Estilarme, Coastal and Shelf Science 136 (2014) 26—34

ELSEVIER

C onten ts lis ts ava ilab le at S clenceD lrect

Estuarine, Coastal and Shelf Science

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / e c s s

» E S T U A R IN E

C O A ST A L

SH E L F S C IE N C E

Seasonal changes in m ollusc abundance in a tropical intertidal ecosystem , Banc d’Arguin (Mauritania): Testing the ‘depletion by shorebirds’ hypothesis^ 1̂

Mohamed Vail Ahmedou Salem a, Matthijs van der G eestb, Theunis Piersmab c,Younès Saoud3, Jan A. van Gilsb *1Laboratoire de Biologie Appliquée et Pathologie, Département de Biologie, Faculté des Science, B.P. 2121, Tetouan, Morocco b Royal Netherlands Institute for Sea Research (NIOZ), P.O. Box 59, 1790 AB Den Burg, Texel, The NetherlandscAnimal Ecology Group, Centre for Ecological and Evolutionary Studies, University o f Groningen, P.O. Box 11103, 9700 CC Groningen, The Netherlands

A R T I C L E I N F O A B S T R A C T

Article history:Received 15 March 2013 Accepted 9 November 2013 Available online 18 November 2013

Keywords:Banc d'Arguindepletionmolluscsoptimal dietseasonalityshorebirds

At tem p erate latitudes d en sitie s and b iom ass o f intertidal m olluscs ten d to be strongly seasonal. Here w e provide a com parative stu d y o n season a lity o f b ivalves and gastropods in th e tropical intertidal seagrass- covered soft sed im en t en v iron m en t o f Banc d'Arguin, M auritania (20°N , 16°W ). In th is system , b en - th ivorous shoreb irds exert considerab le predation pressure w ith strong season al variations. It has b een proposed th at during th e period w h en (ad u lt) shorebirds are ab sen t (M ay—A ugust) b en th ic b iom ass w ou ld be able to recover, but a first te s t w as inconclu sive. Over a full year (M arch 201 1 —February 2012), each m on th w e sam pled b en th ic invertebrates at s ix teen p erm an en t sites. The tota l o f 3763 sp ecim en s com p rised 20 sp ecies, rep resen ting e ig h t orders and 19 fam ilies. B ivalves w ere m uch m ore co m m o n than gastropods. The b ivalve Loripes lucinalis d om in ated th e assem b lage th rou gh ou t th e year (58% o f total num ber), fo llow ed by Dosinia isocardia (10%), Senilia senilis (8%) and th e gastropod Gibbula um bilicalis (6%). A verage b iom ass a m ou n ted to 32 g AFDM /m2, o f w h ich th e large W est-A frican b lood y cock le Senilia m ad e up three-quarter. Loripes 16%, Gibbula 2% and Dosinia 1%. Across th e 2 0 sp ecies, lo w e st d en sities w ere reached in late spring (M ay) and su m m er (Aug.), w h ereas h igh est d en sitie s occurred in autum n (Oct.). The lo w e st overall d en sity o f 676 sp ec im en s/m 2 in A ugust m ore th an d ou b led to a peak d en sity o f 1538 sp e c im en s/m 2 in October, m ost o f th e increase b e in g d u e to strong recru itm ent in both Loripes (d en sities increasing from 32 2 sp ec im en s/m 2 in Sept. to 785 sp ec im en s/m 2 in Oct.) and Dosinia (d en sities increasing from 18 sp ec im en s/m 2 in Aug. to 2 6 5 sp ec im en s/m 2 in Sept.). Our resu lts su ggest th at by th e tim e th e feath ered m olluscivore predators returned in h igh num bers to Banc d'Arguin (after th eir su m m er breed ing season in th e Arctic), b en th ic anim als w ere at a peak. In order to q u antita tively u n ­derstand th e seasonal changes in m ollusc abundance, w e build u p on a recently p u b lish ed optim al d iet m od el in w h ich th e m o st abu ndant m olluscivore shorebird, th e red kn ot (Calidris canutus), cou ld ch oose b e tw e e n Loripes and Dosinia. O bserved ch an ges in d en sitie s o f th ese tw o bivalves c lo se ly m atch d e p le ­tion trajectories pred icted by th e m od el. W e con clu d e that m olluscivore shorebirds are able to d ep lete th eir food stocks in th e course o f th eir 'w inter' in a tropical intertidal area.

© 2013 The Authors. Published by E lsevier Ltd. All rights reserved.

* Ahmedou Salem e t al.: Seasonal changes in a tropical bivalve community.* * This is an open-access article distributed under the term s of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which per­mits non-com m ercial use, distribution, and reproduction in any m edium, provided the original au thor and source are credited.

* Corresponding author.E-mail address: Jan.van.Gils@nioz.nl Q.A. van Gils).

1. Introduction

It is w ell e stab lish ed th a t b e n th ic in v erteb ra tes living in te m p e ra te in te rtid a l so ft-se d im e n t sy s tem s sh o w p ro n o u n ced seaso n ality in d e n s ity a n d b io m ass (B eukem a, 1974; Z w arts e t al., 1992; Z w arts a n d W anink, 1993; U rra e t al., 2013). In th e tropics, te m p e ra tu re flu ctu atio n s a re less p ro n o u n ced , a n d th is m ay also be tru e for th e seasonal ch an g es in food availab ility for b e n th ic in ­v e rte b ra te s (Sastry, 1979; Alongi, 1990; Lubet, 1994; M andai and H arkan tra , 2013). This cou ld lead to tro p ical in te rtid a l b e n th ic co m m u n itie s rem a in in g ra th e r stab le th ro u g h o u t th e y ear

0272-7714/$ — see front m atter is 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.Org/10.1016/j.ecss.2013.ll.009

M.V. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34 27

(P iersm a, 1982; W olff, 1991; D ittm ann , 2000), ev en in reg ions w ith d is tin c t ra in y a n d d ry seaso n s (Silva e t al„ 2011). H ow ever, b en th ic in v erteb ra tes a re th e food o f c o n su m ers such as crabs, shrim p , fish a n d sh o reb ird s (van de Kam e t al., 2004). The sh o reb ird co m p o n e n t o f th is p re d a to r c o m m u n ity is h igh ly seaso n al by v ir tu e o f th e fact th a t sh o reb ird s b re ed in n o r th e rn a rea s from w h ich th e y m ig ra te so u th to w in te r in coasta l w e tlan d s, e.g. a long th e coasts o f W est- Africa (van de Kam e t al., 2004 ; P iersm a, 2007). This w o u ld re su lt in stro n g ly seasonal p re d a tio n p re ssu re (W olff a n d M ichaelis, 2008).

Such seaso n ality in p re d a tio n p re ssu re cou ld be s tro n g in Banc d ’A rguin (M auritan ia , W est-A frica), w h ich is a n a rea o f over 10,000 km 2 o f sh a llo w in sh o re w a te rs b o rd e rin g th e Sahara (W olff e t al„ 1993a). This a rea has received p ro tec tio n from h u m an over­ex p lo ita tio n by th e e s tab lish m en t in 1976 o f th e Parc N ational d u Banc d ’A rguin (PNBA). PNBA is o n e o f th e m o st im p o rta n t w a te rb ird a rea s on th e W est-A frican coastline, h o stin g o v er tw o m illion w in te rin g sh o reb ird s (E ngelm oer e t al., 1984; Sm it an d Piersm a, 1989; Ens e t al., 1990; Z w arts e t al., 1990 ,1998), o f w h ich th e red k n o t Calidris canutus is by far th e m o st n u m ero u s m olluscivore (2 3 0 ,0 0 0 —3 0 0 ,0 0 0 indiv iduals), fo llow ed by th e m u ch less a b u n ­d a n t o y s te rca tch e r H aem atopus ostralegus (5 0 0 0 —8 0 0 0 individuals; Z w arts e t al., 1998; H agem eijer e t al., 2004 ; D iaw ara a n d Overdijk, 2007). M ore th a n 3/4 o f th e in te rtid a l flats a t Banc d ’A rguin are covered by seag rass b ed s (W olff an d Smit, 1990), w hich , co m p ared to b a re flats, su p p o rt th e h ig h est d e n sitie s o f sh o reb ird s (A ltenburg e t al., 1982). This is for good reasons, since seagrass h a b ita t is a m o n g th e m o st p ro d u c tiv e an d rich est o f sh a llo w m arin e hab ita ts , b o th locally (H onkoop e t al„ 20 0 8 ) as w ell as g lobally (D uarte an d Chiscano, 1999).

D uring th e first s tu d ies o f sh o reb ird s a n d b en th o s o f th e Banc d ’A rguin in 1980, th e h igh ra tio o f b ird s to s tan d in g b en th ic b iom ass, in co m p ariso n w ith ra tio s found e lsew h ere in in te rtid a l

so ft-se d im e n t areas, w e re u n e x p ec te d (A ltenburg e t al., 1982; E ngelm oer e t al., 1984). The h igh ra tio im p lied h igh seco n d ary p roduction , b u t a lso su g g ested th a t sh o reb ird s w o u ld be ab le to su b stan tia lly red u ce th e s tan d in g stocks o f b en th o s in th e cou rse o f th e m ain seaso n o f p red a tio n , i.e. from S ep tem b er th ro u g h April. W olff an d M ichaelis (2 0 0 8 ) te s te d th is by co m p arin g th e b en th ic b io m ass sam p led across an ex ten siv e a rea in S ep tem b er 1988 (i.e. a t th e s ta r t o f a sh o reb ird m ig ra tio n season) w ith th e s tan d in g stocks in F ebruary—M arch 1986 (i.e. a t th e end). The lack o f d ifference b e tw e e n th e tw o v a lu es m ay be ex p la in ed e ith e r by an ab sen ce of d e p le tio n o f b en th o s d u rin g th e seaso n o f sh o reb ird p red a tio n , o r b ecau se o f d ifferences b e tw e e n y ears o b lite ra ted th e m ea su re m e n t o f th e p ro p o sed effect. H ere w e prov ide a s tro n g e r te s t o f th e sea ­sonal d e p le tio n h y p o th esis by ex am in in g m o n th ly ch an g es over a full y e a r in a seagrass bed, in ten siv e ly u sed by shoreb irds, in a n ea rsh o re s ite o f th e Banc d ’Arguin. As m olluscs a re by far th e larg est co m p o n e n t o f th e in te rtid a l b e n th ic co m m u n ity o f Banc d ’A rguin (e.g. P iersm a, 1982; W olff e t al„ 1993a; M ichaelis and W olff, 2001), an d since m olluscivorous sh o reb irds, especially red knots, a re v e ry n u m ero u s , w e lim ited o u r s tu d y to b ivalves an d gastropods.

2. Material and methods

2.1. S tudy area

This s tu d y w as carried o u t from th e PBNA Scientific S ta tion a t Iw ik P en insu la o n a n ea rsh o re a rea o f d e n se seag rass b ed s (m ain ly Zostera noltii H ornem .), locally k n o w n as A belgh E iznaya (Fig. 1 ; see also Leyrer e t al„ 2006). 16 s ta tio n s w e re ch o sen ran d o m ly w ith in a 2 0 0 -m rad iu s from a c en tra l p o in t (19°53.40 'N , 16°18.50'W ), w h ich w e re sam p led a t m o n th ly in te rvals from M arch 2011 to February 2012. A bout 1/3 o f th e s ta tio n s w as loca ted in d e n se seagrass

AbelghEiznaya

presence red knots

Fig. 1. Map of the Banc d’Arguin, with tidal flats given in dark grey, ocean in light grey and land in white. Black dot indicates the centre of our study area, Abelgh Eiznaya. Scale bar is expressed in metres. The pie indicates the presence of red knots in Banc d’Arguin (both adults and young birds present during September—April; adults are leaving and arriving in May and August, respectively; only young birds present during June—July). Note that the sampling area of Wolff and Michaelis (2008) was much larger than ours and covered roughly all tidal flats plotted in this map.

28 M.V. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34

h ab ita t, 1 /3 in in te rm ed ia te seag rass an d 1 /3 in b a re h ab ita t, w ith fine (m ed ian g rain size ~ 5 0 |im ), in te rm ed ia te ( ~ 1 0 0 |im ) and coarse se d im e n t tex tu re ( ~ 1 5 0 |im ), re sp ec tive ly (Fo lm er e t al„2012). Tidal reg im e is th e u su a l M2 type, also k n o w n as th e ‘p r in ­cipal lu n ar se m i-d iu rn a l’ type, w ith a n average am p litu d e o f 1.61 m (W olff a n d Smit, 1990). M eteorological m ea su re m e n ts a t a site 65 km so u th o f th e s tu d y a rea (a t N ouam ghar; h ttp : / /w w w . accuw eath er.co m /fr/m r/n o u am g h ar/1 1 1 2 5 8 5 /m o n th /1 1 1 2 5 8 5 ?) sh o w ed m o n th ly v a ria tio n s in m in im al (15.0—25.7 °C) a n d m ax im al (29 .0—37.8 °C) daily a ir te m p e ra tu re s d u rin g o u r s tu d y period , w ith th e m in im u m in F ebruary an d m ax im u m in October. M ost ra in fell in A ugust (27 m m vs. 3 m m /m o n th o n average).

2.2. Field sam pling and laboratory analyses

Each m o n th a t e ach s ta tio n w e to o k a single sam p le d o w n to a d e p th o f 2 0 cm u sin g a core w ith a 15 cm d iam eter. In o rd e r to d e te rm in e w h e th e r p rey w e re w ith in reach o f th e 3.5 cm long bill o f red k no ts (Z w arts e t al., 1992), w e sep a ra te ly sieved th e to p layer (0 —4 cm ) from th e b o tto m layer (4 —20 cm ), u sin g a 1 -m m m esh ed sieve. The m ate ria l rem a in in g on th e sieve w as p u t in sep a ra te labelled p lastic bags an d frozen w ith in a few h o u rs a f te r sam pling. Over su b seq u e n t days all fleshed b ivalves an d g astro p o d s w ere so rted from th e d eb ris o f d ead shells a n d seag rass m ateria l. Spec­im en s w e re id en tified to species level, th e ir len g th s w e re m easu red to th e n e a re s t 0.1 m m , an d th e y w e re th e n fixed in bu ffered fo rm ald eh y d e (4%). S u b seq u en tly (M ay 2012), th e b io m ass o f th e m o st a b u n d a n t species (D iplodonta circularis, Dosinia isocardia, Gibbula umbilicalis, Loripes lucinalis, M esalia mesal, M odiolus sp., Petricolaria gracilis, Senilia senilis, an d Tellina distorta) w as d e te r ­m in ed in th e lab o ra to ry a t NI0Z as ash -free d ry m ass (AFDM). In o rd e r to do so, sp ec im en s w e re d ried a t 60 ° C for a t least 48 h and w e re a fte rw ard s in c in e ra ted a t 560 °C for 2 h (bivalves w ith o u t shells, g a stro p o d s w ith shells).

2.3. Depletion m odel

W e u sed a n ex isting o p tim a l d ie t m odel th a t has b e en d ev elo p ed for th e m o st a b u n d a n t m olluscivore sh o reb ird p re se n t in o u r s tu d y area , th e red k n o t (van Gils e t al., 2013). Red kno ts ingest th e ir p rey w ho le , a n d h en ce p rey th a t have a c ircu m feren ce larg e r th a n th e ir g ap e c an n o t b e in g ested (Z w arts an d B lom ert, 1992). M ost m ollusc species found a t Banc d ’A rguin d o n o t g ro w v e ry large an d th e re fo re large fractions a re ingestib le for red kno ts (P iersm a e t al., 1993).

W e on ly m o d elled th e d e p le tio n o f Loripes lucinalis a n d Dosinia isocardia a n d left o u t all th e o th e r m ollusc species. This is b ecause Loripes an d Dosinia a re th e tw o m o st a b u n d a n t m olluscs in th e system . The th ird m o st a b u n d a n t m ollusc, Senilia senilis, is a th ick -sh e lled bivalve w h ich g row s to h e ig h ts o f 80 m m a n d is th e re fo re a lm o st e n tire ly unavailab le to red k n o ts (W olff e t al., 1987). Our focus o n Loripes an d Dosinia on ly is also ju s tified becau se th e se tw o species to g e th e r re p re se n t 80% o f th e m olluscs co n su m ed by red k no ts a t Banc d ’A rguin (van Gils e t al., 2012; O n rust e t al., 2013; van Gils e t al., 2013).

Even th o u g h Loripes is by far th e m o st a b u n d a n t bivalve a n d h as a very th in shell, it is n o t th e m o st p re fe rred p rey o f th e tw o (van Gils e t al., 2012; v an Gils e t al., 2013; O udm an e t al., su b m itted for publication). R ecent ev idence suggests th a t th is is d u e to th e specialized m etab o lism o f Loripes th a t lives in sym biosis w ith ch em o au to tro p h ic b acteria inside its gili s tru c tu re s (H erry e t al., 1989). Feeding ex p erim e n ts w ith captive red knots sh o w ed th a t th ey develop d iarrhoea, d eh y d ra te a n d lose ap p e tite w h e n on a m onospecific d ie t o f Loripes (O udm an e t al., su b m itted for publication). The d e p le tio n m odel u sed h ere tak es acco u n t o f th e lo n g -te rm m ax im u m in take ra te th a t red k no ts can su s ta in w h e n

foraging o n Loripes. This m odel, called th e Toxin D igestive Rate M odel (TDRM), no t on ly ex p la ined d ie t choice d ifferences b e tw ee n years, it a lso ex p la ined m o st o f th e v a ria tio n in an n u al survival ra tes in a n 8-y r s tu d y o f red k n o t d em ograph ics (van Gils e t al., 2013). The m o d el is an ex ten sio n o f th e so-called D igestive Rate M odel (DRM) th a t exp la ined d ie t choice in red kno ts feeding on non-tox ic p rey (van Gils e t al., 2005a) an d w e re fer to v an Gils e t al. (van Gils e t al., 2013) for its full m a th em atica l deta ils; h e re w e w ill o n ly in tro d u ce th e basics.

A ccording to H olling’s func tional re sp o n se m o d el (Holling, 1959), e n e rg y in tak e ra te Y o n tw o p rey ty p es (labe lled 1 an d 2) is g iven by:

P iQ iN te t + p 2a2N 2e2 1 + p i a 1N 1h 1 + p 2a 2N 2h 2 ( 1 )

w h e re p is th e accep tan ce p ro b ab ility u p o n enco u n ter, a is search ing efficiency, N is available nu m erica l p rey density , e is e n ­e rgy v a lu e a n d h is h an d lin g tim e for re sp ec tive ly p rey ty p e 1 a n d 2 . U nder som e c ircu m stan ces Y can b e m ax im ized by n o t accep ting ev ery p rey th a t is e n co u n te red . D iet se lec tion m odels m ake p re ­d ic tions ab o u t th e va lues o f p i a n d p 2 in o rd e r to m ax im ize Y.

The classical d ie t m odel (S tephens an d Krebs, 1986), som etim es called th e con tingency m odel (CM), ranks prey ty p es on th e basis of profitability, i.e. e i / f l i > e2j h2 .A forager should on ly accep t p rey type 1 (p i = 1 ) and neglect prey ty p e 2 (p2 = 0 ) w h e n th e en erg y in take rate on prey ty p e 1 a lone exceeds th e profitability o f ty p e 2, i.e. w hen:

at Niei ^e21 + a iN ih i h 2

(2 )

If not, it sho u ld accep t b o th p rey ty p es in its d ie t (p i = 1 ; p 2 = 1 ).The CM ignores d igestive p rocessing co n stra in ts , w h ich real-

w o rld fo ragers o ften face, especially th e o n es such as m ollusci- v o res feed ing o n bu lky prey. The d igestive ra te m o d el (DRM), dev elo p ed by H irakaw a (1995) a n d successfully ap p lied to red kno ts (van Gils e t al., 2005a; v an Gils e t al., 2 005b ; Kraan e t al., 2009 ; Q u a in ten n e e t al., 2010), does tak e acco u n t o f such a c o n stra in t. D efining k as a p rey ty p e ’s in d ig estib le b u lk m ass, ballast in tak e ra te X is d e fin ed as:

Xp í a í N í k í + p 2a2N 2k2

1 + p i a 1N 1h 1 + p 2a2N 2h 2 (3 )

D efining c as th e m ax im u m ra te a t w h ich bu lky m ate ria l can be p rocessed by a fo rager’s d igestive system , th e follow ing co n d itio n sh o u ld a lw ays be m et:

X < c (4 )

The o p tim a l so lu tio n for p i a n d p 2 can e ith e r be found g ra p h i­cally o r analy tically a n d w e re fe r to th e o rig inal p a p e r b y H irakaw a (1995) o r to th e first ap p lica tio n in red k n o ts b y van Gils e t al. (2005a).

Finally, th e TDRM n o t on ly tak es acco u n t o f a c o n s tra in t on processing bu lk y food, it a lso includes a m ax im ally to le rab le tox in in tak e ra te . D efining s as a p rey ty p e ’s to x in co n ten t, to x in in take ra te Z is g iven by:

P ia iiV tS ! + p 2a2N 2s2 1 + p i a t N t h t + p 2a2N 2h 2

(5 )

D efining q as th e m ax im ally to le rab le to x in in tak e rate , th e follow ing co n d itio n sh o u ld a lw ays be m et:

Z < q (6 )

P aram e te rs u sed (Table 1 ) a re as in van Gils e t al. (van Gils e t al.,2013), ex cep t for e n erg y va lues e an d ba lla st m asses k, w h ich a re

M.V. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34 29

Table 1Overview of the parameters used in the depletion model.

Parameter Definition Value Unit Source(s)

aL Searching efficiency on Loripes 4 cm2/s 1, 2aD Searching efficiency on Dosinia 4 cm2/s 1, 2c Max. digestive processing rate 5 m g DMsheii/s 3eL Energy value Loripes 5.91 m g AFDMflesh 4eD Energy value Dosinia 2.68 m g AFDMflesh 4hL Handling time Loripes 1 s 1, 2

Handling time Dosinia 1 s 1, 2kL Ballast mass Loripes 41.56 m g DMsheii 4ko Ballast mass Dosinia 34.31 m g DMsheii 4q Max. tolerable toxin intake rate 0.1 m g AFDMflesh/s 5SL Toxin contents Loripes eL m g AFDMflesh 5Sd Toxin contents Dosinia 0 m g AFDMflesh 5

Sources: 1 de Fouw et al. (in preparation); 2 Piersma et al. (1995); 3 van Gils et al. (2003); 4 this study; 5 Oudman et al. (submitted for publication).

es tim a te d in th e c u rre n t study. T hese p a ram ete rs , in co m b in a tio n w ith th e re la tive ly low d e n sitie s o f Dosinia found in th is s tu d y (see R esults below ), sim plify th e m ath e m a tic s o f th e TRDM. Nam ely, on all m o n th ly average d en sitie s o f Loripes a n d Dosinia found in th is study, red kno ts w o u ld n ev er face a d igestive c o n s tra in t c b u t a l­w ays a to x in c o n stra in t q (u sing Eqs. S8—S9 in van Gils e t al. (2013) revea ls th a t for th e c u rre n t p a ram e te r se ttin g s d igestive co n stra in t c occurs above 365 available D osin ia/m 2, w h ile to x in c o n s tra in t q is reach ed above 49 available Loripes/m 2). This im plies th a t all avail­ab le Dosinia c an b e accep ted (po = 1 ), w h ile on ly a p ro p o rtio n o f th e available Loripes can b e accep ted (i.e. p L < 1 as long as d en sitie s a re above 49 sp ec im en s /m 2). This la tte r p ro p o rtio n p¿ can b e calcu la ted b y assu m in g a to x in in tak e ra te th a t is m ax im ally to le rab le , i.e.:

Z = q (7)

Since o n ly Loripes is toxic (si = e¡.) a n d Dosinia is n o t (sd = 0), an d since p D = 1, th is can be w r itte n as:

P lQlN lSl

1 + PlOlN lIil + aDN DhD

R ew riting th is, p i equals:

q ( l + aDN DhD)P l aLN L(sL - qh L)

(8 )

(9 )

Since d e p le tio n is ex p ressed as ch an g es in p rey num bers, w e n eed to calcu la te th e num erical in tak e ra te R g iven b y H olling’s functional response , w h ich o n Loripes eq u als (w ith p D falling o u t since it eq u als 1):

RiPlOlN l

1 + PLa¡N¡hL +( 10 )

A long th e sam e lines, n u m erica l in tak e ra te RD on Dosinia equals:

RdaDN D

( 1 1 )1 + P iO iN ih i + aDN DhD

S ub stitu tin g Eq. (9) in to Eqs. (10) a n d (11), respectively , gives:

Rd

q_sL

_ apN p (sL - hLq) s t ( l + OßNDhD)

( 12 )

(1 3 )

T hese nu m erica l in tak e ra te s a re m u ltip lied by th e red kno t feed ing tim e p e r m 2 (ex p ressed as V) for o u r A belgh E iznaya s tu d y

area . The average red k n o t feed ing d e n s ity o f 18.3 b ird s /h a w as de riv ed from u n p u b lish ed obse rv a tio n s o f feed ing red k n o ts carried o u t in M arch 20 0 7 an d F ebruary 20 0 8 over tw o a n d four days, re spec tive ly (van Gils, v an d e r G eest & Folmer, unpub l. data ). U n­fortunately , w e have no d e ta iled co u n ts available d u rin g th e period o f sam p lin g b e n th o s (M arch 2011—F ebruary 2012), b u t th e fact th a t w e d id n o t find a d ifference b e tw e e n th e 2 0 0 7 a n d th e 2 0 0 8 co u n ts (W elch Two Sam ple t-te st, t = -0 .3 4 , d f = 3.11, P = 0.75), suggests th a t in te r-an n u a l v a ria tio n s in red kno t n u m b ers a re neglig ib le for o u r p u rpose . N evertheless, w e w ill ex p lo re th e sen sitiv ity o f o u r re su lts w ith re sp ec t to V, th e m o re so becau se red k n o t d en sitie s a t A belgh Eiznaya te n d to b e h ig h er th a n th e average for Banc d ’Ar­gu in (A ltenburg e t al., 1982; v an Gils e t al., 2009). A ssum ing a daily feed ing p e rio d o f 14 h (A ltenburg e t al., 1982; van Gils e t al., 2005b; Leyrer e t al., 2 0 0 6 ) gives an e s tim a te d V o f 1.07 • 10~3 feed ing sec­o n d s p e r m 2 p e r second b e tw e e n m id -O cto b er an d th e e n d o f M ay (i.e. 92.3 feed ing seconds p e r m 2 p e r day). Since 1 -yr o ld red knots stay a t Banc d ’A rguin d u rin g su m m er (June—A ugust), w e a ssu m ed a p re d a tio n p re ssu re V th a t is on ly 7% o f th e va lue d u rin g O ctober— M ay (van Dijk e t al., 1990). The m odel p red ic ts th e d e p le tio n from m id -O cto b er to m id-A ugust, u sing th e d en sitie s o b se rv ed in O ctober as s ta rtin g values, w ith available d en sitie s u p d a te d for d e p le tio n p e r s tep len g th A t o f 1 s in th e follow ing w ay for Loripes a n d Dosinia, respectively :

U l N,L, t - 1 - r lv

Np, t = n d , t - i - Rdv

(1 4 )

(1 5 )

Next, unavailable d en s itie s a re a d d ed to available d en sitie s in o rd e r to arrive a t total d en s itie s a t tim e t (w h e re unavailab le d e n ­sities a re th e unavailab le d en sitie s m easu red in s ta rtin g m o n th October, b e in g 222 .82 an d 31.83 m "2 for Loripes (> 4 cm d e p th ) an d Dosinia (d e p th > 4 cm, len g th > 13.2 m m ), respectively).

2.4. Statistics

All sta tis tics an d g rap h s w e re carried o u t in R (R D evelopm en t Core Team, 2013), includ ing th e GAM m odels th a t w e re u sed to sm o o th seaso n al tre n d s in n u m erica l d en sitie s (using package m gcv; W ood, 2006). In th e se GAMs w e ad d ed m o n th as a n o n - p a ram etrica lly sm o o th ed function an d sam p lin g s ta tio n as a p a ra ­m etrica lly e s tim a te d p aram eter.

3. Results

From 192 cores co llected d u rin g th e w h o le s tu d y period , a to ta l o f 3,763 m ollusc sp ec im en s w e re re trieved . T hese com p rised 20 species (Table 2) a n d re p re se n ted e ig h t o rd e rs an d 19 fam ilies. Bi­valves o ccurred in h ig h er d e n sitie s th a n g astro p o d s, w ith n ine bivalve species re p re se n tin g 88% o f all spec im ens. Overall, th e lucin id Loripes lucinalis d o m in a ted th e assem b lag e (58% o f all spec im ens), fo llow ed by th e v en erid Dosinia isocardia (10%) an d th e arcid Senilia senilis (8%).

The overall average b iom ass o f th e m ollusc co m m u n ity a m o u n te d to 32 g AFDM/m2, o f w h ich th e W est-A frican b loody cockle Senilia m ad e up th re e -q u a r te r (Fig. 2). H ighest b iom ass d en sitie s w e re reach ed in February a n d July, a n d low est in M ay (Fig. 2A). The c o n trib u tio n o f Senilia w as lo w est in M arch—May, reach ing a p eak in th e re la tive c o n trib u tio n o f a b o u t 85% to overall b io m ass in February, Ju n e an d Ju ly (Fig. 2B). On average, Loripes c o n trib u ted 19%, a p e rcen tag e th a t reach ed a m in im u m in July (9.6%) a n d a m ax im u m in M arch (34.6%; Fig. 2B). Across th e 20 species (Table 2), low est d en sitie s w e re reach ed in la te sp ring (M ay)

30 M.V. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34

Table 2List of all mollusc species encountered (ordered alphabetically), with density and biomass during the month(s) with the lowest and highest numerical densities.

Species Lowest density month(s) Highest density month(s)

Month(s) Density (m 2) Biomass (g AFDM/m2) Month(s) Density (m 2) Biomass (g AFDM/m2)

Bivalvia Abra tenuis 5, 9 -1 2 0.0 0.00 6 81.3 0.12Diplodonta circularis 2 10.6 0.21 6 95.5 0.13Dosinia isocardia 8 17.7 0.12 9 265.3 0.75Loripes lucinalis 9 321.8 2.88 3 838.2 7.69Modiolus sp. 3 - 4 0.0 0.00 7 148.5 0.55Paphia dura 1 -4 , 6 -1 2 0.00 0.00 5 7.1 0.59Petricolaria gracilis 3 17.7 0.15 10 53.1 0.37Senilia senilis 7 49.5 37.14 11 116.7 23.61Tellina distorta 11 7.1 0.02 3 84.9 0.09Gastropoda Bittium reticulatum 2, 5 -6 , 8 -1 1 0.0 0.00 12 212.2 0.70Bulla adansoni 2, 6, 8 0.0 0.00 5 31.8 0.24Clavatula bimarginata 1, 3, 5 -7 , 9 0.0 0.00 11 7.1 0.44Columbella rustica 1, 3 -1 2 0.0 0.00 2 3.5 0.01Crepidula sp. 1 ,3 -1 1 0.0 0.00 12 3.5 0.02Gibberula sp. 3 -5 , 7 -8 0.0 0.00 10 35.4 0.07Gibbula umbilicalis 4 0.0 0.00 10 159.2 1.94Hydrobia ulvae 1 -2 , 4 -5 , 7 -1 2 0.0 0.00 3 3.5 0.00Mesalia mesal 3 0.0 0.00 11 70.7 1.10Nassarius pfeifferi 1 -6 , 8 -1 2 0.0 0.00 7 10.6 0.10Prunum amygdalum 2, 4 - 6 , 11 0.0 0.00 10 7.1 0.08Total 8 675.5 25.57 10 1538.5 30.50

to 785 sp ec im en s /m 2 in O ctober (Fig. 3A). Similarly, Dosinia in creased from 18 sp ec im en s /m 2 in A ugust to 265 sp ec im en s /m 2 in S ep tem b er (Fig. 3B), an d Senilia in creased from 50 sp ec im en s /m 2 in Ju ly to 117 sp ec im en s /m 2 in N ovem ber (Fig. 3C). Two o th e r b ivalves th a t sh o w ed p eak n u m b ers in su m m er: M odiolus sp. (m ax im u m o f 149 sp ec im en s /m 2 in July; Fig. 3D) an d Diplodonta circularis, (m ax im u m 95 sp ec im en s /m 2 in June ; Fig. 3E), w h ile Tellina distorta p eak ed in sp rin g (m ax im u m 85 sp ec im en s /m 2 in M arch; Fig. 3F) an d Petricolaria gracilis in a u tu m n (m ax im u m 53 sp ec im en s /m 2 in O ctober). The sem elid Abra tenuis w as found in several m o n th s b u t a lw ays in low d en sitie s th a t w e re too low to b e ab le to d e te c t c lear seasonal v a ria tion . The v en erid Paphia dura w as e n co u n te re d in M ay only.

N um erically, th e g a stro p o d s re p re se n ted 12% o f th e assem blage, w ith th e tro ch id Gibbula um bilicalis (6%) b e in g th e on ly o n e c o m ­m o n e n o u g h for u s to d e te c t seasonal flu ctu atio n s in th e ir n u m b ers (reach in g its m ax im u m d e n s ity in O ctober; Fig. 3G). The tu rrite lid M esalia m esal a n d th e sm all snail B ittium reticulatum c o n trib u ted 2% each, w h ile th e o th e r g astro p o d s c o n trib u ted ev en less ( < 1% each; in d ecreas in g o rd e r o f a b u n d an ce: Gibberula sp., Bulla adan­soni, Clavatula bimarginata, Prunum am ygdalum , Nassarius pfeifferi, Columbella rustica, Crepidula sp. an d Hydrobia u lvae ; Table 2).

The d e p le tio n m o d el q u a lita tiv ely m atch ed th e p a tte rn o f change o b se rv ed in Dosinia an d Loripes d en sitie s (Fig. 4). The in itial fivefold dec lin e in Dosinia d e n sitie s (O ct—Mar), fo llow ed b y a n early tw ofo ld decline in Loripes d e n s ity (M ar—Aug), stro n g ly suggests th a t p re d a tio n is th e m ain d riv er b eh in d th e se changes, especially p re d a tio n by m olluscivorous red k n o ts w h ich shou ld in itially c o n su m e th ick er-sh e lled Dosinia a t a h ig h er ra te th a n th e th in n e r- she lled b u t po iso n o u s Loripes. In o th e r w ords, th e re la tive c o n tr i­b u tio n o f Loripes to th e red k n o t’s d ie t sho u ld g rad u ally increase as Dosinia stocks b eco m e d ep le ted , an d th is is in d eed w h a t th e b en th o s d a ta seem to suggest. Testing for th e g o o d n ess o f fit o f th e d a ta (O ct—Aug) w ith th e d e p le tio n m odel revea led th a t o bserved d en sitie s do n o t d iffer from p red ic ted d en sities, n o r in Loripes (log io(observed d en sity ) = 0.61 + 0.81 x log io (p red ic ted density ), R2 = 0.65, f 1,9 = 16.7, P = 0 .003; in te rce p t d o es n o t d iffer from 0 (P = 0.29); slope d o es n o t differ from 1 (P = 0.35)), n e ith e r in Dosinia (log io(observed d en sity ) = 0 .62 + 0.75 x log io(pred icted

an d su m m er (Aug), w h e rea s h ig h es t d en sitie s occu rred in a u tu m n (Oct).

R eflecting s tro n g re c ru itm e n t (as ev id en ced by th e su d d e n o ccu rren ce o f sm all, 2—6 m m long sp ec im en s in th e sam ples), Loripes in creased in d e n sity from 322 sp ec im en s /m 2 in S ep tem b er

1 0 0 - • totalO total w ithout Senilia

E 50-

LL

3 2 0 -

1 0 -

5 -

100 —i— i— rSenilia

I other

I Loripes80-

60-

a> 40-

2 0 -

m onth

Fig. 2. Seasonal changes in (A) total and total without Senilia biomass and in (B) community composition (in biomass terms) of molluscs at Abelgh Eiznaya, Banc d’Arguin, Mauritania, from March 2011 to February 2012.

M.V. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34 31

Loripes lucinalis1000 -

8 0 0 -

6 0 0 -

4 0 0 -

2 0 0 -

Senilia senilis1 4 0 -

10 0 -

6 0 -

CM

ê 2 0 -

mœ 1 2 0 -

Diplodonta circularis

8 0 -

4 0 -

Gibbula umbilicalis2 0 0 -

1 5 0 -

10 0 -

5 0 -

0M A M J J A S O N D J F

month

Fig. 3. Seasonal changes in the densities of the seven most common mollusc species at means, bars give SE and smoothed lines give GAM fit (solid line) ± 2 SE (dashed lines).

density ), R2 = 0.54, Fi g = 10.5, P = 0.01 ; in te rce p t d o es n o t differ from 0 (P = 0.15); slope d o es n o t d iffer from 1 (P = 0.30)). T hese re su lts rem a in ed un affec ted w h e n w e carried o u t a sensitiv ity analysis by vary ing p re d a tio n p re ssu re V b y ±10% (V red u ced by 10%: in te rcep ts d o n o t d iffer from 0 (P = 0 .70 an d 0.15 for Loripes a n d Dosinia, respectively), slopes do n o t d iffer from 1 (P = 0.76 an d P = 0.29); V in creased by 10%: in te rce p ts do n o t differ from 0 (P = 0.09 a n d 0.15), slopes do n o t differ from 1 (P = 0.12 an d P = 0.32)).

4. Discussion

In th e p re se n t study, m o st (73%) o f th e m ollusc b io m ass w as co n trib u ted by Senilia senilis. Its re la tive sh a re lies in th e range of 6 3 —85% found earlie r for th e e n tire in te rtid a l Banc d ’A rguin (P iersm a, 1982; W olff e t al., 1993a; se lec ting m olluscs on ly in th ese s tud ies). H ow ever, its ab so lu te v a lu e (24.1 AFDM g /m 2) exceeds p rev ious e s tim a te s th a t ran g ed from 4.7 to 8.1 g AFDM/m2 (P iersm a, 1982; W olff e t al., 1993a). B iom ass d en sitie s o f th e o th e r

Dosinia isocardia3 0 0 -

2 0 0 -

10 0 -

Modiolus sp.2 0 0 - P,1 5 0 -

100 -

5 0 -

Tellina distorta12 0 -

8 0 -

4 0 -

M A M J J A S O N D J Fmonth

Eiznaya, Banc d’Arguin, Mauritania, from March 2011 to February 2012. Dots give

tw o a b u n d a n t bivalves, Loripes lucinalis an d Dosinia isocardia, a re also h ig h er th a n e s tim a te s m ad e for th e 1980s (5.2 vs. 2.6 g AFDM/ m 2 in Loripes a n d 0.3 vs. 0.1 g AFDM/m2 in Dosinia; W olff e t al., 1993a). H ow ever, n o te th a t sam pling in th e 1980s co vered an a rea o f a b o u t 3 0 ,0 0 0 ha, w h ile o u r s tu d y a rea w as re s tr ic ted to ab o u t 50 ha on ly (Fig. 1). H ence, w e c an n o t b e conclusive ab o u t th e se differences.

A lthough n u m ero u s b e n th ic s tu d ies have n o w b e en carried ou t o n th e Banc d ’A rguin tid a l flats (P iersm a, 1982; W olff e t al., 1987, 1993a,b; M ichaelis a n d W olff, 2001; W olff a n d M o n tse rra t, 2005; H onkoop e t al., 20 0 8 ; W olff a n d M ichaelis, 2008 ; v an Gils e t al., 2012), th is is th e first fu ll-year su rv ey o f n u m erica l ch an g es in th e m o lluscan a ssem b lag es in Banc d ’Arguin. H ow ever, such year- ro u n d s tu d ies on seaso n ality a re ra re (B eukem a, 1974; W olff and d e W olf, 1977; D ittm ann , 2002 ; d e Goeij e t al„ 2003).

M ollusc b io m ass an d n u m b ers in o u r s tu d y sh o w ed co n sid er­ab le seasonal varia tion . H ighest ab u n d an c es occu rred in th e a u tu m n a n d w in te r m o n th s a n d low est in sp rin g a n d sum m er, w ith a d o u b lin g o f to ta l ab u n d an c e from A ugust to O ctober (Table 2).

32 M.V. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34

1000-

800-

700-OJE co

~ 600- COc=0)co 500 -CD■9-

coco-4 400-

TOXIN CONS T R AI NT

0 50 100 150 200 250 350 400Dosinia density (n r2)

Fig. 4. Biplot of the observed density changes in the two bivalves Loripes lucinalis and Dosinia isocardia a t Abelgh Eiznaya, Banc d’Arguin, Mauritania, from March 2011 to February 2012. Grey line gives the predicted depletion due to predation by red knots from mid-October to mid-August, taking October densities as starting values. Letters on the grey line denote months in the simulation. Vertical and horizontal bars give critical densities above which molluscivorous red knots face a digestive and a toxin constraint, respectively. At these critical densities red knots achieve an intake rate that the birds can just sustain in the long run when it comes to digestive processing and tolerating the toxin load (for details how these critical densities are derived we refer to van Gils e t al., 2013). Given the densities found in this study, red knots would only face their toxin constraint and not their digestive constraint.

D uring th e a u tu m n m o n th s (S ep tem b er a n d O ctober), th e increase in a b u n d an ce a n d species rich n ess o f m olluscs has b een genera lly re la ted to re c ru itm e n t ev en ts (H odgson, 2010; v an d e r G eest e t ai., su b m itted for publication). This also seem s to b e th e case in o u r study, as can for ex am p le b e in ferred from th e fact th a t th e d o ub ling o f to ta l a b u n d an ce from A ugust to O ctober is acco m p an ied by an increase in to ta l b io m ass o f 19% only. The m arg inal b iom ass d e ­clines d u rin g su m m er Qui—Sep; Fig. 2A) a re m o st likely d u e to h igh te m p e ra tu re s lead ing to desicca tio n (Alongi, 1990) a n d oxygen- lim ita tio n (Ferguson e t al„ 2013).

In co n trast, th e s teep dec lin es in d en sities, o b se rv ed b e tw ee n a u tu m n an d spring , a re m o st likely cau sed by in ten se p re d a tio n by sh o reb ird s sp en d in g th e ir n o n -b ree d in g seaso n in large n u m b ers a t Banc d ’A rguin (an d in sm alle r n u m b ers by y o u n g b irds d u rin g sum m er; v an Dijk e t ai., 1990). In fact, th e o b se rv ed re d u c tio n in b iom ass (36%; exclud ing Senilia; Fig. 2A) an d n u m b ers (45%; again exclud ing Senilia) b e tw e e n O ctober an d M ay paralle ls th e calcu la­tio n s carried o u t o n th e basis o f sho reb ird n u m b ers a n d th e ir e n ­e rgy b u d g e ts for tro p ical in te rtid a l a rea s (p. 228 in van d e Kam e t ai., 2004). O ur find ings d iffered to th o se o f W olff a n d M ichaelis (2008), w h o d id n o t find a d ifference in b e n th ic b io m ass b e tw ee n a u tu m n a n d spring . H ow ever, as a d m itte d by th e se au th o rs , th e ir desig n d id n o t a llow d iscern in g b e tw e e n a n effect o f seaso n an d an effect o f year, since th e ir a u tu m n sam p lin g w as carried o u t in a d iffe ren t y ear (1988) th a n th e sp ring sam p lin g (1986). Furtherm ore , th e ir co m p ariso n in cluded th re e species o f available b ivalves only. In terestingly , am o n g th e se th re e bivalve species, it w as Dosinia th a t d eclin ed th e s te ep e s t in th e ir s tu d y (by 64%), a re su lt th a t is in line w ith o u r d e p le tio n -b y -re d -k n o ts m odel.

Finally it is o f n o te th a t p re d a tio n tak es p lace o n ind iv idual p rey an d h en ce it m ay b e b e tte r to c o m p are n u m erica l d en sitie s ra th e r th a n b io m ass d ensities. As p re d a tio n m ay lead to co m p etitiv e re lease am o n g prey, it m ay lead to h ig h er p er capita b io m asses a n d /

o r g ro w th ra te s am o n g surv iv ing p rey (G urev itch e t al„ 2000). As a co nsequence, to ta l b iom ass m ay ev en in crease a f te r p re d a tio n (de Roos e t ai., 2007 ; S chröder e t ai., 2009). M ore to th e p o in t and b e ing an ex am p le o f re lease o f in terspecific com p etitio n , it has recen tly b e en sh o w n th a t Loripes g row s faste r w h e n Dosinia b e ­com es d ep le ted , p re su m ab ly becau se Loripes is m ixo troph ic and also relies p a rtly o n th e sam e resources as su sp en sio n -feed in g Dosinia (van Gils e t al„ 2012; v an d e r G eest e t al„ su b m itte d for publication). N ote th a t o th e r p re d a to rs th a n shoreb irds, no tab ly fish, m ay have also p layed a ro le in d im in ish in g m ollusc d en sities. H ow ever, a lth o u g h m o st fish species o b se rv ed o n th e tid a l flats a t h igh tid e feed on b e n th ic fauna (W olff e t al„ 2005), o u r tw o focal p rey species, Loripes an d Dosinia, do no t seem to be im p o rta n t as p rey species for fish (W.J. W olff, pers. com m .). Shrim p, a n o th e r p o ten tia l p re d a to r o f m olluscs, a re also co m m o n ly found a t th e Banc d ’A rguin tida l flats (S chaffm eister e t ai., 2006). H ow ever, th e size c lasses fed u p o n b y sh rim p a re sm alle r th a n th e size c lasses co n sid ered h e re (A ndresen e t al„ 2013). M ost c rabs th a t o ccu r a t Banc d ’A rguin a re d ep o sit-feed in g Uca sp. (W olff e t al„ 1993a).

M ore d e ta iled s tu d y on th e m o st im p o rta n t av ian m olluscivore in o u r s tu d y system , th e red knot, a llow s u s to exp la in d ifferences in th e seasonal tim in g o f d e p le tio n o f several species o f m olluscs. Red kno ts acco u n t for 80% o f all p re d a tio n on m olluscs b y v e rte b ra te s (van Gils e t al„ 2012), w h ich ju stifie s o u r focus o n th is p re d a to r o f m olluscan inverteb ra tes . It has b e en sh o w n e lsew h ere th a t th e m o st im p o rta n t p rey for red k n o ts in Banc d ’A rguin is Dosinia and no t th e m u ch th in n e r she lled Loripes (van Gils e t al„ 2012, 2013). Since red k no ts can on ly su s ta in an in tak e ra te o f o n ly 0.1 m g AFDM/s o n d ia rrh o ea-in d u c in g Loripes (O udm an e t al„ su b m itted for pub lication), th e d e p le tio n ra tes on th is species a re re la tively low (Fig. 4). U pon arrival from th e ir b reed in g g rounds, red k n o ts in O ctober have th e lu x u ry to co n su m e Dosinia a t a re la tive ly h igh ra te (b ecause o f a h igh Dosinia d en sity ) an d ad d Loripes a t a re la tively low b u t co n s tan t ra te o f 0.1 m g AFDM/s. H ow ever, by th e tim e th a t m o st available Dosinia have b e en d e p le ted (an d o th e r sm all b u t less a b u n d a n t no n -to x ic bivalves), in th e second h a lf o f w in te r (Fig. 4), th e re la tive co n trib u tio n o f Loripes to th e red k n o t d ie t increases (b u t still a t th e ab so lu te ra te o f 0.1 m g AFDM/s) an d peaks d u rin g sp rin g /su m m er. This seasonal d ie t sh ift m im ics d ifferences in d ie t co m p o sitio n b e tw e e n years. In y ears th a t Dosinia is re la tive ly poor, red kno ts include re la tive ly m u ch Loripes in th e ir d ie t (65%), w h ile in rich Dosinia y ears th e p ro p o rtio n o f Loripes in th e d ie t levels o u t to a ro u n d 20% (O n ru st e t al., 2013; v an Gils e t al., 2013).

This d e te rio ra tio n o f feeding conditions for m olluscivorous shoreb irds m ay explain w h y fuelling ra tes o f red kno ts a t Banc d ’Ar­gu in a re relatively low (P iersm a e t al„ 2005). W ith an in take ra te o f a t m ost 0.1 m g AFDM/s on Loripes only, red knots m ay find it difficult achieving an in take ra te o f a t least 0.2 m g AFDM/s, w h ich is th e requ ired level for fuelling a t th e Banc d ’Arguin (van Gils e t al„ 2009). Thus, carry ing capacity o f tropical in tertidal ecosystem s for m igratory shoreb irds m ay be se t by th e spring food d ensities th a t rem ain a fte r a long w in te r o f in tense p red atio n by shoreb irds exerted on m olluscs.

Acknowledgements

W e a re g ra te fu l to th e Parc N ational d u Banc d ’A rguin (PNBA) for th e ir p e rm iss io n to w o rk in th e p a rk a n d to u se th e ir facilities, w ith special re fe ren ce to L em haba Ould Yarba a n d th e PNBA c rew a t th e Biological S ta tion in Iwik. A m adou A b d erah m an e Sail an d A nne D ekinga h e lp ed w ith sam pling , T hom as O udm an a n d Jim d e Fouw took care o f som e logistics, w h ile Eva Kok, E m m a Penning, Els van d e r Zee a n d Lisa K lunder d e m o n s tra te d g irl-p o w er in th e lab o ra ­tory. T hom as O udm an an d five an o n y m o u s re fe rees c o m m en ted v e ry con stru c tiv e ly o n earlie r versions, w h ile Dick V isser po lished th e figures. This w o rk is p rim arily fu n d ed by a n NWO-VIDI g ran t

MV. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34 33

864 .09 .002 aw ard ed to JAvG, b u t also by an NWO-WOTRO g ran t W .01.65.221.00 to TP.

References

Alongi, D.M., 1990. The ecology of tropical soft-bottom benthic ecosystems. Oce- anogr. Mar. Biol. 28, 381^496.

Altenburg, W., Engelmoer, M., Mes, R., Piersma, T., 1982. Wintering Waders on the Banc d’Arguin, Mauritania. Report of the Netherlands Ornithological Expedition 1980. Stichting Veth tot steun aan Waddenonderzoek, Leiden, The Netherlands.

Andresen, H., Dorresteijn, I., van der Meer, J., 2013. Growth and size-dependent loss of newly settled bivalves in two distant regions o f the Wadden Sea. Mar. Ecol. Prog. Ser. 472,141-154.

Beukema, J.J., 1974. Seasonal changes in the biomass o f the macro-benthos of a tidal flat area in the Dutch Wadden Sea. Neth. J. Sea Res. 8, 94—107.

de Fouw, J., van der Heide, T., Oudman, T., van Gils, J.A. 2014. Structurally complex seagrass beds obstruct sixth sense in a molluscivore specialist (in preparation).

de Goeij, P., Lavaleye, M., Pearson, G.B., Piersma, T., 2003. Seasonal Changes in the Macrozoobenthos of a Tropical Mudflat. NIOZ-Report 2003-4. Royal Netherlands Institute for Sea Research (NIOZ) & WA Department o f Conserva­tion and Land Management.

de Roos, A.M., Schellekens, T., van Kooten, T., van de Wolfshaar, K.E., Claessen, D., Persson, L, 2007. Food-dependent growth leads to overcompensation in stage- specific biomass w hen mortality increases: the influence of maturation versus reproduction regulation. Am. Nat. 170, E59—E76.

Diawara, Y., Overdijk, O., 2007. Wader count in the Banc d’Arguin National park (Mauritania). In: Diagana, C.H., Dodman, T. (Eds.), Coastal Waterbirds along the West African Seaboard, January 2006. Wetlands International, Wageningen, The Netherlands.

Dittmann, S., 2000. Zonation o f benthic communities in a tropical tidal flat of north­east Australia. J. Sea Res. 43, 33—51.

Dittmann, S., 2002. Benthic fauna in tropical tidal flats o f Hinchinbrook Channel, NE Australia: diversity, abundance and their spatial and temporal variation. Wet­lands Ecol. Manag. 10, 323-333 .

Duarte, C.M., Chiscano, C.L, 1999. Seagrass biomass and production: a reassessment. Aquat. Bot. 65 ,159-174.

Engelmoer, M., Piersma, T., Altenburg, W., Mes, R., 1984. The Banc d’Arguin (Mauritania). In: Evans, P.R., Goss-Custard, J.D., Hale, W.G. (Eds.), Coastal Waders and Wildfowl in Winter. Cambridge University Press, Cambridge, pp. 293-310.

Ens, B.J., Piersma, T., Wolff, W.J., Zwarts, L., 1990. Homeward bound: problems waders face w hen migrating from the Banc d’Arguin, Mauritania, to their northern breeding grounds in spring. Ardea, 1—364.

Ferguson, N., White, C.R., Marshall, D.J., 2013. Competition in benthic marine in­vertebrates: the unrecognized role o f exploitative competition for oxygen. Ecology 94 ,126-135 .

Folmer, E.O., van der Geest, M., Jansen, E., Olff, H., Anderson, T.M., Piersma, T., van Gils, J.A., 2012. Seagrass-sediment feedback: an exploration using a non­recursive structural equation model. Ecosystems 15,1380—1393.

Gurevitch, J., Morrison, J.A., Hedges, L.V., 2000. The interaction between competition and predation: a meta-analysis o f field experiments. Am. Nat. 155,435—453.

Hagemeijer, W., Smit, C., de Boer, P., van Dijk, A., Ravenscroft, N., van Roomen, M., Wright, M., 2004. Wader and Waterbird at the Banc d’Arguin, Mauritania. January 2000, WIWO Report 81. Foundation Working Group International Waterbird and Wetland Research (WIWO), Beek-Ubbergen, The Netherlands.

Herry, A., Diouris, M., Lepennec, M., 1989. Chemoautotrophic symbionts and translocation o f fixed carbon from bacteria to host tissues in the littoral bivalve Loripes lucinalis (Lucinidae). Mar. Biol. 101, 305—312.

Hirakawa, H., 1995. Diet optimization w ith a nutrient or toxin constraint. Theor. Popul. Biol. 47, 331-346.

Hodgson, A.N., 2010. Reproductive seasonality o f southern African inshore and estuarine invertebrates — biogeographic review. Afr. Zool. 4 5 ,1 —17.

Holling, C.S., 1959. Some characteristics o f simple types o f predation and parasitism. Can. Entomol. 91, 385—398.

Honkoop, P.J.C., Berghuis, E.M., Holthuijsen, S., Lavaleye, M.S.S., Piersma, T., 2008. Molluscan assemblages o f seagrass-covered and bare intertidal flats on the Banc d’Arguin, Mauritania, in relation to characteristics of sediment and organic matter. J. Sea Res. 60, 255—263.

Kraan, C., van Gils, J.A., Spaans, B., Dekinga, A., Bijleveld, A.I., van Roomen, M., Kleefs tra, R., Piersma, T., 2009. Landscape-scale experiment demonstrates that Wadden Sea intertidal flats are used to capacity by molluscivore migrant shorebirds. J. Anim. Ecol. 78,1259—1268.

Leyrer, J., Spaans, B., Camara, M., Piersma, T., 2006. Small home ranges and high site fidelity in red knots (Calidris c. canutus) wintering on the Banc d’Arguin, Mauritania. J. Ornithol. 147, 376-384.

Lubet, P., 1994. Reproduction in molluscs. In: Barnabé, B. (Ed.), Aquaculture: Biology and Ecology of Cultured Species. Ellis Horwood Limited, London, pp. 138—173.

Mandai, S., Harkantra, S., 2013. Changes in the soft-bottom macrobenthic diversity and community structure from the ports of Mumbai, India. Environ. Monit. Assess. 185, 653—672.

Michaelis, H., Wolff, W.J., 2001. Soft-bottom fauna o f a tropical (Banc d’Arguin, Mauritania) and a temperate (Juist area, German North Sea coast) intertidal area. In: Reise, K. (Ed.), Ecological Comparisons o f Sedimentary Shores.

Onrust, J., de Fouw, J., Oudman, T., van der Geest, M., Piersma, T., van Gils, J.A., 2013. Red knot diet reconstruction revisited: context dependence revealed by ex­periments at Banc d’Arguin, Mauritania. Bird Study 60, 298—307.

Oudman, T., Onrust, J., de Fouw, J., Spaans, B., Piersma, T., van Gils, J.A. 2014. Digestive capacity and toxicity cause mixed diets in red knots that maximize energy intake rate (submitted for publication).

Piersma, T., 1982. Macrobenthic fauna o f the intertidal flats. In: Altenburg, W., Engelmoer, M., Mes, R., Piersma, T. (Eds.), Wintering Waders on the Banc d’Arguin, Mauritania. Stichting Veth tot Steun aan Waddenonderzoek, Leiden, pp. 49—66.

Piersma, T., 2007. Using the power of comparison to explain habitat use and migration strategies o f shorebirds worldwide. J. Ornithol. 148 (Suppl. 1), S45— S59.

Piersma, T., de Goeij, P., Tulp, I., 1993. An evaluation o f intertidal feeding habitats from a shorebird perspective: towards relevant comparisons between temperate and tropical mudflats. Neth. J. Sea Res. 31, 503—512.

Piersma, T., Rogers, D.I., Gonzalez, P.M., Zwarts, L., Niles, L.J., De Lima Serrano do Nascimento, I., Minton, C.D.T., Baker, A.J., 2005. Fuel storage rates before northward flights in red knots worldwide: facing the severest ecological constraint in tropical intertidal environments? In: Greenberg, R., Marra, P.P. (Eds.), Birds of Two Worlds: Ecology and Evolution of Migration. Johns Hopkins University Press, Baltimore, pp. 262—273.

Piersma, T., van Gils, J., de Goeij, P., van der Meer, J., 1995. Holling’s functional response model as a tool to link the food-finding mechanism o f a probing shorebird with its spatial distribution. J. Anim. Ecol. 64, 493—504.

Quaintenne, G., van Gils, J.A., Bocher, P., Dekinga, A., Piersma, T., 2010. Diet selection in a molluscivore shorebird across Western Europe: does it show short- or long­term intake rate-maximization? J. Anim. Ecol. 79, 53—62.

R Development Core Team, 2013. R: a Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.

Sastry, A.N., 1979. Pelecypoda (excluding Ostreidae). In: Giese, A.C., Pearse, J.S. (Eds.), Reproduction o f Marine Invertebrates. Academic Press, New York, pp. 113-292.

Schaffmeister, B.E., Hiddink, J.G., Wolff, W.J., 2006. Habitat use of shrimps in the intertidal and shallow subtidal seagrass beds o f the tropical Banc d’Arguin, Mauritania. J. Sea Res. 55, 230—243.

Schröder, A., Persson, L., de Roos, A.M., 2009. Culling experiments demonstrate size- class specific biomass increases w ith mortality. Proc. Natl. Acad. Sei. U S A 106, 2671-2676.

Silva, R.F., Rosa, J.S., Souza, S.R., Souza, P.W., 2011. Spatial and temporal changes in the structure of soft-bottom benthic communities in an Amazon estuary (Caete estuary, Brazil). J. Coastal Res., 440—444.

Smit, C.J., Piersma, T., 1989. Numbers, midwinter distribution, and migration of wader populations using the East Atlantic flyway. In: Boyd, H., Pirot, J.-Y. (Eds.), Flyways and Reserve Networks for Water Birds. IWRB, Slimbridge, pp. 24—63.

Stephens, D.W., Krebs, J.R., 1986. Foraging Theory. Princeton University Press, Princeton.

Urra, J., Marina, P., Salas, C., Gofas, S., Rueda, J.L., 2013. Seasonal dynamics of Molluscan assemblages associated w ith littoral soft bottoms o f the NW Alboran Sea (Western Mediterranean Sea). Mar. Biol. Res. 9, 645—660.

van de Kam, J., Ens, B., Piersma, T., Zwarts, L., 2004. Shorebirds: an Illustrated Behavioural Ecology. KNNV Publishers, Utrecht, The Netherlands,

van der Geest, M., Sali, A.A., Ely, S.O., Nauta, R.W., van Gils, J.A., Piersma, T. 2014. Nutritional and reproductive strategies in a chemosymbiotic bivalve living in a tropical intertidal seagrass bed (submitted for publication),

van Dijk, A.J., de Roder, F.E., Marteijn, E.C.L, Spiekman, H., 1990. Summering waders on the Banc d’Arguin, Mauritania: a census in June 1988. Ardea 78 ,145—156.

van Gils, J.A., de Rooij, S.R., van Belle, J., van der Meer, J., Dekinga, A., Piersma, T., Drent, R., 2005a. Digestive bottleneck affects foraging decisions in red knots Calidris canutus. I. Prey choice. J. Anim. Ecol. 74,105—119.

van Gils, J.A., Dekinga, A., Spaans, B., Vahl, W.K., Piersma, T., 2005b. Digestive bottleneck affects foraging decisions in red knots Calidris canutus. II. Patch choice and length of working day. J. Anim. Ecol. 74,120—130.

van Gils, J.A., Piersma, T., Dekinga, A., Dietz, M.W., 2003. Cost-benefit analysis of mollusc-eating in a shorebird. II. Optimizing gizzard size in the face of seasonal demands. J. Exp. Biol. 206, 3369—3380.

van Gils, J.A., van der Geest, M., Jansen, E.J., Covers, L.L., de Fouw, J., Piersma, T., 2012. Trophic cascade induced by molluscivore predator alters pore-water biogeochemistry via competitive release o f prey. Ecology 93,1143—1152.

van Gils, J.A., van der Geest, M., Kraan, C., Folmer, E.O., Jansen, E.J., Piersma, T., 2009. How the carrying capacity o f the Wadden Sea regulates the number of wintering waders at Banc d’Arguin. Limosa 82 ,134—140.

van Gils, J.A., van der Geest, M., Leyrer, J., Oudman, T., Lok, T., Onrust, J., de Fouw, J., van der Heide, T., van den Hout, P.J., Spaans, B., Dekinga, A., Brugge, M., Piersma, T., 2013. Toxin constraint explains diet choice, survival and population dynamics in a molluscivore shorebird. Proc. R. Soc. B 280, 20130861.

Wolff, W.J., 1991. The interaction of benthic macrofauna and birds in tidal flat es­tuaries: a comparison o f the Banc d’Arguin, Mauritania, and some estuaries in the Netherlands. In: Elliott, M., Ducrotoy, J.P. (Eds.), Estuaries and Coasts: Spatial and Temporal Intercomparisons. Olsen & Olsen, Copenhagen, pp. 299—306.

Wolff, W.J., de Wolf, L, 1977. Biomass and production o f zoobenthos in the Gre- velingen estuary, The Netherlands. Estuar. Coast. Mar. Sei. 5 ,1 -2 4 .

Wolff, W.J., Duiven, A.G., Duiven, P., Esselink, P., Gueye, A., Meijboom, A., Moerland, G., Zegers, J., 1993a. Biomass o f macrobenthic tidal flat fauna of the Banc d’Arguin, Mauritania. Hydrobiologia 258,151—163.

34 M.V. Ahmedou Salem et al. / Estuarine, Coastal and Shelf Science 136 (2014) 26—34

Wolff, W.J., Gueye, A., Meijboom, A., Piersma, T., Sali, M.A., 1987. Distribution, biomass, recruitment and productivity of Anadara senilis (L) (Mollusca: Bival­via) on the Banc d’Arguin, Mauritania. Neth. J. Sea Res. 21, 243—253.

Wolff, W.J., Michaelis, H., 2008. Do shorebirds graze down zoobenthic biomass at the Banc d’Arguin tidal flats in Mauritania? Estuar. Coast. Shelf Sei. 79,491^495.

Wolff, W.J., Montserrat, F., 2005. Cymbium spp. (Gastropoda: Mollusca) as bivalve predators at the tidal flats of the Banc d’Arguin, Mauritania. J. Mar. Biol. Ass. UK 85, 949 -953 .

Wolff, W.J., Smit, C., 1990. The Banc d’Arguin, Mauritania, as an environment for coastal birds. Ardea 78,17—38.

Wolff, W.J., van der Land, J., Nienhuis, P.H., de Wilde, P.A.J.W., 1993b. The functioning of the ecosystem of the Banc d’Arguin, Mauritania: a review. Hydrobiologia 258, 211- 2 2 2 .

Wolff, W.J., van Etten, J.P.C., Hiddink, J.G., Montserrat, F., Schaffmeister, B.E., Vonk, J.A., de Vries, A.B., 2005. Predation on the benthic fauna of the tidal flats of the Banc d’Arguin (Mauritania). In: Symoens, J.-J. (Ed.), Coastal Ecosystems of West-Africa: Biological Diversity-Resources-Conservation, pp. 43—59.

Wood, S.N., 2006. Generalized Additive Models: an Introduction w ith R. Chapman and Hall/CRC, Florida, USA.

Zwarts, L, Blomert, A.-M., 1992. Why knots Calidris canutus take medium-sized Macoma balthica when six prey species are available. Mar. Ecol. Prog. Ser. 83,113—128.

Zwarts, L., Blomert, A.-M., Ens, B.J., Hupkes, R., van Spanje, T.M., 1990. Why do waders reach high feeding densities on the intertidal flats of the Banc d’Arguin, Mauritania? Ardea 78, 39—52.

Zwarts, L., Blomert, A.-M., Wanink, J.H., 1992. Annual and seasonal variation in the food supply harvestable by knot Calidris canutus staging in the Wadden Sea in late summer. Mar. Ecol. Prog. Ser. 83 ,129—139.

Zwarts, L., van der Kamp, J., Overdijk, O., van Spanje, T., Veldkamp, R., West, R., Wright, M., 1998. Wader counts o f the Banc d’Arguin, Mauritania, in January/ February 1997. Wader Study Group Bull. 86, 53—69.

Zwarts, L., Wanink, J.H., 1993. How the food supply harvestable by waders in the Wadden Sea depends on the variation in energy density, body weight, biomass, burying depth and behaviour o f tidal-flat invertebrates. J. Sea Res. 31,441^476.