel markers located 100 m away sessseasonal and interspecific variation in day/ for day censusesand 50 m away for night cen- night patterns of birds. suses.Overall, 3.5% of censuses(three day and ten night) began or ended within 15 minutes of nautical sunrise or sunset. We conducted each STUDY AREA AND METHODS censusfor 45 minutes beginning when the edge We studied shorebirds from 10 January 1992- of the rising tide crossedthe comer stake of a 10 January 1993 at the Arcata Marsh Project in plot. North Humboldt Bay, Humboldt County, Cal- We observed birds with binoculars from a ve- ifornia (Fig. 1). North Humboldt Bay is the larg- hicle parked 2 l-5 1 m away on dikes. We arrived est of three basins comprising Humboldt Bay at observation points at least 30 minutes before with approximately 12.2 km2 of exposed tidal a census to minimize disturbance to birds. At mud flat at mean low tide (Costaand Stork 1984). night, we verified observations by briefly (three Local tides are characterizedby two uneven high to five seconds)illuminating plots with a 400,000 and low tides each 24 hours. Gerstenberg (1972) candlepower spotlight (Black Max Q-beam, and T.J. Evans (1988) provide detailed descrip- Brinkmann Corp., Dallas, TX) covered with a tions of the study area. red snap-on filter (Brinkmann Corp.). We used We established four 20 x 50 m study plots a red filter becauseit reduced spotlight glare and (marked with four 5 x 5 cm wooden comer did not startle birds as often as unfiltered light stakes)on high elevation mud flats (Gerstenberg (Dodd, unpubl.). However, during foggy condi- 1972, 1979; Evans and Harris 1994) within 50 tions, we used an amber filter because amber- m of the shoreline in areas of high shorebird use filtered light best penetrated fog. We illuminated (Dodd, pers. observ.) (Fig. 1). For each plot, we plots immediately after we had recorded data on censused twice within a 24-hour period each species,abundance, and behavior (see below) of week: once during daylight and once during dark- birds. Additionally, we illuminated a plot when- nesson successiverising tides. We censuseddur- ever we: 1) heard, but could not see shorebird(s) 198 SARAH L. DODD AND MARK A. COLWELL

HUMBOLDT COUNTY

CALIFORNIA

NORTH HUMBOLDT BAY

, ARCATA I r I \

FIGURE 1. Location of four 20 x 50 m plots (shaded rectangles)within North Humboldt Bay, Humboldt County, California. Dashed lines represent four areas encompassingplots.

or 2) had not detectedshorebirds for five to seven lows: 1) loafing; 2) comfort movements including minutes. bathing and preening; and 3) other (e.g., alert and For each census, we recorded the maximum agonistic behavior). number of foraging birds of each species.We also At night, we had difficulty identifying some categorizedbehaviors of nonfeeding birds as fol- closely related species. Consequently, we com- SHOREBIRD FORAGING ECOLOGY 199 bined observations of Western and Least Sand- the months of March and May. Consequently, pipers (Calidris mauri and C’. minutilla, respec- we averaged counts from l-8 May and all of tively) into the group “small sandpipers” and April together, whereaswe averagedcounts from Long-billed and Short-billed Dowitchers (Lim- 1-17 March and 18-3 1 March separately. Ex- nodromus scolopaceusand L. griseus, respec- cluding summer, this produced averagesfor each tively) into the group “dowitchers”. In subse- speciesper plot for 11 day and 11 night monthly- quent analyseswe refer to thesegroups as species. intervals (see Dodd 1995 for distribution of To evaluate whether or not data from plots counts among plots and monthly intervals). A were representative of surrounding mud flats, we problem with averaging data acrossmonthly in- recordedthe presenceof shorebirdson four areas tervals was that we conducted only one census (area 1 = 1.8 ha, area 2 = 3.7 ha, area 3 = 1.5 for plots 3 and 4 during l-17 March and 18-31 ha, area 4 = 2.3 ha) encompassingplots (Fig. 1). March, respectively. We used observed values Additionally, at night, we recorded whether or (rather than means) for these data points. Here- not we heard shorebirds calling from any loca- after, all referencesof abundance are to foraging tion on mud flats encompassingplots. We noted individuals. the species and approximate location of vocal- Comparisonsof Day/Night and Seasonal Pat- izing birds; we disregarded calls coming from terns. For each species,we compared day/night known shorebird roosts and from birds flying and seasonal abundances using nonparametric overhead. (rank-transformation; Conover 1980, Dowdy and Wearden 199 1) two-factor ANOVA (Hintze 1992). We used day/night, season (fall, .winter, DATA SUMMARY AND ANALYSIS and spring) and the interaction term (day/night Data Summary. We omitted summer observa- x season)as main effects;and ranked abundance tions from analyses because we observed few (monthly-interval means) as the response vari- shorebirds during this period (Dodd 1995). Ad- able. We used nonparametric ANOVAs because ditionally, we removed July-September censuses data did not meet assumptions of normality and (n = 51) from analyses of Dunlin (Calidris al- homogeneity of variances (Dowdy and Wearden pina) because this speciesis a late fall migrant 199 1). We used a relatively conservative alpha and usually does not arrive at Humboldt Bay level of significance of co.01 because we con- until late September (Gerstenberg 1972, Dodd ducted a separate comparison for each species. 1995). With the exception of American Avocet (Recurvirostraamericana), we used plot data for analysesof all speciesbecause plot and area data RESULTS obtained similar day/night and seasonalpatterns of frequency of occurrence. For American Avo- OVERALL PATTERNS. cet, we used area data (instantaneous day and We observed 12 and 11 speciesduring day and night counts [n = 1331of foraging birds) because night censuses,respectively (Table 1). A few un- data from plots did not representthe surrounding common speciesoccurred exclusively in the day mud flat (Dodd 1995). We examined day/night (Long-billed Curlew, Numenius americanus; patterns of foraging shorebirds using two re- Wilson’s Phalarope, Phalaropus tricolor; and sponse variables: frequency of occurrence and Baird’s Sandpiper, Calidris bairdii) or night foraging abundance. (LesserYellowlegs, Tringa falvipes and Killdeer, Frequency of Occurrence. For each species,we Charadrius vociferus).Most shorebirds (93% of determined seasonal (fall, winter, and spring) diurnal observations and 99% of nocturnal ob- percent of day and night censusesin which at servations) observed on plots foraged. We ob- least one shorebird occurred on plots. served nearly seven times as many shorebirds Foraging Abundance. To reduce temporal de- during the day (22,180) than at night (3,237). pendence (of abundances) among multiple cen- Moreover, shorebirds occurred in a greater pro- susesconducted close in time (Hurlbert 1984) portion of day censuses(87%) than night cen- we calculatedmonthly averagessummarized from suses (48%) (Table 2). However, frequency of data on the abundance of foraging shorebirds for occurrence of shorebirds differed less between each plot. However, seasonal intervals divided day and night during fall (day/night: 82%/64%) 200 SARAH L. DODD AND MARK A. COLWELL

TABLE 1. Occurrence, number of birds, and behaviors of shorebird speciesobserved on plots from 10 Jan 1992-10 Jan 1993 (n = 186 paired [day and night] censuses).

Percentof birds” Comfort OCCUIRllC-+ Number birdv FCGigiIlg LCJafing movement Other Species Day Night Day Night Day Night Day Night Day Night Day Night

American Avocet 93 45 883 1,213 98.3 97.0 0.8 2.9 0.8 0.0 0.1 0.0 Marbled Godwit 100 11 2,618 37 81.0 100.0 2.6 0.0 13.1 0.0 3.3 0.0 Willet 67 5 824 7 62.0 100.0 20.9 0.0 16.7 0.0 0.4 0.0 Dowitchers 59 20 1,571 159 93.6 100.0 6.1 0.0 0.2 0.0 0.2 0.0 Black-bellied Plover 18 22 47 23 34.0 100.0 66.0 0.0 0.0 0.0 0.0 0.0 Semipalmated Plover 10 17 21 43 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 Dunlin 62 18 4,447 736 86.9 99.3 10.6 0.0 2.2 0.7 0.3 0.0 Small sandpipers 80 32 11,752 1,015 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 Killdeer 0 1 0 0.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 Long-billed Curlew 4 0 10 :, 80.0 0.0 20.0 0.0 0.0 0.0 0.0 0.0 Greater Yellowlegs 2 2 100.0 50.0 0.0 0.0 0.0 0.0 0.0 50.0 Lesser Yellowlegs d 1 :, 1 0.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 Baird’s Sandpiper 1 0 1 0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Wilson’s Phalarope 1 0 1 0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Shorebirdsd 144 79 22,180 3,237 93.0 98.7 3.8 1.1 2.7 0.1 0.5 0.0 a For eachspecies, percent of numberof birds exhibiting eachbehavior. b Total number of censusesin which eachspecies occurred. cFor eachspecies, total number of shorebirdsobserved during 186 day and 186 night censuses. d All speciesof shorebirdsincluded.

than during winter (day/night: 100%/42%) or semipolmatus)occurred only at night on plots in spring (day/night: 79%/ 14%). spring (Table 2). Six of eight speciesoccurred at night mainly during fall (Table 2). Moreover, seven of eight speciescalled at night from mud INDIVIDUAL SPECIES PATTERNS flats during a greaterproportion of fall than spring We confined subsequent analyses to the eight or winter censuses(Fig. 2). By contrast, during most numerous and frequently occurring species the day, five of eight species’ occurrencespeaked on plots and areas. in winter (Table 2). Frequency of OccurrencePatterns. Most spe- Comparisonsof Day/Night and Seasonal Pat- cies occurred primarily during the day during fall terns. For five of eight species, day abundance (six of eight species),winter (seven of eight spe- significantly exceeded night abundance (Tables cies), and spring (seven of eight species)(Table 3,4). A significant interaction between day/night 2). However, Semipalmated Plover (Charadrius and seasonal effects resulted for Marbled God-

TABLE 2. Temporal (day/night and seasonal)variation in frequency of occurrenceof shorebirdson plots. For each species,values are the percent of censusesin which at least one shorebird occurred on plots.

Fall (n = 85) winter (n = 50) Spring(n = 29) Total- (n = 164) Species Day Night Day Night Day Night Day Night American Avoce@ 77.6 60.0 79.6 63.2 69.0 32.1 76.7 56.2 Marbled Godwit 55.3 12.9 74.0 ::: 55.2 ::: 61.0 6.7 Willet 25.9 4.7 58.0 55.2 40.9 3.1 Dowitchers 29.4 22.4 56.0 20.7 36.0 12.2 Black-bellied Plover 9.4 24.7 16.0 ::: :.: 00:: 11.0 13.4 Semipalmated Plover 10.6 16.5 2.0 2.0 6.9 6.1 10.4 Dunlinc 73.5 32.4 56.0 8.0 31:o 10.3 54.9 15.9 Small sandpipers 55.3 21.2 56.0 22.0 17.2 10.3 48.8 19.5 Shorebird@ 82.4 63.5 100.0 42.0 79.5 13.8 86.7 48.2

’ Summercensuses (n = 22) excluded. ’ Area data usedinstead of plot data. Number of censuses:fall = 85; winter = 49; spring = 28 (night), 29 (day);total = 163 (day), 162 (night). r Early fall censusesexcluded. Number of censuses:fall = 34; total = 113. d All speciesof shorebirdsincluded.

202 SARAH L. DODD AND MARK A. COLWELL

TABLE 3. Comparison of diurnal and nocturnal abundances(response variable) of eight shorebird taxa presented as results of nonparametric two-factor ANOVAs with day/night, season(fall, winter, and spring), and the interaction term (day/night x season)as main effects.

SpXlt% Source of variation df F P-Vhle

American Avocet” Day/night 1 2.62 0.11 Season 2 1.57 0.21 Day/night x season 2 1.21 0.30 Error 80 Marbled Godwit Day/night 1 115.01 <0.0001 Season 2 0.17 0.84 Day/night x season 2 9.48 0.0002 Error 82 Willet Day/night : 93.22 10.000 1 Season 3.41 0.04 Day/night x season 2 6.71 0.002 Error 82 Dowitchers Day/night 1 27.50 <0.0001 Season 2 2.65 0.08 Day/night x season 2 9.67 0.0002 Error 82 Black-bellied Plover Day/night 1 0.18 0.67 Season 2 4.54 0.013 Day/night x season 2 6.24 0.003 Error 82 Semipalmated Plover Day/night 1 1.79 0.18 Season 2 4.58 0.013 Day/night x season 2 0.54 0.58 Error 82 Dunlin Day/night 1 20.81

Robert et al.‘s study occurred during a three- Mauritania. Most species(including Dunlin and month winter period, so a complete evaluation Black-bellied Plover) foraged at higher densities of seasonalvariation was not possible. Recently, during the day, although nocturnal densities of Romprt and McNeil (1994) compared the pro- Red Knot (Calidris can&us) and Bar-tailed God- portion of Willets foraging in the day and night wit (Limosa lapponica) exceededdiurnal densi- from October-May in northeastern Venezuela. ties during their premigration period. Manseau Willets foraged as often at night as during the and Ferron (199 1) compared abundance of Semi- day, with the exception of fall (October and No- palmated Sandpiper (Calidris pusilla) between vember) when they fed more at night. In our day and night on four 50 x 50 m plots at the study, nocturnal abundance of Willets also peaked Bay of Fundy during a fall migratory stopover. in fall, but day abundance exceedednight abun- More Semipalmated Sandpipers occurred in the dance during all seasons. day than at night. Recently, Mouritsen (1994) Other southern and northern latitude studies compared diurnal and nocturnal densities of have reported shorebirdsforaging at night during Dunlin in September and October on a mud flat migratory periods. For example, Zwarts et al. and a Corophium-bed at the Danish Wadden (1990) compared densities of several shorebird Sea. Interestingly, diurnal densities exceeded speciesduring spring on a 6.9 ha area in the day nocturnal densities on the mud flat, but the re- and a 2.5 ha area at night on the Bane d’Arguin, verse occurred on the Corophium-bed. Mourit- SHOREBIRD FORAGING ECOLOGY 203

TABLE 4. Mean (+SE) abundance(prior to rank transformation)and total number of birds of eachspecies on plots.

Mean* k SE (total numb-+) Species Daylniaht Fall Winter Surinp.

AmericanAvocet” day 21.93 f 6.32 (1,404) 38.36 f 12.62 (989) 25.10 + 8.20 (561) night 24.55 + 8.14 (1,604) 49.83 f 14.93 (1,750) 11.75 f 10.22 (212) Marbled Godwit day 4.91 +- 2.36 (406) 24.73 f 8.44 (1,352) 20.78 + 12.92 (362) night 0.44 f 0.15 (37) 0.00 + 0.00 (0) 0.00 + 0.00 (0) Willet day 0.59 + 0.20 (50) 4.96 t 2.12 (277) 11.49 + 7.07 (184) night 0.05 f 0.03 (5) 0.03 * 0.03 (2) 0.00 + 0.00 (0) Dowitchers day 4.78 + 2.04 (403) 16.89 + 4.05 (967) 2.69 + 1.69 (100) night 1.89 + 0.91 (149) 0.14 + 0.14 (9) 0.00 + 0.00 (0) Black-belliedPlover day 0.08 * 0.05 (7) 0.18 + 0.09 (8) 0.04 * 0.04 (1) night 0.27 + 0.07 (22) 0.03 * 0.03 (1) 0.00 & 0.00 (0) SemipalmatedPlover day 0.23 i 0.11 (19) 0.04 * 0.04 (2) 0.00 +- 0.00 (0) night 0.28 ? 0.12 (24) 0.21 + 0.21 (10) 0.33 f 0.28 (9) Dunlind day 39.23 + 18.71 (1,308) 39.48 + 16.89 (2,268) 7.51 + 4.19 (289) night 13.96 ? 10.79 (454) 0.92 + 0.54 (49) 8.77 i 7.03 (227) Small sandpipers day 99.83 ? 25.94 (8,682) 44.62 * 15.25 (2,094) 20.89 + 17.36(976) night 9.81 +- 4.67 (876) 1.32 ? 0.92 (66) 2.79 f 2.34 (71) aFor eachspecies, averages of countsof foragingbirds on eachplot in day and night per monthly-intervals.Number of monthly-intervalmeans: 44 day and 44 night divided into 20 (fall), 16 (winter),and 8 (spring). bTotal numberof foragingbirds per scwce of variation. r Area data usedinstead of plot data. Number of monthly-intervalmeans: 43 day and 43 night divided into 20 (fall), 15 (winter),and 8 (spring). 4 Early fall censusesexcluded. Number of monthly-intervalmeans: 32 day and 32 night divided into 8 (fall), 16 (winter),and 8 (spring).

sen (1994) argued that greater prey availability Pienkowski et al. 1984, Evans 1987, Robert and at night on the Corophium-bed explained this McNeil 1989, Mouritsen 1994). The following finding. observations offer insight into the explanatory Despite methodological differences among power of these two hypothesesin Northern Cal- studies,one generality emergesfrom studiescon- ifornia. ducted at northern latitudes: nocturnal foraging First, shorebirds typically go through a com- is lessprevalent than diurnal foraging and it var- plete body molt during fall, whereas spring mi- ies throughout the annual cycle. However, there is less agreement regarding the season in which nocturnal foraging becomes more common. We TABLE 5. Taxonomiccomparisons of day and night suspect that differing methodologies as well as abundanceand frequencyof occurrenceof shorebirds on plots,presented as ratiosof day/nightvalues. differences among locations contribute to variation in these patterns. Ratio Day/night Mean Frequencyof CAUSES OF SEASONALVARIATION TaXa abundance’ occurrenceb

Two non-mutually exclusive hypotheses have Family Charadriidae: been proposed to explain seasonal variation in Semipalmated Plover 0.5 0.6 nocturnal foraging patterns of shorebirds (see Black-bellied Plover 0.7 0.8 McNeil 199 1). The “supplemental” hypothesis Family Recurvirostridae: postulates that shorebirds forage at night when American Avocet” 0.9 1.4 diurnal intake of prey is not sufficient to meet Family Scolopacidae: daily energetic requirements (e.g., mid-winter Dunlin 5.1 3.5 periods at northern latitudes, Heppleston 197 1). Dowitchers 9.7 2.9 By contrast, the “preferential” hypothesis posits Small sandpipers 12.0 2.5 that shorebirds forage at night to avoid diurnal Marbled Godwit 75.1 9.1 predators(Robert et al. 1989, Morrier and McNeil Willet 138.7 13.2 199 1) or to take advantage of increased prey ac- ’ Calculatedfrom monthly-interval meanspresented in Table 4. b Calculatedfrom frequencyof occurrencedata presentedin Table 2. tivity at night (Dugan 198 1, Pienkowski 1983b, EArea data usedinstead of plot data. 204 SARAH L. DODD AND MARK A. COLWELL

portunities (Townshend et al. 1984). Shorebird O.B- 1 use of coastal pasturesadjacent to North Hum- boldt Bay increases in winter and spring when n nighht precipitation increases availability of earth- worms (Lumbricidue) (Colwell and Dodd 1995). Consequently, the foraging successof fall migrants may depend more on intertidal habitat availability than wintering or spring-migrating birds, increasing the likelihood of fall migrants using exposed mud flats at night. Energy requirements associated with spring migration and molt are also high (Kersten and Piersma 1987, P. R. Evans 1988, Zwarts et al. 1990), so why did spring migrants of most species forage at night on Humboldt Bay mud flats relatively infrequently and in relatively lower num- L bers than fall? We suspectthat spring migrants IFall winter sprim3 obtained their energetic requirements without FIGURE 3. Seasonaland day/nightvariation in mean feeding much at night because: 1) lower low tides tide height(m) at low waterfrom 10 Jan 1992-10 Jan occurredfrequently in daylight; 2) day length was 1993 (Tidelogs1992, 1993). Fall, winter, and spring low tides greaterthan one half in daylight(n = 260) long, and consequently, intertidal areas re- and darkness(n = 245) wereincluded. mained exposed for long daylight periods; and 3) when mud flats were inundated during the day, coastal pastures offered alternative foraging op- portunities. grants replace only body feathers (Hale 1980, Alternatively, shorebirds may respond to sea- Prater 198 1, Evans and Pienkowski 1984). The sonal changes associated with nocturnal inver- pre-basic molt (fall) is energetically more costly tebrate prey availability (Pienkowski et al. 1984, than the pre-alternate molt (spring) (Hale 1980). Townshend et al. 1984). On mud flats of north- Fall-molting birds may forageat night more than western Humboldt Bay, Carrin (1973) observed spring-molting birds to meet higher energy re- little vertical movement of invertebrates be- quirements associated with the pre-basic molt tween day and night samples taken during two (sensu Evans and Pienkowski 1984, McNeil fall (July, August) tidal cycles. Carrin (1973) did 1991). Alternatively, a temporal expansion of not sample at night during other seasons.How- foraging by fall-molting birds may result from ever, he reported that diurnal invertebrates ex- reduced mobility (sensu Hale 1980, Evans and hibited lowest densities in winter and lowest bio- Pienkowski 1984). mass in spring; invertebrates exhibited moderate Second, flocks of fall-migrating shorebirds at to high biomass and densities in fall. These find- Humboldt Bay are composed of both juveniles ings suggestthat nocturnal foraging by shorebirds and adults (Gerstenberg 1972). Immature birds occurred more frequently when diurnal inver- are relatively inefficient foragers compared to tebrate prey were equal to or more available than adults (Puttick 1979,1984; Pienkowski and Evans other seasons.This observation contradicts find- 1984) and juveniles may need to forage at night ings from other studies conducted at northern in the fall to supplement diurnal intake of food. latitudes which suggestshorebirds forage more Finally, seasonal variation in environmental frequently at night in winter (Goss-Custard 1969) conditions contribute to variation in day/night when diurnal invertebrate prey availability is low patterns of shorebirds at North Humboldt Bay (Evans and Dugan 1984). (Dodd and Colwell, unpubl.). Fall tides tend to be lowest at night, whereas, spring and winter INTERSPECIFIC VARIATION tides tend to be lowest in the day (Fig. 3). Fall Day/night patterns of abundance and occurrence migrants may use mud flats at night becausethe of shorebirds varied considerably among taxa duration of mud flat exposure during the day is (Table 5). Members of the family Charadriidae relatively short, or alternatively, because lower and Recurvirostridae were relatively more noc- low nocturnal tides offer profitable foraging op- turnal than members of the family Scolopacidae. SHOREBIRD FORAGING ECOLOGY 205

Robert et al. (1989) also found interspecific vari- shorebirdsbecause they forage on relatively larg- ation associated with day/night patterns of wa- er prey or because relative intake rate is high terbirds, but with different resultsfrom our study. (Zwarts et al. 1990). For example, they reported that Black-bellied Plover foraged primarily during daylight, where- CONCLUSIONS as small sandpipers, small plovers, and medium- At North Humboldt Bay, the occurrence and sized shorebirds foraged mainly at night. abundance of nonbreeding shorebirds varied Different foragingmaneuvers are typically used considerably between day and night. Moreover, by different shorebird taxa. For example, plovers d&y/night patterns varied greatly among seasons searchfor prey visually (Pienkowski 198 1, 1983a, and across shorebird taxa. Shorebirds of most 1983b; McNeil and Robert 1988; Robert and common species foraged primarily in the day, McNeil 1989) whereas sandpipers locate prey although plovers and American Avocet foraged using visual and tactile maneuvers (Pienkowski as often at night as in the day. Nocturnal foraging 198 1, McNeil and Robert 1988, Robert and by most speciespredominated during fall migra- McNeil 1989). American Avocet use mainly tac- tion. Thus, the choice of a researcher’stemporal tile maneuvers (Evans and Harris 1994). Pre- scale or taxa may influence results considerably. sumably, visual foragersshould be more affected Elsewhere,(Dodd and Colwell, unpubl.), we show by darkness than tactile foragers (Dugan 198 1, that variation in environmental conditions such Robert et al. 1989, Zwarts et al. 1990, Martin as moonlight also influence day/night patterns of 1991). But at Humboldt Bay, plovers were the shorebirds. Explanations for seasonal variation most nocturnal taxa (Table 5). in day/night patterns of shorebirds at North According to Pienkowski (1983a), most plov- Humboldt Bay require further research. Studies ers evolved at southern latitudes under arid con- examining energeticsof shorebirdsand variation ditions where invertebrate prey are active pre- in availability of their invertebrate prey would dominately at night. Interestingly, a comparison best evaluate seasonal variation hypotheses. of rod/cone ratios from a Black-bellied Plover, a Short-billed Dowitcher, and a Greater Yellow- ACKNOWLEDGMENTS legsrevealed that the plover had the greatestrod/ We thank M. Dodd for field assistance;L. George for cone ratio and the highest density of rods (Rojas advice on nonparametricANOVA; and A. Dambmann de Azuaje et al. 1993). This suggeststhat the for graphicallyportraying the study figure. This study benefitted from discussionswith W. Kristan, D. Kris- Black-bellied Plover may be better adapted for tan, L. Shannon, and 0. Williams; and reviews by R. nocturnal vision than either of the Scolopacids. Golightly and D. Kitchen. We thank the City of Arcata Additionally, plovers have comparatively great- for allowing accessto public works property. A Frank er olfactory bulb development than diurnal M. Chapman Memorial Grant partially funded this members of Charadriida (Healy and Guilford study. 1990). By contrast, Scolopacids have relatively LITERATURE CITED smaller olfactory-bulb sizes than exclusively diurnal members of Scolopacida (Healy and Guil- BAKER,J. M. 1981. Winter feedingrates of Redshank (Tringa totanus) and Tumstone (Arenaria in- ford 1990). We are not aware of any literature terpres)on a rocky shore. Ibis 123:85-87. that documents the role of olfaction in foraging BARNARD,C. J., ANDD. B. A. THOMPSON.1985. Gulls shorebirds, but other bird taxa are known to use and plovers: the ecology and behavior of mixed- olfaction for locating prey (see review by Martin speciesfeeding groups. Columbia Univ. Press,New York. 199 1). Several authors have suggestedthat plov- BURGER,J. B. 1984. Abiotic factors affectingmigrant ers sometimes locate prey using acoustic cues shorebirds, p. l-72. In J. Burger and B. L. Olla (Pienkowski 1983a, Martin 1991) but most ev- [eds.], Shorebirds:migration and foraging behav- idence suggeststhat plovers continue to forage iour. Plenum Press,New York. visually at night (Pienkowski 1983a, McNeil and CARRIN,L. F. 1973. Availability of invertebrates as shorebird food on a Humboldt Bay mudflat. Robert 1988, Robert and McNeil 1989). M.Sc.thesis, Humboldt State Univ., Arcata, CA. Finally, among Scolopacids, smaller-bodied COLWELL,M. A., AND S. L. DODD. 1995. Waterbird specieswere relatively more nocturnal than larg- communities and habitat relationshipsin coastal er-bodied species(Table 5). Zwarts et al. (1990) pastures of Northern California. Conserv. Biol. 9:827-834. reported a similar pattern among Charadriidae, CONOVER,W. J. 1980. Practical nonparametric sta- Scolopacidae, and Haematopodidae taxa. Large tistics. John Wiley and Sons, New York. shorebirds may be less active at night than small COSTA,S. L., ANDJ. W. STORK. 1984. Humboldt Bay 206 SARAH L. DODD AND MARK A. COLWELL

prototype data collection. U.S. Army Corps of En- em Mendocino, and western Siskiyou counties, gineers,San FranciscoDistrict. ContractDACW07 California. Humboldt State Univ. Press, Arcata, 8 l-C-0029. CA. DODD, S. L. 1995. Seasonalvariation in diurnal and HARTWICK,E. B., AND W. BLAYLCXK. 1979. Winter nocturnal distributions of nonbreedingshorebirds ecologyof a Black Oystercatcherpopulation. Stud. at North Humboldt Bay, California. M.Sc.thesis, Avian Biol. 21207-215. Humboldt State Univ., Arcata, CA. HEALY, S., AND T. GUILFORD. 1990. Olfactory-bulb DOWDY, S., AND S. WEARDEX 1991. Statistics for size and noctumality in birds. Evolution 44:339- research.John Wiley and Sons, New York. 346. DUGAN,P. J. 1981. The importance of nocturnal for- HEPPLESTON,P. B. 1971. The feeding ecologyof Oys- aging in shorebirds: a consequenceof increased tercatchers(Haematopus ostralegusL.) in winter invertebrate prey activity, p. 251-260. In N. V. in northern Scotland. J. Anim. Ecol. 40:65 l-672. Jonesand W. J. Wolff [eds.],Feeding and survival HINTZE,J. L. 1992. Number CruncherStatistical Sys- strategiesof estuarine organisms. Plenum Press, tem (NCSS), Kaysville, UT. New York. HOCKEY,P. A. R. 1984. Behaviour patterns of non- EVANS,A. 1987. Relative availability of the prey of breeding African Black OystercatchersHaema- wading birds by day and by night. Mar. Ecol. Prog. topusmoquini at offshore islands, p. 707-727. In Ser. 37:103-107. J. A. Ledger ted.], Proc. Vth Pan-Afr. Ornithol. EVANS,P. R. 1976. Energy balance and optimal for- Congr. aging strategiesin shorebirds:some implications HURLBERT,S. H. 1984. Pseudoreplication and the for their distributions and movements in the non- design of ecologicalfield experiments. Ecol. Mon- breeding season.Ardea 64: 117-l 39. ogr. 54:187-211. EVANS,P. R. 1988. Predation of intertidal fauna by KERSTEN,M., AND T. PIERSMA. 1987. High levels of shorebirdsin relation to time of day, tide and year, energy expenditure in shorebirds; metabolic ad- p. 65-79. In G. Chelazzi and M. Vannini [eds.], aptations to an energeticallyexpensive way of life. Behavioral adaptation to intertidal life. Plenum Ardea 75:175-187. Press, New York. MANSEAU,M., AND J. FERRON. 1991. Activite ali- EVANS,P. R., ANDP. J. DUGAN. 1984. Coastal birds: mentaire nocturne des Becasseauxsemipalmes numbers in relation to food resources,p. 8-28. In (Calidris pusilla) lors dune halte migratoire dans P. R. Evans, J. D. Goss-Custard,and W. G. Hale la baie de Fundy. Can. J. Zool. 69:380-884. [eds.], Coastal waders and wildfowl in winter. MARON, J. L., AND .I. P. MYERS. 1985. Seasonal Cambridge Univ. Press, Cambridge, UK. changesin feeding success,activity patterns, and EVANS,P. R.,-AND M. W. PIENKOWSKI:1984. Popu- weightsof nonbreedingSanderlings (Calidris alba). lation dvnamics of shorebirds. v. 83-123. In J. Auk 102:580-586. Burgerand B. L. Olla [eds.], Shorebirds:breeding MARTIN,G. R. 1991. The sensorybases of nocturnal behaviour and populations. Plenum Press, New foragingin birds. Acta CongressusIntemationalis York. Ornithologici 20:1130-l 135. EVANS,T. J. 1988. Habitat use and behavioral ecol- MCNEIL, R. 1991. Noctumality in shorebirds. Acta ogy of American Avocets (Recurvirostraameri- CongressusIntemationalis Ornithologici 20:1098- cana) wintering at Humboldt Bay, California. 1104. M.Sc.thesis, Humboldt State Univ., Arcata, CA. MCNEIL, R., ANDM. ROBERT. 1988. Nocturnal feed- EVANS,T. J., AND S. W. HARRIS. 1994. Status and ing strategiesof some shorebird speciesin a trop- habitat use by American Avocets wintering at ical environment. Acta CongressusIntemationalis Humboldt Bay, California. Condor 96: 178-l 89. Ornithologici 19:2328-2336. FEARE,C. J. 1966. The winter feeding of the Purple MCNEIL, R., P. DRAPEAU,AND R. PIEROTTI. 1993. Sandpiper. British Birds 59: 165-l 79. Noctumality in colonial waterbirds: occurrence, GERSTEN~E~G,R. H. 1972. A study of shorebirds special adaptations, and suspectedbenefits. Cur- (Charadrii) in Humboldt Bav. California- 1968 rent Ornithol. 10:187-246. to 1969. M.Sc.thesis, Humboldt State Univ., Ar- MORRIER,A., AND R. MCNEIL. 1991. Time-activity cata, CA. budget of Wilson’s and Semipalmated Plovers in GERSTENBERG,R. H. 1979. Habitat utilization by a tropical environment. Wilson Bull. 103:598-620. wintering and migrating shorebirdson Humboldt MOURITSEN,K. N. 1992. Predatoravoidance in night- Bay, California. Stud. Avian Biol. 2:33-40. feeding Dunlins Calidris alpha: a matter of con- GOSS-CUSTARD,J. D. 1969. The wintering feeding cealment. Omis Stand. 23:195-198. ecologyof the Redshank(Tringa totanus).Ibis 111: MOURITSEN,K. N. 1994. Day and night feeding in 338-356. Dunlins Calidrisalpha: choiceof habitat, foraging GOSS-CUSTARD,J. D., R. A. JENYON,R. E. JONES,P. technique and prey. J. Avian Biol. 25:55-62. E. NEWBERY,AND, R. LE B. WILLIAMS. 1977. The PIENKOWS~,M. W. 1981. How foragingplovers cope ecology of the Wash. II. Seasonalvariation in the with environmental effectson invertebrate behav- feeding conditions of wading birds (Charadrii). J. iour and availability, p. 179-192. In N. V. Jones Appl. Ecol. 14:701-719. and W. J. Wolff [eds.], Feeding and survival strat- HALE, W. G. 1980. Waders. Collins, London, UK. egiesof estuarine organisms.Plenum Press, New HARRIS,S. W. 1991. Northwestern California birds: York. a guide to the status,distribution, and habitats of PIENKOWSKI,M. W. 1982. Diet and energy intake of the birds of Del Norte, Humboldt, Trinity, north- Grey and Ringed Plovers, Pluvialissquatarola and SHOREBIRD FORAGING ECOLOGY 201

Charadriushiaticula, in the non-breeding season. ditions and significanceof night feeding in shore- J. 2001. 19751 l-549. birds and other water birds in a tronical laeoon. PIENKOWSKI,M. W. 1983a. Changesin the foraging Auk 106:94-101. pattern of plovers in relation to environmental ROJAS DE &UAJE, L. M., S. TAI, ANDR. MCNEIL. 1993. factors. Anim. Behav. 31:244-264. Comparison of rod/cone ratio in three speciesof PIENKOWSKI,M. W. 1983b. Surfaceactivity of some shorebirds having different nocturnal foraging intertidal invertebrates in relation to temperature strategies.Auk 110:141-145. and the foragingbehaviour oftheir shorebirdpred- ROMPR&,G., ANDR. MCNEIL. 1994. Seasonalchanges ators. Mar. Ecol. Prog. Ser. 11:141-150. in day and night foragingof Willets in northeastern PIENKOWSKI,M. W., AND P. R. EVANS. 1984. Migra- Venezuela. Condor 96:734-738. tory behavior of shorebirds in the western Pale- STENZEL,L. E., H. R. HUBER,AND G. W. PAGE. 1976. arctic, p. 73-123. In J. Burgerand B. L. Olla [eds.], Feeding behavior and diet of the Long-billed Cur- Shorebirds: migration and foraging behaviour. lew and Willet. Wilson Bull. 88:314-332. Plenum Press, New York. SWENNEN,C. 1990. Oystercatchersfeeding on giant PIENKOWSKI,M. W., P. N. FERNS,N. C. DAVIDSON, bloody cockleson the Bane d’Arguin, Mauritania. AND D. H. WORRALL. 1984. Balancingthe bud- Ardea 78:53-62. get: Measuring the energyintake and requirements TIDELOG. 1992. Northern California, Pacific Publish- of shorebirdsin the field, p. 29-56. In P. R. Evans, ers, Bolinas, CA. J. D. Goss-Custard,and W. G. Hale [eds.],Coastal TIDELOG. 1993. Northern California, Pacific Publish- waders and wildfowl in winter. Cambridge Univ. ers, Bolinas, CA. Press, Cambridge, UK. TOWNSHEND,D. J., P. J. DUGAN, AND M. W. PI- PRATER,A. J. 1981. A review of the patterns of pri- ENKOWSKI.1984. The unsociableplover-use of mary moult in Palaearctic waders (Charadrii), p. intertidal areas by Grey Plovers, p. 140-l 59. In 393-409. In J. Cooper [ed.], Proceedingsof the P. R. Evans, J. D. Goss-Custard,and W. G. Hale symposium on birds of the sea and shore, 1979. [eds.], Coastal waders and wildfowl in winter. Cape Town: African Seabird Group. Cambridge Univ. Press, Cambridge, UK. PUTrICK, G. M. 1979. Foraging behaviour and ac- TREE, A. J. 1979. Biologyofthe Greenshankin south- tivity budgets of Curlew Sandpipers. Ardea 67: em Africa. Ostrich 50:240-25 1. 111-122. ZWARTS,L. 1990. Increased prey availability drives PUTTICK,G. M. 1984. Foragingand activity patterns premigration hyperphagia in Whimbrels and al- in wintering shorebirds, p. 203-231. In J. Burger lows them to leave the Baned ’ Arguin, Mauritania, and B. L. Olla [eds.], Shorebirds: migration and in time. Ardea 78:279-300. foraging behaviour. Plenum Press, New York. ZWARTS,L., A. BLOMERT,AND R. HUPKES. 1990. In- ROBERT,M., AND R. MCNEIL. 1989. Comparative creaseof feedingtime in wadersperparing for spring day and night feeding strategiesof shorebird spe- migration from the Bane d’ Arguin, Mauritania. cies in a tropical environment. Ibis 13 1:69-79. Ardea 78:237-256. ROBERT,M., R. MCNEIL, AND A. LEDUC. 1989. Con-