Estuaries

VoL 22, No. 1, p. 47-54

March 1999

Predation on the Southwestern Atlantic Fiddler Crab

(Uca uruguayensis) by Migratory Shorebirds (Pluvialis dominica, P. squatarola, Arenaria interpres, and Numenius phaeopus) OSCAR O . IRIBARNE ~

MAm~L'~OM. MARTINEZ

Departamento de Biologfa (FCd~'yN) Universidad Nacional de Mar del Plata CC 573 Correo Central (7600) Mar del Plata, Argentina ABSTRACT: In Bahia Samborombon (SW Atlantic; 35~ 56"45'-57023'W) migratory American golden plover (Pluvialis dominica), black-billed plover (P. squatar0/a), ruddy turnstone (Arenar/a interpres), and whimbrel (Numenius phaeopus) forage on fiddler crabs (Uca uruguayens/s). These shorebirds have dislinct patterns of feeding behavior and capture different proportions of each sex and size. P. domin/ca made short runs, capturing mostly juvenile crabs, which are eaten whole. P. squatarola use a similar feeding strategy but capture females primarily. A. interpres walks continuously, capturing almost exclusively large male crabs (71%). N. phaeopus walk in the Uca patch and probe burrows by inserting the biB; they primarily capture females. The feeding rate of A. interpres is higher than that of P. dominica, P. squatarola, and N. phaeopus. Handling time shown by P. dominica and A. interpres was greater than the other two species for all categories o f crabs. In all cases, handling time of male crabs was greater than those of either females or juveniles. All the evidences suggest that [7. uruguayensis is an important f o o d source for all these species and should be accounted in any conservation endeavor.

ited by several resident shorebirds (i.e., Haematopus palliatus, Himantapus melanurus, and Charadrius col-

Introduction

Fiddler crabs provide food for a number of migratory shorebirds, including Arenaria interpres, Ca-

/aris) and is an important stopover and wintering site for North American migratory shorebirds such as Pluvialis dominica, P. squatarola, C. semipalmatus,

toptropharus semipalmatus, Charadrius collaris, C. hiaticula, C. leschenaultii, C wilsonia, Numenius phaeopus, Pluvialis squatarola, Tringa cinereus, T. macularia, and T. totanus. (Morrier and McNeil 1991,

Tringa melanoleuca, T flavipes, Arenaria interpres, Calidris alba, C. canutus, C. melanotos, C. bairdii, C. fuscicollis, Micropalama himantopus, Numenius phaeopus, Limosa haemastica, and Phalaropus tricolor (e.g., Dab-

Spaans 1979, Strauch and Abele 1979, Summers 1980, Thibault and McNeill 1995, Zwarts 1985). Some of these shorebirds overlap in their distribution with the southwestern Atlantic fiddler crab, Uca uruguayensis. This small crab species (up to 20 mm carapace width) is found from southern Brazil (33~ to the northern coast of Argentina (38~ Boschi 1964) inhabiting the upper part of estuarine intertidal areas (Boschi 1964) in wet muddy substrate (Spivak et al. 1991). Individuals dig a 1520 cm deep burrow and spend most of the low tide period near the burrow entrance eating or, in the case of males, waving their large chelae (personal observation) The southernmost permanent population inhabits Bahia S a m b o r o m b o n (35030 ' 36~ 56~176 mouth of Rio de la Plata, Argentina) where it is the dominant intertidal species (Boschi 1964) reaching densities of up to 140 crab m 2 (personal observation). The bay is inhab-

bene 1920; Gibson 1920; Olrog 1967). Some of these shorebirds prey on other fiddler crab species along their migratory pathway. Pluvialis squatarola feeds on Uca tangeri in Guinea Bissau (Zwarts 1985). Ruddy turnstone feeds on Uca lactea in the Seychelles (Summers 1980) and on U. tangeri in Guinea Bissau (Zwarts 1985). Pluvialis squatarola, A. interpres, Numenius phaeopus, and P. dominica were observed feeding on Uca in Bahia Samborombon (personal observation). However, despite the importance of crabs in this area (Boschi 1964), the shorebird-crab interaction has not been studied. Here we document predation on Uca uruguayensis in Bahia S a m b o r o m b o n by several migratory shorebirds species and describe their feeding behavior, prey selection, and intensity of predation. Material and Methods The study was conducted from December 1994 to April 1995 and from February to March 1996

1 Corresponding author; fax: 54 023 753150; e-mail: [email protected]. 9 1999 Estuarine Research Federation

47

48

o . o . Iribame and M. M. Martinez

Atlantic Ocean

o

50 Km

Fig. 1. L o c a d o n of the study site, Arroyo San Clemente, in the Bahia S a m b o r o m b o n (BS). Horizontal line shows the southern limit of distribution of the fiddler crab Uca urugua~,ensis.

n e a r the m o u t h o f o n e o f the easternmost tidal c h a n n e l s , A r r o y o San C l e m e n t e (36~ 56 ~ 45'W), of Bahia S a m b o r o m b o n (an approximately 100 km long coastal basin within the Rio de la Plata estuary, Argentina) (Fig. 1). T h e area is characterized by large, dense populations o f the fiddler crab Uca uruguayensis and the burrowing crab Chasmagnathus granulata (Boschi 1964). T h e littoral zone extends into an extensive, Spartina-dominated salt marsh (Cabrera and Zardini 1978). T h e fiddler crab p o p u l a t i o n inhabits the u p p e r o p e n intertidal n e a r the Spartina densiflora fringe, with few individuals living in the vegetated areas. A beach (500 m long and 60 m wide at low tide) with large patches o f Uca was selected for evaluating bird activity in relation to f o o d items. To evaluate the presence of f o o d sources in the Uca beds that may be used by shorebirds, 10 r a n d o m l y located cores (0.03 m 2 area; 0.3 m deep) were taken in the Uca b e d in b o t h study seasons (April 1995; March 1996) and sieved t h r o u g h a 0.5-mm screen. Individuals o f all species were identified and counted. Observations o f bird behavior were m a d e during low tide f r o m m i d - m o r n i n g to late "afternoon on two different patches (20 m x 5 m and 80 m • 20 m) located approximately 500 m apart. Bird species were identified using Narosky and Yzurieta (1987). Two individuals, hiding in different sites (usually at less than 40 m f r o m the bird activity arena) observed bird behavior using 10 • 50 binoculars and 18-36 • spotting scopes and r e c o r d e d data with a portable tape recorder. Each bird was

observed for a m a x i m u m o f 15 min. T h e following categories o f behavior were recorded: (1) walk: forward locomotion, (2) pause: standing motionless, (3) peck: striking at the substrate with the bill, and (4) prey capture: prey c a p t u r e d f r o m the m u d or f r o m a crab b u r r o w followed by a swallowing action. H a n d l i n g time is the time elapsing f r o m the prey being pecke d until the prey is swallowed. T h e following categories o f prey items were identified: (a) small unidentified items, (b) male Uca crabs (individuals with the notorious red chelae), and (c) female Uca (large individuals without the larger chelae). Since previous samples o f the Uca population (1994, n = 420, 1995 = 372) had few males (1994: 0.5%, 1995: 0.8%) lacking large chelae, we assumed that large crabs without the notorious red chelae were females. T h e null hypothesis o f n o difference in the overall d e g r e e o f activity in relation to success or failure in the capture o f juvenile, male, or female Uca was evaluated by a two-way ANOVA (Zar 1984). T h e null hypothesis o f n o difference in handling time in relation to shorebird species and prey items was evaluated by a two-way ANOVA (Zar 1984). During each observation p e r i o d we also recorded intraspecific and interspecific aggressive interaction between individual shorebirds: displacem e n t (when an i n t r u d e r moved into an area, supplanting a n o t h e r ) , chase (when o n e individuals actively pursues a n o t h e r ) , or fight (when an i n t r u d e r actively contacted a n o t h e r bird). To evaluate if shorebirds prey differentially o n the Uca population, crab density and size frequency distribution was estimated f r o m 24 (first period) and 20 (second period) r a n d o m l y located sampling units (0.5 m 2 area, 0.5 m deep, sieved t h r o u g h a 1-mm screen) in the shorebird feeding site. Sampled individuals were categorized as juvenile (when we were not able to sex t h e m ) , female, or male. Chesson's alpha (Chesson 1978), an index o f food selection, was used to evaluate prefe r e n c e for these food items. T h e i n d e x measures an invariant d e g r e e o f p r e f e r e n c e on tile part o f the p r e d a t o r (Pearre 1982). T h e i n d e x is $

Chesson's alpha = ( r i / p i ) / ~ i

(ri/Pi)

where ri and p~ are the p r o p o r t i o n o f prey item i in the diet and the environment, respectively (Strauss 1979). To evaluate the null hypothesis o f n o difference in the risk o f p r e d a t i o n o f Uca individuals due to sex and size, o n e observer walked linearly in a pred e t e r m i n e d 2-m-wide track collecting crabs that failed to hide. T h e intention was to simulate a disturbance that may resemble the o n e created by

Predation on Fiddler Crabs by Shorebirds

walking shorebirds, particularly Arenaria interpres (see Results). Crabs were sexed and m e a s u r e d and the null hypothesis was evaluated by c o m p a r i n g the p r o p o r t i o n s o f males and females that failed to hide (Zar 1984). Previous observations had indicated that crabs may fail to hide by being stacked in a b u r r o w entrance, or f r o m being chased f r o m a b u r r o w by its inhabitant. We quantified the occ u r r e n c e o f these two patterns and evaluated the null hypothesis of n o difference in their f r e q u e n c y by c o m p a r i n g the p r o p o r t i o n o f o c c u r r e n c e of each event (Zar 1984). Previous o b s e r v a t i o n s (personal o b s e r v a t i o n ) showed that when Arenaria interpres prey on males (n = 123 attacks), they rarely (n = 2) eat the chelae. T h e null hypothesis o f n o selection by shorebirds on the size o f males was evaluated by collecting the male's chelae remains after a predation event. Crab size was back-calculated using a body size-male chelae size relationship obtained from a sample o f 100 individuals. Crab body size was expressed as m a x i m u m carapace width (CW, measured between anterolateral margins) and was used as an i n d e p e n d e n t variable. Major cheliped size was expressed as p r o p o d u s length (PL). Both dimensions were m e a s u r e d to the nearest 0.1 mm. A regression between CW and PL (best met the statistical assumptions) was p e r f o r m e d and the estimated parameters were used to back-calculate the size of crabs. T h e null hypothesis was then tested by c o m p a r i n g the size f r e q u e n c y distribution o f crabs estimated f r o m chelae remains against the population's size f r e q u e n c y distribution using the Kolmogorov-Smirnov test (Zar 1984). T h e null hypothesis o f n o difference between the size of wandering males and the size o f p r e y e d crabs (estim a t e d f r o m their chelae remains) was also evalua t e d u s i n g t h e K o l m o g o r o v - S m i r n o v test (Zar 1984). Results

Fauna density in the Uca b e d was low in both years. Fewer than 2 ind m -2 o f the polychaetes Neanthes succinea (first year: :~ = 1.3 ind m-Z; SD = 1.4; n = 10; second year: ~ = 2.1 ind m -2, SD -1.8; n = 10) and Laeonereis acuta (first year: ~ = 1.1 ind m 2; SD = 1.5; n = 10; s e c o n d year: s = 1.8 ind m 2; SD = 1.2; n = 10) were present. Crab density was 97 crab m- '-' (SD = 23 crab m-Z; n = 15; 42% juveniles, 18% females, 39% males) during the first year and 81 crab m 2 (SD = 21 crab m-2; n = 14; 49% juveniles, 14% females, 37% males) d u r i n g the second year, and sex ratio was always male biased ( 1 9 9 4 : 0 . 7 males: 0.3 females; 1995:0.7 males : 0.3 females). Several bird species were observed feeding o n the fiddler crab Uca uruguayensir. American gold-

49

en-plover (Pluvialis dominica), black-bellied plover (Pluvialis squatarola) , r u d d y t u r n s t o n e ( Arenaria interpres), whimbrels ( Numenius phaeopus), gull-billed tern (Sterna nilotica; which dive f r o m the air to catch crabs), b r o w n - h o o d e d gull (Larus maculipennis), American stilt (Himantopus melanurus), great kiskadee (Pitangus sulphuratug which hover over the Uca patch using a hawk-like t e c h n i q u e and occasionally p e r c h or stand nearby waiting for crabs to surface), and two-banded plover (Charadrius falklandicus). T h e species studied here, which were the ones most frequently observed feeding in the study site, were American golden plover, black-bellied plover, r u d d y turnstone, and whimbrels. Behavior and feeding p e r f o r m a n c e s o f these species were different. AMERICAN GOLI)EN PLOVER (PLUVIALISDOMINICA)

T h e behavior o f American g o l d e n plover in the study area was observed for 20 h. During focal sampling o f 27 individuals for a total time of 149 minutes, most o f t h e m (90%) were solitary. All these observations were obtained d u r i n g the first sampling season. These plovers feed in the Uca bed during the whole tidal cycle, even during low tide when the soft bottoms are uncovered, pursuing crabs most o f the time. T h e i r feeding m e t h o d is based on short runs and stops for inspection of the area. This is similar to the " r u n and pick" or plover strategy cited by Pienkowski (1980) but begins with a standing position inspection o f the area (stop-run-stop when no success and stop-run-pick when prey were captured). Sometimes, they also run, h e r d i n g crabs, and capture those that fail to find their burrow (similarly to the r u d d y t u r n s t o n e behavior). If there is success, crabs are eaten entirely. T h e selectivity index (Fig. 2C) showed that unidentified items (which we believe to be juveniles) were the p r e f e r r e d prey, while females and males were attacked at m u c h lower p r o p o r t i o n than their availablity. American g o l d e n plover have a cyclical use o f the crab bed and the nearby intertidal, spending a variable time (range: 3-9 rain) in the Uca bed, moving to the lower intertidal, and t h e n r e t u r n i n g to the Uca b e d "after a similar a m o u n t o f time (range: 6-10 rain). T h e y shared the same patch with r u d d y turnstone (Arenana interpres) without aggressive interactions. BLACK-BELLIEDPLOVER (PLUVIAl,ISSQUATAROLA) This species was observed feeding on Uca patches d u r i n g b o t h periods, but they were m o r e active during the second sampling period. T h e behavior of black-bellied plovers was observed for 28 h. All information r e p o r t e d h e r e are f r o m the second year. Focal sampling o f 20 individuals was perf o r m e d for a total time o f 136 rain; all individuals

50

=E

O.O. Iribame and M. M. Martinez

'IA

P. domlnl~ ~z~ p. iquatlcokl A. lnt~Drlxu

E~3 N . ~

FEMALE8 6O

| LU

o

Z

|

UNKNOWN

IB {

I I

;:i.i:i;:i:;:; ~

oL

r

o iii

O;

-

MALES

FEMALES

UNKNOWN

P R E Y ITEM Fig. 2. Number of prey captured per minute (A), prcy handling time (B), and a selectivity index (C) of shorebirds preying on juvenile, male, and female Uca uruguayensis.

were solitary. T h e i r feeding behavior was similar to the American g o l d e n plover but h a d l o n g e r standing times b e f o r e starting to run. Several times they waited until crabs surfaced, capturing crabs very close to their standing position (n = 15). Crabs were always eaten entirely. T h e selectivity i n d e x (Fig. 2C) showed that females were the p r e f e r r e d prey, while unidentified items and males were attacked at m u c h lower p r o p o r t i o n than their availablity. This species was the only species that def e n d e d its territories by singing displays a n d chasing o t h e r conspecifics or American golden plovers when their feeding area was a p p r o a c h e d . Blackbellied plover were observed sharing a feeding patch only with whimbrel (Numenius phaeopus) and r u d d y turnstones ( Arenmia interpres) . RUDDY TURNSTONF (ARENARIA INTERPRES)

This species was observed feeding on the Uca patch only d u r i n g the first study period. During the second year they were feeding in the lower intertidal area, mostly eating small animals h i d d e n u n d e r n e a t h gastropods ovisacs. Behavior o f r u d d y turnstones was observed in the study area for a total time o f 22 h. Focal sampling o f 22 birds was p e r f o r m e d for a total time o f 123 rain. Most are

solitary feeders, the most c o m m o n a b u n d a n c e in the Uca patch was two individuals (80%); rarely one bird (18%) or 3-5 birds (2%) were counted. W h e n there is m o r e than o n e individual feeding in the same patch, the birds frequently used the opposite parts o f the patch, occasionally walking in parallel, but always being apart. Birds always walk, some turning over bivalve shells, gastropods ovisacs, and o t h e r invertebrate shelters. T h e y were active d u r i n g ttie low tide period (we lack n o c t u r n a l observations), spending most o f their time searching for crabs (99%), with only short stops, usually at the water's edge. T h e i r constant m o v e m e n t over the Uca bed p r o d u c e d a running-away behavior in the crabs. During low tide, male crabs are n e a r their b u r r o w entrances trying to attract females, but due to the presence o f birds, some of the males fail to find their own b u r r o w when disturbed and are c a p t u r e d by turnstones. T u r n s t o n e behavior differs f r o m plovers: they work m u c h h a r d e r looking for crabs in their burrows a n d trying to pull crabs out o f their burrows, in some cases spending minutes with a large p o r t i o n o f their bill (and even head) dug into the b u r r o w entrance. As a result, after crabs are attacked, most b u r r o w entrances are left with a larger opening. T h e larger chelae o f males were rarely eaten (1%) and were left in the attack site. These remains were used to evaluate possible selection o f males according to size (see below). T h e selectivity index (Fig. 1C) showed that males were the p r e f e r r e d prey and that females a n d unidentified items were attacked at m u c h lower p r o p o r t i o n than their availablity. T h e r e were some aggressive interactions (n = 4) between birds, mosdy when a crab was captured. WHIMBREL ( NUMENIUSHIAE~OPUS)

This species was only observed feeding o n the Uca patch during the second sampling period, and most days only o n e individual was seen feeding o n the whole beach. However, the small n u m b e r o f whimbrel always r e p o r t e d for this area (Blanco et al. 1988) makes it difficult to assure that we observed different individuals. T h e behavior o f w h i m brels was observed in the study area for a total time of 15 h. Focal sampling o f nine birds was perf o r m e d for a total time o f 86 min; all were solitary. T h e i r behavior differs f r o m the o t h e r shorebirds since they actively search for crabs walking slowly and p r o b i n g crab burrows by inserting their bill. Crabs were c a p t u r e d inside the burrow. Forty percent of the times that males were captured the larger chelae was not eaten. In only 2% o f the cases were individuals c a p t u r e d outside. T h e selectivity index (Fig. 2C) showed that females were the pref e r r e d prey, and males and unidentified items were

51

Predation on Fiddler Crabs by Shorabirds

attacked at m u c h lower p r o p o r t i o n than their availablity. During the first sampling season the feeding success o f Pluvialis dominica was partially d e p e n d e n t o n the interaction with the most aggressive P. squatarola, which usually chase P. dominica (80% o f interactions). However, the intensity o f aggressive interactions was site-dependent: all aggressive interactions between Pluvialis species were r e c o r d e d in a small patch, while n o aggressive interaction was r e c o r d e d in the larger patch. T h e feeding o f Pluvialis dominica in the small patch was always interr u p t e d by the arrival o f even o n e o f P. squatarola. Since shorebird species feeding in our study area were different d u r i n g both study periods, the statistical c o m p a r i s o n o f prey capture per unit o f time and handling time was p e r f o r m e d by c o m p a r i n g the species that were feeding d u r i n g the same period (i.e., Pluvialis dominica versus Arenaria interpres, and Pluvialis squatarola versus Numenius phaeopus). Although sex-ratio and size f r e q u e n c y distribution o f crabs were similar between years, these comparisons avoid possible differences between years in availability or behavior o f crabs. For the first study year the overall d e g r e e o f activity o f P. dominica a n d A. interpres, in relation to failure or success in capturing unidentified items, female crabs, or male crabs was n o t significantly different (two-way ANOVA: species effect F = 0.014; df = 1; p > 0.05; Fig. 1A). T h e two species display significant differences when the a m o u n t o f each activity per unit o f time is c o m p a r e d (activity effect F = 7.21; df = 1; p < 0.05; Fig. 1A). Pluvialis dominica eat mainly small individuals, and a relatively low p r o p o r t i o n o f large males. Arenaria interpres feed almost exclusively on male crabs (Fig. 1A). T h e n u m b e r o f failures per unit o f time was m u c h lower for r u d d y turnstones (Fig. 1A), probably due to their feeding behavior and their robust bill. H a n d l i n g time was not significantly different between shorebirds (twoway ANOVA: species effect F = 3.25; d f = 1; p < 0.05), a l t h o u g h r u d d y turnstones t e n d to have a longer handling time. However, b o t h shorebirds showed longer handling time for males than for females or unidentified items (Fig. 1B; crab stage effect F = 23.17; d f = 2; p > 0.05). For the second study year the overall d e g r e e o f activity of Pluvialis squatarola and Numenius phaeopus, in relation to failure or success in capturing, juvenile, female, or male crabs was significantly diff e r e n t (two-way ANOVA: species effect F = 4.04; d f = 1; p < 0.05; Fig. 1A). However, the two species display significant differences when the a m o u n t of each activity per unit o f time is c o m p a r e d (activity effect F -- 8.24; d f = 1; p < 0.05; Fig. 1A). P. dominica eat mainly small individuals and a relatively low p r o p o r t i o n o f large males. Arenana interpres

40L 9 PREYED

.........

]I I

30

ILl (3

I

*

o POPULATION

. . . . .

1 ..~"

1

3

5

-

7

9

11

13

15

CARAPACE WIDTH (mm) Fig. 3. Comparison of" the size frequency distribution of the entire population of male fiddler crabs (black squares) with the wandering males (cross), and tile preyed males (empty diamonds) estimated from chelae remains. Tile difference between wandering males and preyed males is not statistically significant (KS test, p > 0.05).

feeds almost exclusively on male crabs (Fig. 1A). T h e n u m b e r o f failures per unit o f time was m u c h lower for r u d d y turnstones (Fig. 1A), probably due to their feeding behavior. H a n d l i n g time was significantly different between species (two-way ANOVA: species effect F = 69.29; df = 1; p < 0.05) and for crab stages (Fig. 1B; crab stage effect F = 17.12; d f = 2; p > 0.05). During both study periods male Uca crabs spend most o f the time waving n e a r the e n t r a n c e of their b u r r o w while females w a n d e r inspecting burrows. Crabs react fast to the presence o f shorebirds by r u n n i n g and hiding, resulting in a 2-3 m " e m p t y " circle a r o u n d the shorebird. W h e n disturbed (as h a p p e n s with r u d d y turnstones), some crabs stop in or get stacked in the b u r r o w e n t r a n c e (x = 0.3 crab m-2; SD = 0.1; n = 70). No female was ever observed in this position. O u r walking e x p e r i m e n t showed that 80% o f the crabs (n = 56) were individuals that e n t e r a burrow, back out fast, and c o n t i n u e testing o t h e r burrows. T h e o t h e r 20% were individuals that got stacked in a b u r r o w entrance. T h e size f r e q u e n c y distribution of these crabs is significantly different f r o m the population's size f r e q u e n c y distribution (Fig. 3; Kolmogorov-Smirnov test; p < 0.05). Size o f crabs f o u n d wandering or stacked in a b u r r o w e n t r a n c e were the largest individuals in the p o p u l a t i o n (Fig. 3). T h e regression between crab carapace width (CW) and p r o p o d u s length (PL) was CW = 4.5 + 0.38 PL; r 2 = 0.87, n = 100

52

O.O. Iribame and M. M. Martinez

Using this e q u a t i o n we calculated the size o f crabs f r o m chelae remains, which were also f r o m the largest individuals in the p o p u l a t i o n (Fig. 3). T h e i r size was not statistically different f r o m the size o f wandering crabs f o u n d in o u r "walking" experim e n t (Kolmogorov-Smirnov test n = 100; p >

o.o5). Discussion Although the Uca uruguayensis p o p u l a t i o n in Bahia S a m b o r o m b o n has n o t b e e n quantified, it is a large resource spread over m o r e than 100 km o f coastal marshes with an intricate system o f tidal creeks and channels. Four shorebirds were conspicuously observed feeding on these crabs: blackbellied plover, American golden plover, r u d d y turnstone, and whimbrels. These species are comm o n immigrants to the S a m b o r o m b o n area during the s o u t h e r n spring-autumn season ( a b u n d a n c e in the P u n t a Rasa area: black-bellied plover: -< 10 individuals, American g o l d e n plover: < 120 individuals, r u d d y turnstone: < 29 individuals, and whimbrels: -< 16 individuals; Blanco et al. 1988). T h e Bahia S a m b o r o m b o n is o n e o f the southe r n m o s t sites where r u d d y t u r n s t o n e are r e p o r t e d (Vila et al. 1994). T h e activity o f Arenaria interpres in o u r study site was restricted to Uca beds during the first year. But they were very rarely seen feeding o n Uca d u r i n g the following year, although they were feeding in a nearby area. T h e i r feeding strategy is different than the sandpiper or a plover strategy (see B u r t o n 1974; Pienkowski 1980). T h e i r constant m o v e m e n t over the Uca bed elicited the crab's running-away behavior. Although male crabs are usually n e a r their b u r r o w attempting to attract females, in a high density p o p u l a t i o n and u n d e r external disturbance, some of t h e m fail to find their own burrow. This is exploited by r u d d y turnstones. W h e n disturbed, males a t t e m p t to e n t e r o t h e r burrows but their risk o f mortality is actually increased. Males have two problems: (a) larger males are less likely to find a b u r r o w with an entrance wide e n o u g h to fit their size; and (b) even when burrows are o f p r o p e r size, interactions with b u r r o w inhabitants generally prevent males f r o m entering. T h e s e two problems are highly depend e n t o n the p o p u l a t i o n sex structure. In a population with a highly skewed sex ratio (70% males), such as the o n e in o u r study site, male-male competition should be very intense (Emlen and Oring 1977). Thus, males may go f u r t h e r away f r o m their b u r r o w to attract females, increasing the chances o f missing their own b u r r o w if a disturbance occurs, and thus risking p r e d a t i o n by shorebirds. T h e behavior we observed in the S a m b o r o m b o m area differs f r o m observations in o t h e r areas. T h e behavior observed o n the beaches o f Mauritania,

where turnstone are kleptoparasites a n d scavengers, taking Uca remains left by whimbrels (Ens et al. 1990), has n o t b e e n observed in o u r study site. F u r t h e r m o r e , they discard the male's large chelae, a part that they scavenged in Guinea-Bissau (see Ens et al. 1990). However, the different behavior shown in both years also suggests that their behavior is highly plastic. T h e foraging strategy o f both Pluvialis species c o r r e s p o n d e d to the exclusively visual strategy c o m m o n for plovers (Burton 1974; Pienkowski 1980). Both species showed a slightly different strategy o f prey capture, with b o t h strategies having some benefits and disadvantages d e p e n d i n g on the activity o f prey items (Pienkowski 1980). T h e i r feeding behavior results in p r e d a t i o n o f a larger n u m b e r o f females and small items that we believe to be juveniles, with only very small p r o p o r t i o n of males in their diet. T h e feeding efficiency o f golden plovers is lower than r u d d y t u r n s t o n e a n d their diet are mostly c o m p o s e d o f small individuals. T h e y are able to capture crabs by making quick dashes. In some cases, crabs are c a p t u r e d by stopping and waiting in an area devoid o f crabs for the Uca to r e t u r n to the surface. T h e same behavior has b e e n r e p o r t e d for shorebirds preying o n Uca tangeri in Guinea-Bissau (Zwarts 1985). Thus, the effect on the crab p o p u l a t i o n structure should be different than the o n e r e p o r t e d for turnstones. T h e behavior o f black-billed plover, Pluvialis squatarola, in o u r study area differs f r o m their behavior seen in Mauritania (Zwarts and Dirksen 1990), where they scavenge Uca remains left by whimbrels or fish left by h u m a n s (Ens et al. 1990). O f the species identified in o u r study, Pluvialis dominica are the most plastic foragers, since they feed o n organisms f r o m intertidal worms to insects in grazing farmlands (Myers and Myers 1979). P. squatar0/a, although very similar in shape to P. dominica, is a less versatile species, feeding only in intertidal marine areas (Hayman et al. 1986). Long-billed birds, including the white ibis (Eudocimus albus), willet (Cat0ptr0phorus semipalmatus), and whimbrel (Numenius phaeopus), take crabs from the marsh surface, as well as f r o m intertidal burrows, which they p r o b e with their bill (Kushlan 1979; Pettit and Bildstein 1987). T h e shore crab Carcinus maenas is a main prey in late s m n m e r when whimbrels pass t h r o u g h the W a d d e n Sea (Zwarts 1990), where fiddler crabs (Uca spp.) and o t h e r crab species are the main prey in tropical a n d subtropical wintering areas (e.g., Summers 1980; Zwarts 1985; Zwarts and Dirksen 1990). Zwarts (1990) describe four feeding behaviors for whimbrels observed feeding on Uca tangeri in Mauritania: (1) while walking, they scanned the surface and m a d e a short dash directed at crabs that were

Predation on Fiddler Crabs by Shorebirds

far from their burrows; (2) waiting motionless near burrows until a crab emerged; (3) capturing crabs that were far from the burrow zone; and (4) scave n g i n g remnants of crabs. Whimbrels feeding on Uca uruguayensis show a different behavior, they insert the bill into the crab burrows and catch the prey inside the burrow. Only juvenile crabs are captured o n the surface, but whimbrels always maintain their slow walking speed, scanning the surface, and going from o n e burrow to the nearest neighbor probing with their l o n g bill like (1) above. This restricted behavior may be due in part to the fact that U. uruguayensis does n o t a b a n d o n the burrow z o n e like U. tangeri. Similarly to white ibises, whimbrels capturing crabs from burrows, catch female crabs twice as often as males crabs (Bildstein et al. 1989). This pattern may be due to gender differences in behaviors within burrows. Frix et al. (1991) showed t h a t . u n d e r a predation threat by long-billed mimics, female Uca pugnax and U. pugilator retract m u c h further into the burrow than males. Although the Uca population is a large resource, interactions and territoriality in the studied shorebirds are expected due to the nature of resource exploitation. Walking redshanks, 7~inga totanus, reduced the n u m b e r of Corophium at the surface in the immediate vicinity, which promotes a tendency to avoid close conspecific associations (Goss-Custard 1970). Similarly, when SW Atlantic fiddler crabs are disturbed by shorebirds, individuals on the surface move down their burrows, creating a circle (1-3 m radius) devoid of Uca on the surface around a feeding bird. This pattern should generate interference competition between individual birds. Species observed foraging on Uca were mostly solitary or at very low dens!ty, and most individuals appear to have high affinity with their feeding place, in some cases displaying territorial behavior. The most territorial species in our study were the Pluvialis species; they realized song displays when a conspecific individual approached. Overall our research shows that several migratory shorebirds intensely use the southwest Atlantic fiddler crab as a f o o d resource. Given that the largest crab populations in Bahia S a m b o r o m b o n are located in areas with tourist activities, we suggest that this crab species should be considered in any conservation endeavor as an important resource for shorebird species. ACKNOWI .EI)GMENTS This work was supported by the Canadian Wildlife Service (contract no. K1868-4-0250 to OI), Fundacion Antorchas, and the Universidad Nacinnal de Mar del Plata. We thank E Botto, I.. Lucifora, G. Palomo, E. Schwind, and J. Gutierrez for field assistance. We are indebted to the reviewers and the Associate

53

Editor Dr. J. Dauvin; the manuscript was greatly improved as a result o f their comments. LITERATURE CITED BILDSTEIN, K. L., S. G. McDOWELL, AND I. L. BRISBIN. 1989. Consequences o f sexual dimorphism in sand fiddler crabs, l/ca pugilator. Differential vulnerability to avian predation. Animal Behavior 37:133-139. BLANCO, D. E., G. D. PUG.~AIJ, AND H. RODRJGU~'Z Gog'l. 1988. Punta Rasa: Su importancie en la conservacion de las aves migratorias. (unpublished technical report). Bosom, E. E. 1964. Los crustaceos decapodos Brachyura del litoral Bonaerense. Boletin d,el ln.stituto de Biologia Marina, Mar del Plata (Argentina) 6:1-99. BuR'rON, P. J. K. 1974. Feeding and the Feeding Apparatus in Waders. British Museum, L o n d o n , United Kingdom. CABRERA, A. L. AND E. M. ZARDINI. 1978. Manual de la flora de los alrededores de Buenos Aires. ACME, Buenos Aires, Argentina. CHESSON, J. 1978. Measuring preference in ~lective predation. Ecology 59:211-215. DABBFN~:, R. 1920. Notas sobre chorlos de Norteamtrica que invernan en la Repfiblica Argentina. IIornero 2:99-128. EMLEN, S. T. A.'~DL. W. ORING. 1977. Ecology, sexual selection, and the evolution o f mating systems. Science 197:21.5-223. ENS, B. J., P. DuIvE.s, C . J . SMIT, AND "I: M. VAN SPA.',3E. 1990. Spring migration o f turnstones from the Bane d'Argnin, in Mauritania. Ardea 78:301-314. FRZX, M. S., M. E. HOSTETLFR, AND K. I,. BILDSTEIN. 1991. lntraand interspecilic differences in responses o f Atlantic sand (Uca pugilator) and Atlantic marsh (U. pugnax) fiddler crabs to simulated avian predators. Journal of Crustacean BioloK~ 11: 523-529. GIBSON, E. 1920. Further ornithological notes from the neighb o r h o o d o f Cape San Antonio, Province o f Buenos Aires. Ibis 2:1-91. (R)ss-CusrARD, J. D. 1970. Dispersion in some overwintering wading birds, p. 3-35. InJ.H. Crook ted.), Social Behavior in Birds and Mammals. Academic Pre~s, L o n d o n , United Kingdom. HAYMAN, E,J. MARtZtlANT,AND 7-'. PRA'IER. 1986. An identification guide to the waders of the world. Croom Helm, L o n d o n , United Kingdom. KUStlLAND,J. A. 1979. Feeding ecology and prey selection in the white ibis. Condor 81:376-389. MORmER, A. AND R. McNFIL. 1991. Time-activity budget o f Wilson's and semipalmated plovers in a tropical environment. Wilson Bulletin 103:598-620. MY~'RS,J. E AND L. E MYERS. 1979. Sborebirds of coastal Buenos Aires Province, Argentina. Ibis 121:186--200. NAROSKY, Y. AND D. YZURIFTA. 1987. Guia para la identificaci6n de aves de Ai'gentina y Uruguay. Asociacion Ornitologica del Plata, Buenos Aires, Argentina. OI.ROG, C. C. 1967. Observaciones sobre ayes migratorias del hemisferio norte. Hornero 10:292-298. PEARRV, JR., S. 1982. Estimating prey preference by predators: Uses of varions indices, and a proposal on a n o t h e r based on X'2. Can:utianJournal of FistunJes and Aquatic Scie~zces39:914-923. PF'FrlT, D. R. A.'~D K. I.. BILDSTEIN. 1987. Effect of g r o u p size and location within the g r o u p on the foraging behavior o f white ibises. Condor 89:602-609. PWNKOWSK1, M. W. 1980. ltow foraging plovers cope with environmental effects in invertebrate behavior and availability, p. 179-192. hz N. V. Jones and W.J. Wolff (eds.), Feeding and Survival Strategies of Estuarine Organisms. Plenum Press, New York. SPAANS,A. L. 1979. Waders studies in Surinam, South Ainerica. Wader Study Group Bulletin 25:32-37. SPIVAK,E. D., M. A. GAVIO, AND C. E. NAVARRO. 1991. I.ife history

54

o . o . Iribame and M. M. Martinez

and structure of the world southernmost Uca population: Uca uruguayensis (Crustacea, Brachyura) in Mar Chiquita lagoon (Argentina). Bulletin of Marine Science 48:679-688. S'rRAUCII, J. G. AND G. ABELE. 1979. Feeding ecology of three species of plovers wintering on the Bay of Panama, Central America. Studies in Avian Biology 2:217-220. STRAUSS, R. E. 1979. Reliability estimation for Ivlev's electivity index, the forage ratio and a proposed linear index of food selection. Transaction of the American Fishery ,~ociety 108:344352. SUMMERS,R. W. 1980. Impressions on a tropical mudflat. Wader Study Grmlp Bulletin 30:16. THmAULT, M. AND R. MCNEIL. 1995. Predator-prey relationship between Wilson's plovers and fiddler crabs in northeastern Venezuela. Wilson Bulletin 107:73-80. V1LA, A. R., ~E. R. BREMER, A.XD M. S. BEADE. 1994. Censos de

chorlos y playeros migratorios en la bahia de Samborombon, Provincia de Buenos Aires, Argentina. Boletin Tecnico # 22, Fundaci6n Vida Silvestre Argentina, Buenos Aires, Argentina. ZAR, J. H. 1984. Biostafistical Analysis. Prentice-ttall, Inc., Englewood Cliffs, New Jersey. ZWARTS, L. 1985. The winter exploitation of fiddler crabs, Uca tangeri, by waders in Guinea-Bissau. Ardea 73:3-12. ZWARTS, L. 1990. Increased prey availability drives premigration hyperphagia in whimbrels and allows them to leave the Banc D' Arguin, Mauritania, in time. Ardea 78:279-300. ZWARTS, L. AYr~S. DIRKSFN. 1990. Digestive bottleneck limit~ the increase in tood intake of whimbrels preparing their departure from the Bane D' Arguin, Mauritania, in spring. Ardea 78:257-278.

Receivedfor consideration,June 6, 1997 Acceptedfor publication, June 30, 1998