Embed Size (px)
Citation preview
Dt~p-~,euR~eurttll ~,,fl 41) '%~ ~ pp Jd~r,-~lo ltml Oqe,7-4~37¢t~3$e, fN)+lll III Printed m (.,real Br=t..n ~ 1~3 Pergamon Pre~.,. Lid
Tempora l persistence o f copepod species groups in the G u l f
Stream
C A R I N J . A S H J I A N * ~ a n d K A R E N F . W I S H N E R *
(Re~et~ed 12 August 1~91. m revt,wd form 14 April 1992, accepted 13 May 1992)
Abs t rac t - -The distributions ot 22 copepod t~pe~ ( 18 spec=e~ and the copcpodtte stage V of four of the species) across a transect ol the Gulf Stream near Cape Hatteras. NC. sampled m May 1983 u,=th a MOCNESS. v, ere anal,,zed and compared ~tth the dt,,tr,butzons found on a similar transect • ,ampled m September 1982 Copepod specie,, de, tnbut tons follo~cd physical charactcrtshcs closely and v, ere s~mdar for the tv, o samphng hines Speoes were found m d=screte environments , and specze~ dr, tnbu .on , , changed acro,,s the Stream with changing phy,,tcal properties Most of the copepod t)pe,, (IS) wcrc placed into four dlstrlbtmonal groups using recurrent group analysis. cluster anal),,t,,, and cxammatam o! ',peele', dl,,tnbuhon,, The September and May transect,', ~ere analyzed Intlcpcntlcntl). ~ct vlrtuall) the ,,amc group,, rc,,ultcd, nnpl,,mg that the,,e ,.pccle~ group,, are conxr, tcnt o,,er tm+c and ,,uggc,,tmg that a pcr,,i,,tcnt communtt ) ,.trutturc may be a general feature of the Gull Stream I he th,.tr~hutlon,, of the ,,pet,c,, group,, ~ere a.,,,ociated with phy,,tcal propcrhc*, acro,,', the Gull Stream. ~,,=th httlc ~v, crlap bctv, ecn grottp cnvmmmcnt , , In both Scpiclnber a l ld May i.ro'~.-',.lrcanl trends nl integrated al~utltl;.tncc,, ot two o| the ',pcclcs groups hfllov, cd Ihe paltcrn~ cxpct tcd it COXltlll|lOtl.,. cro~,'~-4tre,llll Itll%tllg protc~.~.c'~ had occurred Cold- W.ltcr ,,pcclC., had higher ,l['~tlllddtltC ~, ,It the llorlhcrfl ',.IdllOll,., ~,lth al~Ullddncc~, tlccrca.,.ing acro,,*, fl~c Stream to the ,~ottth W,trn |-~atcr ,,pccic,, abtmdantc, , followed the opl'~,atc pattern (lugh abtmdancc,, ot~ the S,trga',,,o Sea ',~dc of the Stream and Io~cr al~ttntl,incc,, on tim northern *,tile) Ccr t , ,n ,,pcctc,, or ,,pc~.m', group,, inay bc rchablc tl~tht,itor,, ol w,ttcr t.H~c and could bc u,,ctl a,, tr,itcl,, ol ~,~,llCl ii1,1~.% lllt~.l]lg dtro,,,, the (hill ,~lrC,llll
I N ' r R O D U C T I O N
TIlE Gulf Stream is usually rcgardefl as a biogcographical botmdary separating the cokl-water eutrophic Slope Water fattnal a.~semblages to the north and west from the warm-water oligotrophic S;trgasso Sea a,;,,cmblagcs to the south and east (GRIcz and HART, 1962; [lvaLauRr. 1964; BACKUS et a l . , 1970; JalIN and BACKUS, 1976; CHENEY, 1985). However. studies of zooplankton populations in the northwest Atlantic have noted the presence of non-endemic species in both the Slope Water and the Sitrgasso Sea. Their transport across the Gulf Stream hits been attributed either to an unspecified large-scale lateral mixing or to thc formation of Gulf Stream rings (e.g. GalCE and HAR'r, 1962; BOW,~tA,'~. 1971; DEEVr.V, 1971; WtraE et a l . , 1976b; DLEVEV and BkooKs, 1977; Cox and WtE,r,, 1979; JOYCE et t t l . . 1984). Smallcr-~calc physical procc',scs, also identified its agents
"Graduate School of Occanography. Um~,cr,,ay ol Rhode i,,land. Narragan,,ctt, RI 02882. U S A "l'Prcsent addr~:w,: ()ceanographlc and Atmo,.phcnc Science*, Divl,,ton. Department of Apphed Science.
Brookhaven N,monal Laboratory, Upton. NY 11973. U S A
4N~
48-1. C J. AsHJtan and K F '~tSH~,eR
of cross-stream water transport, include shingle formation (Slope Water entrainment) and decay (e.g. LILLmRIDGE and ROSSBY, 1986: HrrCHCOCK et al.. 1986: GARFIELD and EvaNs. 1987). formation of streamers associated ~ith ring-Stream interactions (NoF, 1988). "'leakage" from the Stream at depths beio~ the thermochne (SHAw and ROSSBY, 1984), and water transfer at meander crests and troughs (BowER etal.. 1985; BOWER and ROSSBY. 1989; HOGG, 1986). Therefore the Gulf Stream functions as a leaky membrane (BowER et al., 1985), permitting some cross-stream exchange of water and plankton populations.
Despite the importance of the Gulf Stream as a biogeograph,cal boundary, there are few in-depth studies of the zooplankton within the current itself. Most studies of Gulf Stream zooplankton populattons have been done near the Straits of Florida, g here the current is still discrete (e.g. MOORE and O'BERRY, 1957: ROEHR and MOORE. 1965: WORrHI.',GTON. 1976). Earlier studies in areas north of Cape Hatteras have usuallv been hmlted to a few samples, taken either on a transect of the area or in conjunct~on with investigations of Gulf Stream rings, and have concentrated on species assemblages in the Slope Water and Sargasso Sea (GR[cE and HART, 1962: DEEVE¥. 1971: DEEVE¥ and BROOKS. 1971. 1977: Cox and WtEBE, 1979).
Strong associations between spccics distributions and physical characteristics and a consistent community structure (species groups) were found across the Gulf Stream on a September 1982 transect (ALLtSON. 1986: WlSHNER and AtLISON, 1986) Despite the changing environmcnt, copcpod species di~tnbution~ could bc grouped into a few distinct patterns at all points along the transect using recurrent grt~t,p analysis (FA(;ER, 1957). suggesting a stable commtinity structure (WIsHN~:r and At I ison, 1986). Furthcrmorc. the abundance trends of some copcpod species groups across the Gulf Stream follo~cd the patterns expected if continuous cro,,s-~tream mixing procex,,cs wcrc operating (WlbtlNI R and Al.l ISON, 1986). Cold-water specie,, wcrc in lughcr abundances at the northern stations, with abundance,,, decreasing across the Stream to the south. In contrast, ~a rn l -
water species abundances h~llowcd the opl~osltc pattern, with highest al~untlance~ on the Sargasso side of the Stream. ! Iowcvcr, tlu,, study was hmitcd t~ a ,,ingle transect ol the Gull" Stream.
The present study analyzed coPCl~od species th,,trd~ut,ons and ct}nmumity structure lrom a spring (May 1983) transect of the (;ull Stream near (_'ape t lattcras, in comparison with the fall transect (September 1982) at the same h~cation (73"W), previously described by WISHNt:R and AH ISON (1986) and At.l.lSON (1986). "l'hl~ paper dc~cribcs the May results, includes further analyses ol September data, and prcscnt~ comparisons between the two times. Specilic tlttCSttons include: ( 1 ) Is there a con,,i.~tcnt conlmututy structure m the copepod assemblages across the Gulf Stream, and was thts structure the same at the two timcs? (2) Do copepod species" groups abundances follow changing environmental characteristics across the Gulf Stream. and arc these associations observed consistently ovcr time'? (3) Which copcpod species arc most usclul over time as water mass indicators? (4) Arc there patterns m total abundance of copcpod species groups across the Gulf Stream, and what arc the m~plications of these patterns for crt~,,s-strc,m~ tran.sport of plankton?
METHODS Sampling
Samples were collected from two transects of the Gulf Stream. done in September 1982 (WlsHNER and At.t.iSON, 1986; ALt.tSON. 1986) and in May 1983. off Cape l-latteras. NC
Temporal per'~tstence of copepod t~ pes m the Gult Stream 4 8 5
Tab& 1. Samphng data for the three stattons conducted m May 1983 See Wtst-t~'E,~ and At t t sox ( 19861 for September 1982 samphng mJbrmatum
Start latttude Start longitude Date Local time Local time Max. depth Region Operanon (°N) (°W) (lqS3) Start End (dbar)
Northern edge MOCNESS 15 36 07 11 73 17 79 19 Nla,, l I 117 12 11 200 MOCNESS 16 36 06 04 73 21 86 19 May 13 53 16 21 1{XXI MOCNESS 17 36 05 8{) 73 23 S8 19 Ma? 21 42 22 44 2{H) MOCNESS 18 36113 78 73 22 56 211 Ma,, 0 l I 2 54 q3l)
Hydrocast 36118 114 73 11~ 91 It, ~ May I~ 42 5t)l
Central Stream MOCNESS 19 35 56 4q 73 12 Ill 211Ma~, Ill 46 10 46 IOIII) MOCNESS 211 35 56 53 73 I 1 711 211Md} 15 (Iq 15 119 21111 MOCNESS 21 35 56 95 73 13 I111 211Ma~, 21 20 21"29 Iq3 MOCNESS 22 35 56 93 73 12 93 20 May 23 42 23 42 11J110
H)drota,,t 35 57 12 73 (I t) 95 211Ma} l~ 3t~ 773
South edge MOCNESS I11 35 43 72 73 01 96 12 Ma~ 23 40 I 28 200 MOCNESS I 1 35 44 115 73113 32 13 Ma} I 18 4 116 I(Xl(I MOCNESS 12 35 44 211 73 OI 38 13 Ma~ 9 34 12 112 I(~1(I MOCNESS 13 35 42 33 73 05 24 13 Ma,, 13 23 14 12 200
tlydroca,,t 35 44 45 73 Ill h"; 13 M,I} 7 117 q28
(Table I ). These 2 month,, were cho,~en to provide the maxlnuml contrast in temperature gradient aero,,,, tile North Wall (between the Gulf Stream and tile Slope Water), which might atfcct plankton di',persion and distributions. The tran,,ccts co11,~v~ted of three stations, located 211--.4(I km apart, and oriented such that tile transect was perpendicular to tile mean direction of life Gt, lf Stream. The stations were sampled intensively, from 0 to Ill(10 m. at three location,,: a North Wall station (located where the 15°C i,,othcrm was at 211(I m). a Centr;,I Stream station, and a station at tile Sarga,,so Sea edge ol the Stream. The transect was Ioc;,tcd across a fairly straight, non-meandering ,~cgment of the Gulf Stream at both sampling time,, (Fig. I; WISHNI:R and AIt ISON. 1986).
At each station, h)ur MOCNESS (Multiple Opemng and CIo,,ing Net and Environmen- tal Sensing System: WH.BE el al., 1976a) plankton tows, us,ng 335 ltnl mesh, were conducted, two dr, ring tile day and two at night. The MOCNESS was equipped with temperature and pressure sensors and a flow meter. For each tmle period (day and mght) both a shallow, fine-scale tow, from 0 to 2(111 m at 25 m intervals, and a deep, coarse-scale tow, from 0 tt) I()()0 nl at 100 or 15(I m intervals, were done ( I(X)(I-850, 850--7()0, 700-550, 55()--41X), 4110-311(I, 3110--200.200-I(X), 100-0). resulting in eight d,screte net smnples from each tow. Zooplankton samples were preserved m 4% buffered tormahn tmmedmtely fl~llowmg collection, l lydrocasts were done at each of the three stations to obtain temperature, sahnity, density (sigma-t), oxygen, and nt, trient information. Downstream and cro.,,s-strcam velocities tor all stations were available lrom tile Pegasus program (t [ALKIN et al., 1985).
&,nph" a,talwt.~
Samples were analyzed fl~r the abundances of 22 copcpod types. 18 species and the copepodite stage V of fl~ur of these species. The same copcpod types were enumerated
486 C J ASHJIA~ and K F ~'(sa,'.,Ea
from each transect to maintain consistency between the two sampling times (ALLISON. 1986: WISHNER and ALLISON, 1986). The species selected were common in the samples. had different depth distributions, and represented 1_L-95% of the adult copepods in a sample, at least for the September 1982 samples (ALLISON. 1986). Copepod species gere identified according to the keys ot S i eur (1932), Rose (1933]. HULSEMANN (1966). and Owre and Fo~o (1967). Anal)sis was done by first splitting the sample down to a fraction ~,,here at least 2(XI of the target copepod t)pes were present. The counts were then extrapolated to No. animals/1000 m 3 according to the number of splits and the volume of ~ater filtered during that particular tow. Ninety-six samples were counted from the September transect and 83 samples from the May transect. For the upper 200 m. onl,~ the data from the fine-scale sampling were used.
Total integrated abundances (animals m-Z) over the 1000 m sampled were calculated for each species at the three cros,,-stream locations from both day and night for both data sets The fine-scale samples (25 nl Intervals) were used for the top 200 m. and the coar',e-,,cale ',ample,, ( 100 and 150 in intervals) were used for the rest of the water column.
Statistical analv.~e.~
Species abundances (animals/10(}0 in ~) I'roln all samples (84 for Scptenlber and 83 tot May) were compared for each ,,pecics bct~een September and May using the Mann- Whitney U test (SOKAI. and Rotn r , 1981'. ZAK. 1984). Multivariate statl,,tlCS ~ere employed to determine grouping of copcpod types. In these analyses, till abundances < 21)% of the n~e(.llan al')tmdatlcc of that species were counted as "IF" for statistical
smlphlicatlon and to emphasize the major dlstril+ut,ons (McGow,xN and WAI KLr. 1979, AI I.IbON, 1986: WI,hlINI,R and A I I ISON. 1986).
Recvrrent group allalysi~ ([:AC,t g. 1957; FA(;I I~ and Me. GO'WAN. 1963) was used to place ,,pccles into group,, on the bils~S ol co-occurrence using prescncc/al)senc¢ data. An moles ot ,ltfinity, /, was calcuhltcd Io~ each species pair according to the formula in FA(;I r and Mc GowAn (1963). The nmmnunl vahle ol / bch)w wluch the species were not considered to have strong altinity wan 0.5 (WlSllnVr and AI.IISON. 1980). Groups were formed between those species that had indices ol alfinity (I) >0.5 lot all pairwlse combinations. The degree ol as,,ociatk)n between species groups or between groups and individual species was determined from the percentage of pal rwise conlbnmtions of species froin that particular group that h:_ld sigmlicant indices (Jl affinily with another group or species. Recurrent group analysis lot the Scpteml~er data was presented in ALl ISON (1986) and WisiINl'r and ALl ISON (1986).
Copepod species groups also were i(.lentilied usnlg cluster analysis. A h,erarchlcal agglomerative, polythetlC cluster analysis was done separately for each tran.,,ect using the clu.ster procedure ol the Statistical Analysis System (SAS) (PIELOU, 1977; SAS [NSlIIUIL ['~C , [985: KrLas. 1989). Copcpod species were clustered using Spearman's Rho between species pair.,,, rather than differences ol abundance. The T R E E procedure was used to draw dcndrograms ol the clusters, based on the distance coelIiclent bet~een species calculated by the cluster procedure (SAS INS Ill U I r. l'qc., 1985). The dendrogr,mls ~erc then ~ Isually e,~amincd to identify the groups of species.
The species groups described by the cluster :lnal)sls and the recurrent group analysis were compared independentl) for the two data nets to describe one net of groups for each samphng time. If dlflcrcnt group compositions were dciincd by tile two statistical analyses.
l c m p ~ r a l p c r , , l , , t c n c ¢ o t c ~ , p c r o d t x p c , , m t h e G u t : ~ . : r c . am ~t, 'q"
+ + - f + + . . ~ : R ~ , + -+
t l B I ,%.~lclhl'., m h , t t , . d t t n . tg¢ +t:.l tilt." ,,.tmt',htl,.t. ,Lrc,t l m m M,t',. I'~,'n ~, ( ,q+,¢ II, t t l , . . .r, I , . "~( l', l ~ . , d ~ . d in t h e ]+++¢+k I It+ II , l l | t1111¢ ( J t i l l "~[It+,ll'l! i',, llt+ h g h l t . l ~.t)lt Hk.d ,it t+.l t. \ t~_l>thn~ ,t~.zt~n,+ lilt_ ll~tll'.-
I h u t h l c c , , t , l | l t l l ls, , l I e Ilhllk.,.<.j I+,~, Ih,,. \ ( ) .zltd ~ ( \ \ t + I t h x.k.d] ( ) (+ - I I t z , d "~tl,,.,t l l l
~t = %,H~,i','.L~ %'..,1 ",ldu )
Temporal persistence of copepod types m the Gulf Stream 489
then the distributions of the individual component species across the Gulf Stream were compared to insure that species grouped together had common distribution patterns. Copepod species groups for the September transect were redefned [from the groups of ALUSON (1986) and WtSHNER and ALLtSON (1986)] based on the new statistical analysis. The species groups from each sampling time were then compared with each other.
Total integrated group abundances (animals m -2) were summed for each group at the three cross-stream locations from both day and night for both data sets. Maximum group abundances (animals/1000 m 3) and integrated group abundances (animals/m:) were compared for each species group between (1) September and May and (2) day and night for each cross-stream location using Wilcoxon's signed-rank test (abundances of the group members were used as variables. SOKAL and ROHLF. 1981; ZAR, 1984). For tests of day versus night abundances, both September and May data were used m each category (day and night), with pairings between the same species and the same month, resulting in a sample size of eight for Groups l-3 (four species, each with 2 months) and of six for Group 4 (three species). Similar calculations were employed for the September versus May comparisons, with both day and night abundances used. Integrated and maximum abundances were compared separately for September and May between cross-stream Iocattons using Wdcoxon's signed-rank test and appropriately paired abundances of the component species (SOKAL and ROHLF, 1981; ZAR. 1984).
Kendall's coefficient of concordance (W: ZAR, 1984) wa'~ calculated, using integrated abtmdances for each species, to indicate if there wa~ significant agreement between the species in a group in abundance trends with cross-~trcam location. Day and night data were both tt~etl in the same test for groups wtth no dttfcrcncc~ in day-night integrated abundance. Correlation in cro~s-strcam trend,,, in grottp integrated abtmdance between September anti May wa', calculated u,,Ing the Spearman rank order cocflicicnt (ZAR, 1984)
RESU I,TS
En t'ironmt'ntltl dllllt
The physical structure across tile two transects was similar at both times, and typical of tile Gulf Stream (Fu6us'rER, 1960; SroMMrL, 1966). Temperature, salinity and density i,~opleths in the top 2(~) m approximately paralleled the surface, while isopleths below 200 m sloped downwards across the stream from the North Wall to the Sargasso Sea [Fig. 2; for September section,,,, see WISHNER and ALLIbON ( 1986)1. The top of the main thermoclinc sh)pcd down from 20(I m at the North Wall station to 550 m at the Sargasso Sea station. Characteristic "'Eighteen Degree Water" (ScHRO~:t)~R et al., 1959; WoR'rHtNG'rON. 1959, 1976) was found in a Ions at the Sargasso Sea statton and, with reduced range, at the Central Stream station. Density (sigma-t) in the surface layers of the September transect was lower (< 24.1; WISUNER and ALt,lSON, 1986) than in the May transect (24.1-24.5). The 27.0 isopycmd was located slightly deeper in the water column during the May transect than in September, especially in the core of the Gulf Stream and on the Sargasso Sea side.
The surfilce temper.'lture gradient between the North Wall station and the Slope Water to the north was greater in May than in September. The mean Slope Water surface temperature 55 km to the north of the North Wall station was 23.65°C during the September cruise and 15.25°C during the May cruise (tlAt.KL~ et al., 1985). The surface
"1'(~ C . J . A~ ,HJ I~ and K. F WI$H",,ER
t O O .
2 0 0 .
3 0 1 .
410
51e
6410
7 0 0 .
BOO-
• ,,,., [],o.,.,,,
[]. ,.,$ o [] , . , . , . ,
[ ] , , , . t o , [ ] , ,
T e m l M r a t n r s ( " C ) S a l i n i t y ( p p t ) D e n s i t y ( s i g m a - t )
N C S N C S N C S I l I ! I I
,ooJ ! ~ il ~ + ! . . . . - . . . . . tOO- , +
. . . I \ \ . 3 . + " " F , \ ~ I ' ' ° * ' ° ' 1 - " \ \ " N ,oo.
"'1 \ ' ~ . 1 " ' ' " I \ 1 " '
'° ' I . ~ " i "' N . I ~ Ol l - f ~ ' ~ J leO"
[] ~,~6.5 [] J5.0. .15 49 ~ : ,27 $ [] 26 1-27
[] ~ , . , . 3 , s [] < . , [] ' , ,--., [] 2s t . , ,
[]. ,..o [] ~, ,.~s
.i:1
11
n_
Nitrate and Nitrite (IJ.M)
P+ C S
P h o + p h s l e (~NI)
N (" s ! ! !
l i e - : ;:
2OO -
3OO -
400 -
see-
6 1 o .
?O0 -
l l O -
900 -
lOOP
Osyxen (mil l .)
p,l C S
No t~uL
I t o o .
l e e -
. lee -
4 0 0 -
+O1-
t O 0 .
? l l -
O I l -
9 0 0 -
Ne Ue~ 1 0 0 0 "
O s . , - , , O.,s O o + . , o • =.zo o • . s.5 [ ] , 1.4 5
D - o - t , , 0 , , - , , 0 ' " ' ~ I-]<os 13+ , .~ 13~+.,,
[ ] , , . , . , , . , [ ] . , . , D ° "+" 13"+'+ Fig 2. Cro.*,.,,-.,,trc,ml .,,cottons o f env i ronmen ta l data col lected w i th hydro~.a.,,t.~ m M a y 19~3 The three ,, l : l t lon Ioeatio,.~ arc dc ' , lgnatcd N ( N o r t h %Vail). C (Cent r :d Sire,ro l l a , d S (Sarga,,,~o Sea ,,tde) No te that the (.l:lta do not ex tend down to [iXK) m at all Ihrc~: Iocatzon,,. See WI.MINI:R and
A t I.ISON (1¢)~(+) f o r S e p t e m b e r ~,ccttor'l,,
Temporal pcrststence of copcpod t~pe,, ,n the Gull Stream 491
temperatures at the North Wall stations, located 20 km from the Stream axis. were 28.0 and 25.5°C for September and May. respectively. This resulted in a mean temperature difference between these t~o locations of 10°C (0.18°C km -1) in May but only 4.35°C (0.08°C km -]) in September (HALKIN el al., 1985). In contrast, there was httle difference in the surface temperature gradient between the North Wall and Central Stream stations at the t~o times. Very warm (>25°C) surface water extended across the top 50 m of the September 1982 temperature section (WISHNER and ALLISON, 1986), while in the May 1983 temperature section the surface water was cooler (24.88°C) at the Sargasso Sea side station (Fig. 2).
Nutrient sections from May were similar to the September sections (WtsHNER and ALLISON, 1986) with nutrient distributions following the sloping isotherms across the Stream (Fig. 2). Lowest concentrations were found near the surface with increasing concentrations at depth. Lowest oxygen concentrations were found in the deep core (225- 325 m at North Wall) and along the mare thermocline across the Stream. with high concentrations found at depth in and below the main thermochnc and in the Eighteen Degree Water.
The do~nstream velocity field in May was "'t}pical'" for the Gulf Stream (HALKIN and Rossini. 19~5) [Fig. 3; see WISIINER and ALI.ISON (1986) for September sections].
uq t.
t,.
Downstream Velocity (cm/s)
N C S
Cross-Stream Velocity (cm/s)
N C S • I l ~
3OO-
4011-
5O0-
600-
700.- , • , ,
100(] i
] >200 [ ] 81- 100
[ ] 151 - 200 [ ] 6 1 - 8 0
[ ] 126- 150 [ ] 41 . 60
[ ] !01 • 125 [ ] 2 0 - 4 0
] O - 10 [ ] (. 0.1) - (.1o)
[ ] 11-20 [] (.11).(.20)
0 • > (- 20)
] Ig .~ Cro.~,.-,,tr¢;it11 '~¢Ltlt)n'.. ~.ht)~.~.lllg the do~.nMruanl ,llltI cro,..,-,.trc.tm vuiocK~ Iluld,. for M,t~ 19~3 .~t,lta)n,, dc,,ign.ttton,, (N. C ,rod S) dr¢ a,, dc,.~rtb~:d for [-tg 2. Section,, are. compo,,it¢,, o! the
M,.~ I ' ~3 (A) .rod M,ly 19~3(B) ~¢lo~:tty field,. (Rcdr,t~ n from l l~ t ~.r,, c l a l . 19~5)
492 C J AsHJta.~ and K F WISHNER
Maximum downstream velocities ( 150-200 cm s - i) were found from 0-150 m at the North Wall station of the Gulf Stream. Downstream velocity in the core of the Gulf Stream ranged from 100-150 cm s -t and decreased with depth. Velocities were greatest at the North Wall station and least at the Sargasso Sea side of the Gulf Stream. The cross-stream velocity field was similar to the mean Gulf Stream cross-stream velocity section derived by HALKIN et al. (1985). except that the magnitude of the cross-stream velocity toward the north in the core of the Gulf Stream was greater in the May section ( ~ 10-20 cm s- l) than in the mean ~elocity section ( ~ 0 - 5 cm s- t ) .
Copepod species groups
Cluster analysis and recurrent group analysis produced similar groupings. Species characteristic of Slope Water ( Calamts finmarchicus. C. finmarchicus C5. Pleuromamma borealis. Rhincalan.s nasutus, and Metridia hwens; DEEVEY and BrooKs. 1977; GrlcE and Hart . 1962) were clearly separated from other species. The vertically migrating Ple.ro- mamma spectes, with the exception of P. borealis, were also well separated from the non- migrating species. Species grouped together were generally at the same depth ranges and exhibited similar behavior (i.e. diel vertical migration: Table 2).
Group,; defined using recurrent group analysis from the two transects had some differences. Twche copepod types were grouped similarly from the two sampli,lg times: ho~evcr, eight type,, were grouped differently in the two analyses (Fig. 4). and two ~pccics wcrc con~i~tcntly not placed ill species groups (Neocahttrtl,~ gracilis and P. borealt~} Thc groups delincd for tire May analysis were more biologically meaningful in that associations ol specie., w~th similar behaviors were |denttfietl. For example. Plettrotllamnta pi.w~t, a vertical migrator, was associated will1 other vcrtically-n|igrating l'letrromatn.ta species (Group 3) in the analysis of the May tlat:l but was tmassociatcd with those species in the September analysis l..ctc .tia/htvtcorni.~, a vertical migrator, was associated with shallow- living, non-tmglatmg species in the September analysis, but more logically grouped with N. gracih.s and Ieht.cahtntt.~ cornttttL~ as an association of "'ubiquitotl~" species in thc May an,dy,,is.
The root|trent species groups were more distinct in the May data than in the September data. as ind|catcd by the lower levels of association between all groups. For example, in September. Group 3 was associated at the 3,~'V,, level with Species Pair 2 (Group 4 m the May data), at the 411';/. level with Group 1, and at the 5% level with Group 2. flowcvcr, in the May analysis. Group 3 was associated only at the 16.6% level with G,'oup 4. at the 6._~ V,, level with Group I, and not at all with Group "~
Smfilar species groups occurred at the two sampling times according to the cluster analysis (Fig. 5). in contrast to the results of the recurrent group analysis. Nineteen out of 22 copcpod types clustcrctl into the same groups in both September and May. Certain species were more closely associated than others, and these associ,ltions were consistent between both data sets.
Summary groups tor 15 of tire 22 copcpod types were defined based on the two grouping ;.Inalyscs and the species" thstrlbutions (Fig. 6). in compiling the [inal groups, the results from the two data sets (May and September) were compared and species" dtstributions and behavior were considered. Species not consistently :issociated with a specitic group between the two data sets were not placed in the final groups (Lucicutia chtttsi. L. flavicorni.s. Metridta hrevtcattda. Metridia vemtsta. N. gracili~. I'. boreali.¥ and R corntt-
Tab
le 2
D
txtr
tbut
ton~
o]
the
22 l
op~
7~od
tt p
e~ i
n .l
la~
1983
S
epte
mt)
er
1982
dt~
trtb
utto
n~ a
re p
rew
nte
d Á
n A
t I t
~ON
(19
86)
an
d a
~ gr
aph~
tn
Wtn
IIN
tR a
nd
At
i tS
ON
(198
6)
Lo
t at
loll
~ a
nd
tlo
le~
for
~ ht
t h t
here
~ e
re n
o o
~ er
hq;
i;l
a pa
rltt
tth
ar ~
lw~
te~
th~
trth
tatt
m ~
l tth
lhe
Sep
tem
lwr
dl~
lrlh
ullo
n tl
re d
e~tg
tlal
ed w
tth
all
a~le
t l~l
t ~
leth
an
abun
thH
it e
, (
~'o
amt~
lals
/ li1
110 m
a) h
~r l
he $
't']
ltem
ber
data
are
pre
~e~
lted
to
r t o
mpa
rl~
otl
( Jl
t lso
,¢,
1980
) St
gnlJ
lt a
tltl
t' g
real
er m
edtl
ln a
llun
thln
t ¢'
s or
e de
s#gt
lale
d l~
t, a
~hlg
gt'r
( P
~
11 It
S,
$h
mn
- II
hta
tlt't
~"
tt'i
t t.
r a
tta
r.h
int
e~ ]
or t
he w
et w
* b
etw
een
the
m,o
Itn
Wi)
Dep
th r
ange
D
epth
o1
max
mlu
m a
bu
nd
ance
M
axm
mm
M
edm
n
Sep
t m
cdh
m
Nor
th
Cen
tral
S
arga
,,so
N
orth
C
entr
al
Sar
g,r,
,,o
,tbu
nd,,
nce
abu
nd
ance
,t
l'~uJ
ld,l
llC
e S
peci
e,,
Spec
ie,,,
| l
ille
~V
,III
Sire
,til
l Se
,i ~%
,ill
Str
e,lm
Sc
,i (N
o/1(
1111
1 ill I
) (N
il /11
1111
1 Ill
I)
(Nil/
1111
111 i
ll ~)
gr
oup
( ah
mu~
It
,iv
2110
- IIX
N)
--°
31K
1-II
N10
551
1-71
111
--
4111
1-55
11
8211
27
1 I
]a
m;I
clrt
'hlc
lL~
Nig
ht
IIX
1-2I
NI
2111
1--3
1111
11
111-
125
175-
200
5511
-711
11
IIN
)-12
5 31
111-
930
5511
-851
1 71
111-
8511
('
ahlt
lu~
D
,ly
2111
1-11
1111
) 4(
N1-
5511
55
(1-1
1111
11
5511
-711
11
--
850-
1110
11
1451
77
54
]l
nm
arc
lmu
~
('5
Nig
ht
125-
2111
.1
411(
I-85
11
11-5
0 55
11-7
110
711(
)-,"
;50
25-5
11
5511
--931
1 55
11-7
511
Cuh
mu~
l)
ay
511-
175
75
-21
1()
15
11-2
110
1(11
1-12
5 12
5-15
11
175
1763
92
15
2 te
nlm
otnt
.~
2111
1-31
111
7111
1--8
511
Nig
ht
75-1
75
74-4
00
75-2
00
75-1
1X)
75
--1
11
0
1111
1-12
5 2(
111-
3(1(
I
('ah
mu~
D
a)
75-1
75
7.~
-21
10
51
1-11
111
75-1
1111
1(
111-
125
511-
75
648
86
70
tem
aco
rm~
( 5
15
11-1
75
Nig
ht
75-1
75
75-1
25
75-2
110
75-1
(1(I
7
5-1
11
(I
125-
150
15(1
-175
Ll
l(tt'l
llla
D,I}
12
5--1
511"
75
-511
75
-1(1
0"
2t11
1--3
1X1
3111
1--4
1111
75
-1(1
11
227
17
241
thm
.~t
2(11
1-31
111
175-
550
Nig
ht
100-
-3(1
11"
1111
1-85
0 15
0-55
0 21
11)-
3110
12
11-1
511
2111
1-31
~ 55
0--8
511
7111
)- i !
11 I(I
Lu
clcu
tm
Da.~
I)-2
5 75
-711
11
25--
511
125-
1511
11
111-
121t
15
11-1
75
18,5
27
89
266t
/h
it It
ortl
ls
511-
175
850-
-111
011
75--
7~1
21K
)-551
1 N
ight
I)-
2111
1 11
-31K
1 I1
-111
1111
2.
'L-5
tl 25
--511
I)
-25
4111
1-71
111
4111
1-85
11
l.m
aiu
na
l)
,t.x
11
111-
175
511-
7111
1 15
11-7
(111
12
5-15
11
125-
150
175-
2111
1 iX
)2
47
86
ger
ain
t 21
11-3
(XI
$5(1
-11X
I N
ight
75
--11
111
25-7
1111
12
5-11
1111
1 75
-111
11
7.'L
- 111
0 15
11-1
75
3 t~
m
=1-
0 ¢; 3 4
-
('on
llnlle
d k,
~
lah
h,
2 ('
on
ttm
wd
4-
4-
Spec
ie,,
LIlt
I('ll
llll
otdh
J
Ih'tr
tdta
/)
Tel I
t lll
tdd
Met
ruha
lit
( t't
l~
%le
trtd
m
t't'/I
t/'11
(I
~,'ll
Nl|O
g'lll
lllllL
+,
?NIIl
Or
.~ttl
NItl
OC
algl
NIl?
,
mm
or
C 5
• ~+t
'O£'
lil(l
l| It
T
gra(
'ths
grac
ths
(3
Nor
th
Tim
e W
all
Dep
th r
ange
D
epth
o1
max
mm
m a
bund
ance
M
axm
mfn
M
etha
fl
Sep
t m
cdhm
Cen
tral
S
arga
sso
"~or
th
Cen
tral
Sa
rg,t,
,,,o
ahun
d,m
cc
,ll'~t
t I|d
d]l~
'e
ahul
ld,ll
tt.e
%
pt:¢
1~2~
S
trea
m
Sea
W
all
%tr
cam
S
ea
(N~)
111
1111
1 Ill
))
INo.
/IIW
111
III 1
) (N
o 11
1M11
1 Ill
+}
grou
p
Da)
51
1111
1U
1.112
1111
lt
W11
15U
" 75
-111
~1
5111
75
L111
1112
5 14
9g
125-
1511
N
tght
11
1175
11
1175
11
175
25-5
11
75-1
1111
25
-511
55
1117
1N1
151~
175
Da)
41
1111
1111
111
5511
-11K
111
7111
1-11
.1111
1 85
1111
1111
11
S511
-- 1111
+N1
8511
-111
1111
12
2 N
ight
55
1.11
9311
55
11-1
1111
4.1
3111
1--4
1111
71
10--
850
S511
1|1W
111
7111
1185
11
7111
1~50
D
a.~
40
11-1
111+
111
31.W11-41111 +
700-
850
7O0-
850
3111
1---4
1111
71
111-
8511
34
5 N
~ght
15
11-2
1111
55
1)-1
1+)(+
)(.I
7111
-851
1 L
75-2
011
7011
1850
71
w11
8511
31
W11
8311
D
a)
5511
1711
11"
-11.
1111
5511
55
1.11
1111
111
55U
--T
tK)
4111
~55
0 55
1117
00
29
NJg
ln
5511
1719
1"
4111
1171
111
-111
1118
511
55(1
171.
111
5511
171N
) 71
141-
8511
D
,t.~
1.112
5 11
175
1112
0O
5111
75
2.'r~
-50
75-1
1M1
14.9
64
511-
175
ItW
1117
5 2(
N)-
-3(I
P 4(
X11
550
Ntg
ht
1111
51)
(111
(NI
(112
(11)
25
-511
11
125
75-1
1K)
] 75-
20~)
12
5- !
5(|
Day
11
-25
(115
0 0-
-200
5O
-75
25-5
1J
5111
75
3841
1 51
1-- 1
50
Nig
ht
O-I
lK)
(1-1
25
0-2
00
5(
1175
11
125
125-
150
4(M
1155
0 D
ay
5111
1tM
) 25
-75
5(11
125
51.11
75
2:L-
5U
75-1
00
32¢)
2i
1111
41.X
) 17
5-2(
gl
400-
850
41.1
11-7
00
Ntg
hl
1117
5 11
121W
1 50
-111
0 51
1175
(1
125
75-1
(K)
1(X
11.4
110
41K
)-55
(l 17
5-21
X)
7111
1185
0 D
ay
5111
1111
1 25
-151
1 11
1125
51
1-75
25
-511
75
-111
11
694
2111
1---31
111
Nig
ht
1117
5 (1
1175
5(
1117
5 5(
1175
25
-511
75
-111
11
1511
--311
1)
41.1
11-.-
551|
53
(+8
I
15
lO
19
81)t
2
10
13
--
321
194
I
4()2
21
8 I
18
31
44
281
I
J+
'11
.,r.
I
IJh'
llroI
Hlll
Hnl
ll ab
donl
llla
h+
D,tx
75
-111
| 41
WI-
7tll
2l
l1-7
tlJ
N:g
ht
t1-5
50
0-17
5 83
11-9
311
21W
1-11
XlO
I~
leun
nnan
tma
Day
41
111-
1111
11
--
bore
ahs
Ntg
hl
75
-2tl
l --
P
h,ur
oman
lnla
D
ax
2t1/
-550
17
5-85
0 gr
aclh
~ 85
11-1
0Oil
I~h'
ll?O
Ill|ll
llllll
l I,
Isek
t
I~h'
ll¢on
llllll
lllll
uphi
as
Rhln
cala
n#+
s co
rmtt
tt~
RIim
~ ah
ltlll
s
II/IM
IIII.~
Nig
ht
11-4
1111
O
-TtW
I
75-1
1111
41
111-
5511
41
111-
5511
41
111-
-~ gl)
125-
150
4111
1- i 0
(10
0-11
1011
25
-50
75-1
00
7:g-
- IIW
)
--
55t1
-700
--
--
--
151t
-175
--
--
2.
~- 1
00
20~
1-30
11
3011
--41
141
200-
300
125,
-150
21
11-8
5|1
11-7
1111
25
-50
75-1
0tl
5O--
75
850-
101N
| 8
50
-11
1K
) 85
11-1
1K10
l)
ax
2(~1
-550
2
IK~
IOIK
I 2
5-5
0
2(W
1-3
0o
30
0--4
1~1
200-
300
2111
-851
1 N
ight
11
-125
11
-211
11
11-1
25
511-
75
25-5
11
75--
11K
I 15
0--2
00
400-7(W1
175-
7111
1 85
11-1
1111
11
Da~
20
11-7
1111
12
5-15
0 4O
t1-1
INW
) 41
11-5
511
5511
-711
11
4(10
--55
0 21
10-3
00
4111
-711
11
N~g
ht
75
-30
11
75
-85(
I 4
~-8
5(I
D
a)
75-1
25
0-11
1111
.1
25-5
0 15
11-1
75
100-
150
2IW
1-1I
WK
) 20
(1-1
(~10
N
~ght
11
-931
1 11
-75
11-2
5 12
11-t
1100
5t
1-75
10
0-10
00
0-1
till
I1
X1-
125
12i1
-150
75
-1tl
1
3111
-300
30
1MO
II
125-
150
31X
I--4
00
5 ~,
1-71
111
1111
1-12
5
Da.
~
2111
1-11
1111
1 50
-75"
7
5-1
11
0
711>
850
50-7
5 75
-101
1 55
0-7(
I)
850-
1111
11
Ntg
ht
125-
150
55(I
-III
IIII
1
11
1-1
75
83
11-9
311
7111
1--8
51t
1(11
1-12
5 41
111-
9311
71
111-
1111
1
4 ! 8
9
379
5184
4474
48
23
91
48
70
27
123 34
3
1428
19
3(
1t
1
935
264
39
78f
14
22
2
,-t
x2 .7 ,q "d S .q-
ffl
e_
.1--
496 C.J. Asl-~t~ and K. F. ~'ISHNER
I I Me~vad~ ~a~sta
[ Plc~o,,~,,~.a bonmh~ ]
S e p t e m b e r 1982
o_., ~ I ~ ,.-,~.,,/z,,~ ~=,,., ?air..;
NaMac~ ~ C5 C~ ~,~o,a,~ C3 N,oc~ &mc,~r C5
Gmun 3 P/~om~mma gvagt/u P l*womamm~ ~ q ~ u 38 % LNCiC~ISG fe~U4~
Gl-oun C ~ /v.mnuc&cm, CJ Metn~a l~ena RAlm:agam~r cora,alas Rh,~calamu ~ a t a :
May 1983
~33% Gruun l [ [ C~un4 j
i X" I""
J I7%
[ M¢wld~4s .--~ I
I Gn, un 2 CW.a,~, fin,n~chJcv.,
Ma~d~d ~ccna 66e~
Temporal persistence of copepod t)pes m the Gulf Stream 497
SEPTEMBER 1982
C finmarciucus
C finma~c~+c~s C5
M l~cens nas~tu.$
".,7U7
MAY 1983
R corn~t~ [-~R na~aus
I ~ ~.~C.fmmarch,c~ U t.~C.1"mmarOucus C5
,72; ++o"+
114 I
[.-.C tenuicornts I-4 C tenuicoreus C5
i t.~ L. geauna
i . - r N""°"
l i j I I I..~l., oval~
N graclhs
P gracihs
P abdommahs
i ~eki
~L cla. .
I'2 ' l ' O ' 'll ' '6 ' '4 ' '2 ' AvrraKe I)l,dllnl.e IIt, lwecn ( ' lusters
N. gracihs
N. gracihs C5
N minor
N mmorC5
,________~P. gracihs
t.~ p piseh
I V t.,, p. x~p~
~ n a tcnmcorms C5 C ICnu~COrl~5
_~ t.~p. boeealt~
L. cla,,~l
I~g 5
i . 1 . ! i i i | . 1 . I i , !
O 14 12 10 li 6 .4 2 0 Average I)l~bmce Between Clu~'lers
P,c',ult', ill the t~l+ clu,,tcr ,inaly,,c,, ill COl~cpod ,,l~CClc ,, lot Scptciubcr 1982 :llld M,iy 1983 Del|drograll! not tlrtlwll exactly, to ,,talc
tu~') For example, L flavicorms was eliminated from Group I because the distribution of the species was markedly different from those of other group members. L. flavicornis had both a deeper distribution and a different behavioral pattern (dicl vertical migration) than other Group I members.
Spectes atttl spectes g roup tlistribution3
In general , copepod distributions (Table 2) followed the same trends in May as in Sep tember (At t.iso~. 1986). Species distributions followed physical characteristics across the Gul f Stream, with some distribution pat terns paralleling the surface and others fi~llowing sloping isopleths across the Stream with changes in depth range from station to
Fig 4. Copcpod ~pcoc~ group~ dctcrnuncd by recurrent group ,maly~l.~ The numbcr~ linking boxc~ refer to the per cent a~oclatlon ol spc¢lc~ bct~ccn group~ (the per cent a~oci,mon wtthm a group is I(Xi%) (September
19,~2 ,m:dy~l~ redrawn from Wl%ll'qrR and Au ,so~, 19,gfl,)
498 C J ASHJIA% and K F WIbH",,IER
Group 1 Nutmocalanu.r ramor Nannocalanus manor C5 Neocalanus graohs C5 l.,uctcuna ovalts
Group 2 Ca/a~4afinmarcJucua Calanus finmarchtc~ C5 Metridta lucens Rluncalanus nasutus
Group 3 Pleuromamma abdommalts Pleuromamma gracdts Pleuromamma ptselu P leuromamma x~phtas
Group 4 Ca~anus tenutcorms Calan~ temacorms C5 Luczcuaa germna
Not Grouped Lucwutta claua~ Luctcut~a flavtcorms Metndta breucorms Metndta venuata Neocalanus grao&~ Ple ur omamma bore alts RhtncalantLv cornutu£
N C S I I ,°°l
2 0 0 [
3 o o - ~ . .
_400- ! ~5001 N ~ 6oo-~
7oo- N I 8 o o -
900- I
1000-~
' - - ] Group 3 • Night
F~71 Group .11 • Day
|~lg (~ ~LIIllllldl~, group i.Olll['l{IM[IOll~ ,1110 dlll'll,.|~lllhl|l, g roup dl~|lil'ql|lOll '~, d~.'ro~, [h¢: {;ull S[ICdlII
,,tation. Of the 22 copepod type,,, ,,even were Iound prinlarily in the upper 21~) m ( Nannocahm. ~ minor. N. minor ('5. ('ahmu ~ wmucorm ~. (". wm.corm,~ ("5. N. gracilL~ ('5. lmcictma gemina, l.ucicutm ovah.O. ',ix were strong vertical mlglator,, (Pleuromamma ahdominahs. P. t~oreah¥. P. gracilis. P. pi.¥eM, i '..rqduas and L. Jlark'ornis), and nine (C. Jmmarchic..~, C. f immuchic.s C'5. g, chmsi. N. gracdt¥, R. corn.t..~. R. mt.~utu.~, M. I, rericamla, M. hwen~ and M. veml~ta) inhabtted mid-depth,, ( >200 m) ol the OulfStream.
Members of Grot, p I were totJnd in the top 101)m across the Stream (Table 2, Fig. 7) and remained at approximately the same depth ranges a|t all three locations across the transect. Members ol this group (N. minor, N. minor (-'5. N. gracihs C5 and g. orahs) are shallow- living species from warm, high salin,ty water in the Gulf Stream (approximately 19-25°C, 36.1-36.5 ppt) and are not d id vertical migrators (FAK~aN, 1947; RoE, 1972, 1984; WISHNt~ and ALl tsox, 1986, AMBI.ER and MIt.LEk, 1987). The species did not co-occur as frequently in May as in September (in 46 and 28% of all samples in September and May. respectively); however, three out of the four group members were found together trequently. At most locations, there was a single peak in group :,bt|ndance (Fig. 8). usually located in water ol 25°C or greater (Ftg. 2; Wtsttxe~ and At.t Isox. 1986).
Group 2, composed mainly of Sh,pe Water species (C. fimnarchicu.~. C. f inmarchk' t ts (.'5, M. htt'ett.~ and R. ttaattttt.~; GRICE rind HARI, 1962; Rot:, 1972). was fi,t, nd in and below the thermoclinc (10-15°C, approximated by the sigma-t = 27.1)isopycnal) across the Gulf Stream. following the sloping isotherms and isopycnals (Fig. 7). The distributions of these species had broad depth ranges at the North Wall station (~2{R1-10(10 m) which narrowed across the Stream to the Sargasso Seat side (T;,ble 2). Co-occt, rrence of all group members was more frequent in September (in 48 and 35% of all samples in September and May,
Temporal persistence of copepod t}pes m the Gull Stream 499
respectively). Although some of these species may be vertical migrators ( e . g . C . finmar- chtcus), the diel migration pattern was inconsistent between the two times, with no migration in September but limited migration in May at the North Wall and Sargasso Sea (Figs 7 and 8). Night group abundance peaks were progressively deeper in the water column across the Stream from north to south (thts was also true for the day group abundance peaks in September because little diel migration occurred: Fig. 8).
Group 3. composed of four congeneric Sargasso Sea-Gulf Stream species (P. abdomi- nalis. P. gracihs. P. ptseki and P. xiphias; GRICE and HART. 1962: DEEVEY and BROOKS. 19771. showed strong dlel vertical migrauon. Day distributions extended from just above the thermocline to approximately 200 m (Table 2; Fig. 7) and were constrained to narrower depth ranges at the North Wall side compared with the Sargasso Sea side. so that the maximum depths of the vertical migrators were considerably shallower at the North Wall statton than at the Sargasso Sea station. This resulted in a dccreasing vertical migration range for deeper-living members of the popttlation across the Gulf Stream from south to north as the thernlocline shoaled (night abundance maxima were located at similar depths at all three locations). P. abdominahs and P. xiphias had slightly deeper ranges than the other two species, so that all four species did not always co-occur. For example, only three out of four members were ever together in the Central Stream during the day in May (Fig. 7; all four species were tound in 311 anti 50% ot all samples I11 September and May. rc,,pectively). At night. Group 3 species were mainly m the upper 200 m (Ftg. 8) and night dlstributton rangc~ ~crc simdar across the Gull Stream There wa,, evidcncc of rcvcrsc or no migration ol somc mdividttal~ trom all tour species, rcsttltmg m the nighttunc occurrcncc of Group 3 ,,pccic~ at depth (Table 2). The peak ntghttimc abttnd,mcc fl~r the group wit,, always located in thc top 125 m (Fig. S). Day di~tnbutton~ wcrc Ic,,s sharply dclincd than the night th~tributions (T,d~lc 2). being spread over a greater depth range, but pcak group abuntlancc,, for day and night wcrc wcll separated in the water colulnn (Fig. 8).
The distribution ol Group 4. conlt'~oscd of warm-water species (C. tentttcortti~. (:. tentttcornt.s (.'5 and l.. gemina; Rot., 1972: WtSIINI;R and At l.tSON. 1986), wits usually centered at 1(111--21}11 m in water ol 20--25"C across the Stream (Fig. 7) and thd not vitry acro~ the transect. Group coherence act'o~ the Stream wa~ good in both data ~cts (m 57 and 6(1% ol all ,,imq~lcs in September and May, rcspccttvcly). A ,,ubtle dicl pattern in the location ol thc abtmdancc maxin~um cxi~tcd such that at tivc ot, t ol six of the Iocattons, the day group abtmdancc nlaxtnlttnl was 25-50 f l l dccpcr than the night abundance nlaximunl (Fig. 8).
With the cxccption ot Group 3 (vertical migrators), dtstribtttional overlap between groups wa.,, mmtnlal, suggesting habitat partitioning. Group,, 1 and 4 had little overlap in the top 21111 m and Group 2 wit.,, found primarily below thc main thcrmochnc. The distribution of Group 3 (vertical migrators) overlapped complctcly with Groups 1 and 4 at night m the top 21111 m. but during the day Group 3 did not overlap significantly with other group,,.
Seven copcpod types wcrc not placed into copcpod groups because their distributtons or behavior were dissimilar to those of other types. Of these, onc species, !'. bore~di.s, was notable in that it occurred only at the North Wall station at both times./', boreah~, a Slope Water species, hits not bccn found m the Gulf Stream and the Sargasso Sea (e.g. Grace and | tARt. 1962; Wtsit'~t:r and ALI.ISON. 1986). The Gulf Stream functtons as a sharp boundary to the dtstrtbution of thi~ species, in contra,,t to the other 21 copcpod types cxamincd in this study.
5 0 0 C . J . As~UtAN -,nd K F WisH,.ca
(a) September 1982
106- 200. 300-
~,~ 4 0 0 .
$00-
i 60O,. 700-
800-
°°0-,
tOO&~
Group Day
N c s I I I
• °1
1 Night
N C $ I . A . . . . . '
I 0 0 ~ +j, 00,.I...~.,] • ~
300-I 400"I
$00 1 6001 .
7 0 0 ]
"°1
Day c
Group 2 Night
s N c I I I
3oo-1°°" ~ . ' ~ 2oo.
4 0 0 -
$ 0 0 -
6 0 0 -
700 - 800 ~
900. tO00- = - - +- -
Group 3 Group 4 Day Night Day Night
s c s N c s m c s ~ c s
J : q '°'1. . / , o , ~ ,o,1. . f : l " ' t . " " '~ "o 30o 1
i '0't -"x.. ~ '°°1 ' ° '1 '~ '0°1 ~ ' 0 ° 1 ' ° ° i
" ° 1 " " ' ° ° I ' ° ° ~ ,::::I , , : 1 : : ] : : : : 1
- - ' ]Four members present
[ ] Three members present
" ' ] T w o members present
F'Jg 7 [),l}, ,rod mght ,,umm,,ry copcpod ,,peeves group,, distribution dcro,,s the Gull Stream lot (a) September It;g2 and (b) May 1983. Statmns dc~lgnatu.ms (N. C and S) arc as m Fig. 2. Locations of the full group (,dl members prcscm) arc mdic,ttcd by the darkc,,t shading, with hghtcr tones represent,rig depths where some group mcmb,:rs were: ab,,cnt Note: September group d~sm- bulton,, arc dtffcr(:n! than those presented in WlhllNI K :rod ALliSON (1986). because of the
rcdchni t ,m of group members m this study.
T e m p o r a l p e r s i s t e n c e o f c o p e p o d types m the G u l f S t r e a m 5 0 [
(b) M a y 1 9 8 3
++ AD
[
D a y N C
1 0 0 -
20O- :
30O-
4 0 0 -
5 0 0 -
6 0 0 -
7 0 0 -
8 0 0 -
00O-
1000-
Group 1 N i g h t
s N C
too- 2OO"
300"
4OO"
S00"
600"
700"
800"
900"
I I a r ~ - ~
D a y
c I
. /
Group 2 N i g h t
N C I I I
I " 2 0 0 . ~ s
300-1
,,~ 400-4
' $00 -4
, i 6 0 0 - 1
700-4
800-I
900-4
1000-1
Group 3 D a y N i g h t
N C S N C ~
,o0. ! i ~:~
g ,o0. ~+++ " "
o $00 .
i 6 0 0 -
I~ 7 0 0 .
8 0 0 -
0 0 0 -
1 0 0 0 .
Group 4 D a y
N c I I I
100-
200"
3 0 0 -
400"
5 0 0 -
6 0 0 -
700-
800-
000-
II 00O-
N i g h t
N c ,5 , £ I I
,oo .I 400"'1
500-4
600"t
700-4
800-5
[ ] Four members present
[ ] Three members present
I ~ T w o members present
5 ( ) 2 C . J . ASHJIAS, a n d K F. ~, lSHNER
ta) September 1982
North Wall Central Stream Sargasso Sea
Group 1 • Ant~alb I III00 i i~
"
~ 200
P
8(X)
Nqlht ~ y I000
• Anhlulk I 111~0 n 3
+ I $(I)
N,ght D= 9 I000
• Anlmakl I 10410 m3
o * ,
200
~ 0
600
S00
P~ht Dly 1000
Group 2
G r o u p 3
Group 4
• Animal / I I000 II~ 2600 10l]O 0 I0¢0 2000
0 . . . . . . . .
l (XM)
• A n l m a k I I 0 0 0 m3 21X30 I000 I000 2000
0 ; . . . . . . . .
I + o.,i L I , • Animal-/ 000 ~ • Anlmmli/ 0@4 i10
3CIC I~I0 1500 3~00 I~O0 ?~0 ?~0 I~O0
0 ' * . . . . 0 " ' * + " + '
~. 2(m' 2(x)'
4(I} i .- ,
~X)"
~" so<J-
I(XM) Nq~t
4(l}"
NI<)"
NO0 +
Day 104JO
+tltht Day
• Anlmala / 1000 m3 ~o0 2SO0 o 2~o SoI~ 0 , i , - • - -t
Nltht I~y 10OO
P
• AnlmslL I 1000 3000 I ~ 0 0 I ~ 0 3000
Nq[h¢ l ~ y
• Animals I II14~ m3 200o lOOO 10oO 200o
0 + " ' • *
IN+ltht D*y 1000
• Anlmala I 1 0 N m3 IX]OD 7~00 7~00 150~O 0 ~ • t ,
"*00'
400'
600
Niltht D=y |000
# Animal.I / 100O m3 ~13o ZIOO Z~0O ~
!
400" 1
Ntlht Dey Ira30
F i g X T o l a l c.l.t) a n d m g h t . d ' m n d , i n ¢ c , , v, , t h d e p t h o f t l l c s u n m m r y ~.op, . :pod , , p e t : i t +,, grt~urP+, l o r ( a )
S c p t c n l b c r J g S 2 . m d ( b ) M a y ItJ,',;3 N i g h t a h u n d , m < : c s ( d a r k q + a d c ) a r c t o t h e h : f t s t d c o f c a , : h p a n e l
a n d d a > a l + u n d . , n ¢ c , , ( h g h t s h a t h : ) a r c o n t h e r l g h t N o l ¢ d d l + c r c n t ,d+t, n d . t n c c a x e , ,
Temporal persistence ot copcpud t~pcb m the Gulf Strc,~m 503
(b) May 1983
Group 1
Group 2
Group 3
Group 4
North Wall Central So'cam Sargasso Sea
# An/rash / leO0 m.; ~ o O 2 ~ 0 0 ~ 0 0 ~ 0
~ .
8OO
j Night Duy
I Anlnut t, / lOOO m3
° " " ~ l 200"
I
Nq~t I~y J 1000
# Anlmak / 10~4 m3 2ooo0 iocco 0 IO~O 1 ~ o o
o
200"
tOO"
600"
800 i i
1000 ~ N j ~ Day
• Animal*/1040 m3 lOgO O IOGO
0 " " " t '
~ ~ . -.,~
~ .
800 ]
Night f Day 101)~) -
• Animab I 1000 m3 300 150 0 |50 ~ 0
O" - " , . . . .
8 Animak / 1000 m3 ~0 IJO 0 DO
° ~ - - L - ' ' 1 • i
200" [
41111-
~ ,",,3jtht J ~ D.y I IOCX3
# Anlmsk/10G0 mJ # Anlmnk / 10GO m3 ~:~ I~O 0 I~JO0 Xx~o 3(x~O I~0 0 t~O 3~x)
2OO.
,~ 4(1(1"
NtKht Day 10oo
20(
6dx
# AnimaLs / I000 m3 I ~ 7 ~ 0 ?SOD 130~0 0 " ~ ~
~XJ"
I {]()0 Nl|hl Day
• Animak / IO00 m3 3GO0 L ~14X) 0 1.~0
o - . , i _ .
: I *t g" 80o
I ~ Nlg~ Day
INIO'
I
I Anlms&/10O0 m3 3000 )~0 0 I ~ 3CO0
o . . . . - ¢ i
Nlthl
• AnJmak ! 1000 m3
~ Y ICx~ NJO~ i D,y
5 0 4 C J. AsHJb~ and K F WtSH'~Ea
Day-night and between-cruise differences
Overall. both species and summary group abundances (integrated and maximum) were similar in September and May. and few differences were observed between day and night. Only five copepod species had significantly different median abundances between the 2 months (Mann-Whitney U test. P ~< 0.05; Table 2). Maximum and integrated group abundances at the three cross-stream locations were usually not significantly different between day and night or between September or May (only 3/24 and 5/24 comparisons for day versus night and September versus May. respectively, were significantly different: Wiicoxon's paired-sign test: Table 3a: Figs 8 and 9)
Cross-stream trends in abtmdance
Species group abundance (integrated and maximum) differences between cross-stream locations were confined to Groups 2 and 3, the deeper-li~ ing species groups. Integrated and maximum abundances of Group 2 were greatest at the North Wall station in both September and May (P = 0.02-0.05. Wilcoxon's signed-rank test; Figs 8 and 9; Table 3b) and abundances decreased across the Stream from the North Wall to the Sargasso Sea. Cross-stream trends in integrated abundance between the two times were positively correlated (Spearman's Rho = 0.60). In contrast, integrated and maximum abundances of Group 3 were greater at the Sargasso Sea than at the North Wall for both September and May (P = i).02. Wilcoxon's signed-rank test: Figs 8 and 9; Table 3b). Central Stream abundances resembled the Sargasso Sea in September and the North Wall in May. Cross- stream trends in integrated group abundance were positively correlated between the t~t) times (Spearman's Rim = 11.575).
Most cross-stream comparisons of integrated and lll;,IXinllllll group abundances for the shallow-living groups (Groul~s I and 4) revealed similar abundances across the current (Table 3b). Only two exceptions were noted Ior Group 1: the September integrated and the May maximum abundances were different between the North Wall and the Sargasso Sea (Figs 8 and 9; Table 3b). Exceptions for Group 4 were confined to the Central Stream and Sargasso Sea ct~nlparison: both integrated and maxinlum abundances for September but only integrated abundances h~r May wcrc dffl'crcnt between the two locations (Figs 8 and 9; Table 3b). Cross-stream changes in integrated abundance were negatively corre- lated for Group 1 between September and May (Spcarman's rho =-0.771) , probably due to the higher abundances of the group at the Sargasso Sea in May, but were positively correlated for Group 4 for the two sampling dates (Spearman's rho = 0.829).
Within-group species abundance trends across the Stream showed significant agreement for two of four groups from May and for one of four groups from September (Kendall concordance test, Table 4). Only Group 2, composed of Slope Water species, showed significant concordance at both times. The abundances of these species decreased dramatically across the Gulf Stream from north to south (Figs 8 and 9).
D I S C U S S I O N
The distributions of plankton species result from biological and physical processes acting in conjunction. Biological characteristics of the organisms (e.g. behavior, tood requirements, toler:mce to physical environment and reproductive success) respond to
Temporal persl~tcne¢ of cop¢~l t.~pes in the Gull Stream 505
SEPTEMBER MAY
Group I I000"
v II
'<JI ]14oo-
North Wall C¢~U~I Sue.am Sar|auo Sea
800'
600'
200'
0 ¸ Wall CenlTal S~L'a SatlPlmo Sea
Group 2 800
C
j '- ia_l No~'I Will Ccrcml ~ S~p.,um S¢1
800
6 0 0
4 0 0
2 0 0
0 n
NocthWmll C~mlralS~re~u'a ~ ' | w a o ~ a
(; rou p 3 1200
C IIXX) 1
.~,, 200 am~ 0
N ~ h W|ll ('enroll .~tc.mt .~u'l~.tto Sea
1200
1(700
800
60O
400
200
0 NucthWill ('r.,ntraIStrcam .¢~r|~=oSca
Group 4
0" 0 - Nueth Will C~trll St/urn .¢~rlI, auo S¢.a Noclh Wnll Cerltnll Stl'um SltlltU~ Su
ca Day [] Night
Fig t) "lot.'d day ,rod night integrated ;tbund,inc¢,. ({I-llllll)I|1) lot the tour SUlUlnary COllcrR~d specie, , g r o u p s Ior S e p t e m b e r al ld M a y Noghu ( d a r k sh , ldc) alld d a y (hgh t "-,h,idc) illtcgr;.ttcd
;.ihlllld,lllCC,s (No ,ll|llllals, I11 2} ,ire p lo t t ed Ior e a c h cro~,s-Mrc;.tnl It)c,lUl(lll | o r e d d l '.,pccic,*, g r o u p
physical characteris t ics of the wa te r col tmm (e.g. t e m p e r a t u r e , salinity or oxygen) and affect the depth and horizontal distr ibution. Diel vertical migrat ion may result in a species being .,,t.bjcctcd to two different physical and advcct ivc regime,, within a single 24 h per iod. Advec t ion may act to concen t ra te or dilute species abul ldanccs . "l'hc consistent a,~sociation
l al,
le 3
R
esul
ts o
l II
'th
o~
on'~
pm
red
mm
ph"
te,t~
lbr
~ op
epod
gro
up a
lmm
hutt
es
(A I
Dql
erem
es
betw
een
Sepl
cml)
er 1
9~
2 i
S ) a
ml
May
19
83
( M
) a
nd/t
ells
¢¢1
1 Ih#
V ( D
) m
id n
ight
(V
) Ill
legr
aled
(.V
o /1
1121
ond
lilll
lll/ll
llll
(.~,
'O./
I|X
XI
EII
;j ab
.nda
nce$
|te
re t
e~te
d tit
eac
h cr
os s
-~lre
all!
ll,t
llllO
II ]
llg
C~l
th g
rol+
p 11
sIng
the
abut
#dan
t t's
II I
Ih¢
{ olll
PO
llt'll
l ~l
ll't
It%
(1~
) ~
.~
~e
~ ~?~
)~1~
¢~n
~
~h~
~¢~r
~J1t
~d i
~nd
~?~
~ ~n~
H~
~u~
d~C
~
~ ~
~hre
~ ~ ~
i~i~
i~.e
t~
~ ~
i
]~
etlt
h gr
oul~
]or
eat
'h l
llll
llll
lll~
ll
llle
, ll
MIl
~
IhC
tlb
lllld
tlllt
t'~
ol
lhe
l olll
lJol
l¢'ll
l il)
('(
I('i
Bol
h dd
~ ll
lld
lll~
hl t
lbll
lldt
lllt
t%
lt t
'rt"
ll~t'd
lor
lhe
iat
lllt
htlg
lli
tle
{ (lll
llltl
rl~
olls
tl
ilil
llol
h 'W
lIll'
llltt
t'r t
lltd
l~hl
y tl
l)lo
ldal
lt
es I
Vt'r
e
tl~¢d
jor
Ih
e da
i-Il
l~hl
[ I
)llll
)lll'
l$ol
lj (.
~"
= .'~
Orl
h |$
till
Sta
. C
=
('¢l
llrt
d S
lrt'(
IIII
.~'
l~l ,
.~"
=
.StlY
gll$
$l)
.~et
1,5'1
~1 ,
"n
I "
= Il
ol $
lglll
]lt[
Oll)
(A)
hll
eg
rale
d
ah
ull
d,l
n¢
¢
(No
/m:)
.% I,
iMn
lun
l a
bu
nd
an
ce
(No
,'
1110
0 m
;)
{B)
Inte
gra
ted
,t
bu
nd
,mc
c
( No
/m
2)
Ma
xim
um
:tl
)ulk
l, u~
.'u (
No
/11
11
1 ,!
1 ~)
Se
pte
mb
er
Ma
y
Se
ptc
nlb
cr
M,l
y
S~
sM
D~
sN
S~
sNI
D~
sN
N
C
S N
C
S
N
C
S N
("
S
Gro
up
I
Gro
up
2
Gro
up
3
Gro
up
4
.~
n •
n s
ii
s ii
s
--
[1 s
I1
112
--
n
,, n
,, --
tl
s
ii
s --
n
s 11
II.
~
("
ns
ns
ns
O(1
5(D
>N
} --
n
s --
ns
--
ns
--
ns
II •
l')
$
n '~
II
~,
..
..
N
ns
ns
ns
ns
--
(111
5 11
112
--
IIO
2
0(1
2
--
ns.
0
(12
--
1
11
12
I1
115
C
(I 0
2 (
S >
M
I n
s (|
1.1
5 ~
S >
M
) n
s.
--
fl.~
, --
ii
s
--
(111
5 --
n
•
S (1
05
(S>
M1
n
s n
s n
s .
..
.
N
ns
ns
ns
O0
51
N
>D
) --
O
02
0
.U2
--
n
s (1
112
--
ns
1111
5 --
i)
• 11
.112
C
fls.
n
s.
ns
0 (1
2 (N
>
DI
--
ns
--
1111
2 --
ii
s.
--
11
02
S 1
10
21
M>
S)
ns
ns
ns
..
..
C
n s
n s
n s
n s
--
(} 0
5
--
l) 1
15
--
l) 0
5
--
n.s
S I|
•
I1 •
11
115
(S
~>
~
11
II
5
..
..
'.'.2,
t'3
>.
r.
.9
,r. .m
Temporal permstence ot copepod types m the Gult Stream 5 0 7
Table 4 Re~ult~ o f the Kendall Concordance tests. KendaU's coeffictent o f concordam e ( W ) ,~ as used to estab&~h agreement bet)* een spectes m a group m abundant e trends ~ tth cro~-~tream h~£ atmn Cohtmns contain sums o f the t ros s-stream rank.~ o f abundances ol the ¢omponent wec le~for each group. Kendall'~ W and the ~tgm]icance level Abundances )*'ere ranked )~ ttlt "'1"" betng the least abundant ( Sog.tL and R oti tl~. 19Sl ). All SLr "rumples'" (three
lo(attons, day a n d mght) ))ere used tn each concordance test
_ rdnk~
Species North Wall Central Stream Sdrgasso Sea group Da) Night Day Night Da) Night I,', Sigmfic,mce
September
Ma~
l IS 17 16 14 12 7 0 293 >11 2 2 23 IS 14 14 7 5 7 5 0 648 0 01 3 6 5 l0 5 14 I,~ 15 21~ 0 434 0 2{I 4 11 14 ,g 4 12 14 1) 479 >0 20
1 65 15 19 ,~5 1~5 145 o41I 02o 2 22 22 5 14 I 1 IO 0 ,~36 o Ill 3 12 5 ++ g 13 5 22 22 I1 823 00l 4 I11 I I ~ 7 t~ IS tl 4t~2 0 20
of a species v+ith particular envi ronmenta l condition,, results in a predictable dp, tribution of that specie,, in the Gulf Stream.
l l i o l o g i c a l - l d t y . ~ w a l ,.~.w~( ia t io . ,~
l )ur ing both Scptctul+cr 1982 and May 19,~3. the Gull Stream could bc divided into two c n v . tmmcntal ,,one,,: ( I ) The .,,urtacc down through the main thcrnloclinc, typical of the Ul~pcr Sarga,,,,o Sea. The lower boundary of thi,~ region was constricted i,.,, the thcrmocl inc ,dloalcd frt+m ,,outh to north, but the top 211{I n| tff the Gulf Stream showed little change in gnvilt}nnlcnt across the scction~,. (2) A Slope Water +nvlronn1¢llt c×tcnding into the Stream front the north below the main thcrmt}clinc, becoming constricted fr<.+nl the upper bt~und,u'y a,, the thcrntochnc d c c p c n c d from north to ~outh. Thc,,c region,, wcrc iml+ortant (or the distribution ol dd lc rcn t plankton type,, and copcpod ~,pcclc+,, group,,.
The dr4 ributhm,, ol + the COl+cpod specie,, groups wcrc strongly a~so¢iatcd with character- istic cnvironntcntal condit ions acro,,s the Gull Stream. G r o u p , , I and + wcrc a,,,+o¢iatcd with the top 2(}11 m of the t, ppcr zone of the Gulf St ream. where isoplcth,, extended parallel across the section. G r o u p 3, the vertically migrating group, wits a,,sociatcd with the whole upper zone. G r o u p 2 was a.,,socmtcd with the lower. Slope Water zone of the current. Assocmtlons between hydrographic features and distribution,, ol zooplankton have bccn idcntilicd in the northwest Atlant ic for water mas,+cs (e.g. GRICE and H a r r . 1962: Cox and WIEBr, 1979: D.wIs. 1984) :rod Gulf Stream rings (e.g. Wt[m.: et al. 1976b; RIN(; GrouP . 19,";1 : Joke r ¢1 t l l . . 1984: Davis and Wirer.. 1985). The distributions of copcpt)d .species and species group.,, arc also i,s,,ociatcd with the vertical structt, rc of the envi ronment in o ther rcgkm,, of the ocean (c g. McGOwAN and WAI.KEr. 1979: CUMMINGS. 1983: LON(;eIURSr. 1985). Thi.', study and the work of WISIINER and At.[ ISON (1986) arc the lir,+t to identify thc,,c a,,,,ocmtion,,, in chffcrcnt zone,, o f the Gul f Stream and ,,uggcst that conmstcnt ;i,~SOgllttlon,, between plankton specie.,, and the env i ronment may be characteristic of zooplankton communitic.,, in the current .
508 C.J. +'.'X:SHJI-~N and K F WISU'~ER
Temporal stmdanties and differences: physical envtronment
Most differences in the ph}sical environment between the t~o sampling times may be attributed to seasonal changes. The physical structure of the Gulf Stream itself was similar in September 1982 and May 1983. although the temperature difference between the Slope Water and the Gulf Stream was much larger in the spring. The cooler surface water (<25°C) at the Sargasso Sea side station m the May hydrocast section (Fig. 2) may be due to v+mter cooling (ScHROEDER et al., 1959: WORTHINGTON, 1959: MENZEL and R'tTHER, 1960. 196 la). but an alternative explanation may be the occurrence of a recently described Gulf Stream phenomenon, the "warm surge" (GZLMAn and ROTHSTEtN. submitted). The May Sargas,,o Sea station ~as sampled (+--8 da},, earlier than the Central Stream and North Wall stations (Table I) so that the Sargasso Sea station was sampled during the "'cold phase" x~ hile the Central Stream and North Wall stations were sampled during the "'v, arm phase" of a warm surge event (GtI.M~N and ROTllSTEZN. submitted). Ho~se~cr. the potential effect of thi,, event on copcpod species distribution and abundances i.,, tmclear. The Io~er densit} m the sur|acc layers of the September transect, relative to May. ,s inchcatl~e ot stratification of the tipper la?er', during summer and earl~ fall (SIEGt:L et al.. I 0t)())
The cro,,,,-stream ~elocity field indicated that bi-directional cross-stream transport of water parcel,,, and potentially plankton pol~ulatlons, was prc,,cnt during both sampling times In the September tran',ect, convergence of cro,,,,-stream vclocitie,, from the edge', of the Gull Strc,lm to the center indicated that entrainment of water parcels from both ~idcs ol the ct lrrcnt w a,, OcCLII I l l tg (WISIINI R and At It,SON. 19~6). 110~ ever. in ,May three I} pc,, ol cro,,s-strc,uu transpolt WClC occtu'ling sinlull,ulcott,,l+~, (lqg 2): (I) cros~,-strcanl llow from the ilol th~c,,t to ,,OtlthC,l'q (Slope Water to S;ll'gasso Sc;.i) I~clow the thcrnlochnc alld 27.00 isopycmd, suggesting entrahlnlcnt of Slope Water into the Gull Slrcanl at depth, (2) cro,,s-,,trcani llox+ from ,,tmthca,,t to north++cst in the core ol the Gull Strcani +tnd abox e the thcrnIoclinc (except on the Sarg;.tsso Sca 'ddc), suggc,,ting entrainment and cioss-strcam transpt+rt of Sarg;.i,+,,o Sea water, and (3) flow out of the Gl.llf Stream from northwcM to ,,ottthca,,t m thc top 1511 nl at the Sal"gasso Sea side station, indicating dctr:unmcnI of Gulf Strc;.lnl ~,tll lace water to the Sarga,,so Sea.
Temporal ~tmthtrtti<'~ attd dif]erence~': COlWlmd species' attd COItllltllllllV Mructur¢
Copcpod 4pcctc', rc',pondcd in a smfilar nlanncr to the chilnging environmental Colldltlt~n.~ ;.|cro,~.,, the Gulf Stream during both sampling times ,,o that species distributions gencrdll} Iollowcd the same trends ;.icros.~ the Gull Streanl in September and May, ;ilthottgh SOll|C depth rangcs dilfcrcd (T,tblc 2; AIA+IhON, 19N6: WISiINER and AI.LISON, 1986). Vertical di,,trtbutions in the North Atlantic have been described for many ol these +,,peele',, (WIsIINI R and At LI.',,;ON, 1986 and relerenccs therein) and species di,,tributions at the Sarg;.t~,so Sea side of the Gull" Stream wcrc in gcncr~,l agrccnlcnt with these previous reports. I-Iowcvcr. a consistent pattern of environmentally as,~ociated change in species" vertical <.h,,trlbutlon across the Gulf Stream was not described previous to thin Mudy and that ol AI.! ISON (1986) and WISIINLR and ALLISON (1986).
Although total zooplankton bioma.',s lrom the samples was greater in September than in May (Al IJSON and WIS~INLR. 1986). this difl'crenec was not seen in the abundances of most of the sclcctcd copepod species (Table 2). Maximum and integrated group abundances at 4nmdar cro,,4-4trcam locatnon,, were ,dmilar for the two sampling time,, for most specie,,
Temporal persistence ol copcpod t.~pcs m the Gull Stream 51)9
groups. The absence of differences in abundance between the t~o times suggests that. for these species, there may be little seasonal change in abundance. Differences and similarities between September and May species abundances m this study cannot be conclusively attributed to seasonality, however, since only one cycle was sampled.
In two previous year-long studies in the Northern Sargasso Sea. which included data on 14 of the 18 species m the present study, total copepod numbers reached a spring maximum (following the spring phytoplankton bloom), with a second max,mum in fall and minima in early summer and late fall (DEEvE~. 1971. DEEVE~ and BROOKS. 1977). Most species followed this same seasonal cycle. Abundances in 196l (Dee~e~. 1971) ~ere greater in May than September for 11 species, approx,matel} equal for two spec,es, and greater in September than May for only one species. The d,screpancy between the present study (~ith no apparent change in abundance between September and May for these species) and the earlier ones could be due to samphng dfltcrences or diffcrcnccs bct~een the 2 years ,n the timing of the spring bloom and the spring copepod maximum.
It ~as somewhat unexpected to find community structure m the plankton within the Gull Stream. since phys,cal characterlsttc~ change over a relatively ~hort distance and advcctton is ~trong Community structures lor phytoplankton, zooplankton and mesopcla- gic tish ha~c bccn identified u,,ing descriptive statistical techniques in "'conservative cn~ironmcnt¢" (VENRICK. 1971) ~uch as g~re s~,,tcms (c g J~IIN i|,ld B~CKUS. 1976: McGo~.xn. 1977. Veyrlc K. 19S2: M c G o ~ a n and WAi KER. 1979)and 3oung warm-core rings ( G o u , D ( ' l i t [ . 1986). ~hilc Ic,,s distinct species groups were itlctltttictl in cn~iron- mcnt~ ~uch as the North Pacilic Transition Zone (Vt.Nri( K. 1971 ). the Calitornia Current (M( (;(,w,x~. 1977). and the Gull Strc,ml (.I x,,n and B xtKt,S. 1976) [dlthough recent ~ot'k in the C.,h fo, ,11,1 Current ha~ shown i, ssoct,tlit,ns ol plankton ~s flh physical Icalurc', such as jets (11xlrr'~. 19S4; M,x( K,x~ c t a l . . 1991 )1' ( 'onlnluntty stt'ucturc might bc Ic~ likcl~ in ,t "physically tlistul bell'", cgion, such as the (]till Stl cain. than m a more xtal~lc. "'bK~loglcally acconHllodalctl'" ;.i,ca bcc,tusc physical conthtituls lluctuatc t, nl~lcthctdl~l 3, prcsttm,d~ly ptcxcnting the cst~d+lishn,cnt ol a structured b~t+h~g~cal ct+,nnlt, n,t~, (SANI~I r~. 19(+N. 1969) l loxxc+cr, the cross-sttcam vcrt~cal strttcturc ol the Gull Strc,nn is ,tsclt :, pcrs,stcnt lcatutc (S,oMx, I, • 1966: I I xf Kt'~ <'t a l . 1985) and also m,,y bc ct>nsidcrcd prcdict,d+lc (Wll , I ~,.M"~, IiJSS)
(_'onlnitiility stt 'tictu,c in the oceanic 7t~oplankton has bccil t lcsctibcd, however the incchanisn~s rcgul.tting this structure arc poorly ttndcrstt~od (I l,xsw',xrD and M~GowAn. 1979; MtGowxn and Wal KI r. 1979: Lon(, l lurs , . 1985; WI, ,,Axis. 1988) and the cocx,s- tcncc of multiple species at simiktr depths ~s still paradoxical (e.g. I ltlic,~tnson. 1964; GH,t arov. 19,',;4). The vertical structure of the environment appears to inlhtcncc co,umu- nity structure, as evidenced by associ:ttionn observed conntntcntly bct~sccn copcpod species (and spccicn group.~) thstril~uttons and p.trticular depth intervals (e.g. P,,'~. and C(~o.~,IBs, 198(I, I IoI'KIN.',;, 19,',12; CUMMI~(,.s. 1983. SAmLo,o, 1984, Lo,,(,Hursl. 1985; S,x.,,t, o ,o . 1986. Wit.,.,ams. 19,";S). Compctflion and niche partitioning, the spatial thvcrs~ty of the ph~',~c;,l cnv,ronmcnt, and predation have all bccn suggested it,, mechanisms rcgul,tting the observed cotnmunity structures, ho~c~ or. unlike trc,,hwatcr and terrestrial Cll~,lronlllClll,~. it p,trticul,tr nlcch,tni'~m has yet to b¢ COll[irlncd for tin occ,tnic cotnmttnity (e.g. Lox(i~tt;rsI. 19,',15. Wit.t.tA,~.s. 1988).
In the Gulf Stream. species grt~ttps wcrc consistent at the t~o ~amphng times and wcrc btolog,cally mc;mmglul, w,th species grouped together having smdlar distrtbut,ons and bcha~ t~r. t lo~cvcl', because this study consisted of only two sampling times, caution ,s
511) C J AsuJIa.~ and K F. WtSHSER
required when concluding that the similarities in community structure are evidence for longer term consistent or persistent community structure (CONNELL and SOUSA. 1983). May and September are physical extremes in the Gulf Stream environment and would be the most likely times to reveal differences m copepod species distributions and community structure, if these differences exist. The consistency in species groups across the Gulf Stream and between these two sampling times suggests that a stable zooplankton community structure persists over time and space even in the highly advective Gulf Stream.
In summary, both the physical environment and biological distnbutrons in the Gulf Stream were remarkably similar in September 1982 and Ma~ 1983. despite the strong temperature gradient between the current and the Slope Water during May. Few differences were observed between the two times in individual species distribution, species abundances, community structure, and environmentally associated patterns of species (groups) vertical distributions. The structure of the physical environment in the Gulf Stream system strongly affects the d~stributions and abundances of these species, and their d~stributions in the current may be predictable.
Species groups as tmhcators
The con,,i,,tcnt a,,,,ociatton of,,pccles group', with physical characteristics of environment suggested that certain copcpod spcocs or group,, may bc used as tmhcators of spccitic region,, of tire Gulf Stream. In a related ,~tudy, the strong associahons between these copepod ,,pccic,,. ,,pccic,, group,,, and cnviromucntal characteristics wcrc demonstrated by the ,,uccc',,,ful dcvclopnlcnt of linear di,,criminant models that thffcrcntiated between ,,amplcs collected in difl~'rcnt region,, of the Gulf Stream on the basis of the copcpod specie,, or species as,,cmbl:tgcs alone (AbrlJIAN. 1991 ). Other rc,,carchcr,, have previously classilicd some of thc,,c copcpod ,,pccic,~ ,,tudicd ;is indicators of water type (Grlct. ;.rnd I IA~l. 1962; BOWMAN, 1971; l)Lt.Vr V and I:IROOKS. 1977; Cox and WII;BI. 1979). On a smaller scale. I IAukY (1984) and l lau~rv ct al. (1986) utilized changes in abtmdance ol plankton species ch,tractcrNtic of spccilie water ma,,,,c,, as evidence of water mass mixing m ctldics in the Californi,l Current Sy,,tcm.
lnthcator copcpod ,,pccics may bc useful tn identifying water mass exchange across the Gull Stream, especially on the northern ctlge of the current between Gulf Strcam- Sargasso type water and Slope Water (ASrlJIAN, 1991, 1993). For example, I'. boreah~, a Slope Water species, w as found in the Gulf Stream only art the North Wall station (Table 2, GI~It.'E and t lakt , 1962). Its congener, !' gracihs', is similar in size and behavior (thcl migration) and is considered to bca Sargasso Sea-Gulf Stream species (G~It.'E and Flakr. 1962; DLt-vtv and BkooKs, 1977). Thcsc two closely related spccics can be utihzcd a.s reliable inthcators of water mass type and mi,dng (AstiJIAN, 1991, 1993). Ccrtain species arc characteristic of only the top 21111 m of the Gulf Stream (Group 4) while others arc characteristic of middle-deep regions (Groups 2 and 3). Trends in cross-stream abundance of indicator specie.,, o r groups can be inthcative of contintrous cross-stream mi,dng processes occurring at diltcrent depths ira the Gulf Stream.
Rob" o] the Gulf Stream m cross-stream tnm wort o f photkton popuhttion.~
A hypothesis describing possible zooplankton transport routes by snlall-sca[e continu- ous processes across the Gull Stream, based on cross-stream and downstream velocity
Temporal persistence of copepod type,, m the Gult Stream 51 1
sections, was proposed by WISHNER and ALLISON (1986). In this hypothesis, cross-stream mixing of water and populations is most likely in the surface layers of the current (streamers and filaments) and in and below the main thermochne, where loss of water parcels from the Gulf Stream along isopycnals, especially in meanders, has been demon- strated (BowER et a l . . 1985: BOWER and Rosss~. 1989). Recent modeling of Gulf Stream meanders suggested that considerable cross-stream exchange of water parcels occurs in the main thermocline (only 42% of the water ~s retained in the jet in a "'typical" meander). while greater retention of water parcels is found in the surface layers (87% retained: BOWER. 1991).
The vertical d~stributions of the zooplankton across the Gulf Stream would determine the transport regimes the animals would experience. Species from Groups 1 and 4 would be subjected to cross-stream transport and mixing in the top 2(~) m of the current. This type of mixing might include water transfer through the formation of small-scale eddies or shingles as well as diffusion. Species from Group 2 ~ottld experience cross-stream mixing along isopycnals in and below the main thermocline. Species from Group 3 would be subjected to three types of transport during a 24 h period: (1) cross-stream mixing in the surface layers at night. (2) downstream transport in the core of the Gulf Stream during the day. and (3) isopycnal cross-stream transport in the main thcrmocline (deeper-living individuals).
If continuous cro`,s-stream mtxmg proces,~e,~ are occurring, then abundances of indicator `,petit,, would bc highe`,t near their source and decrca,,c acro`,,, the Stream. Cold-water `,pccie`, wottld be found in tughcst abundatlces at the northern statton`, and warm- water species would have highest abttndance`, on the Sargasso Sea side. Two of the four copcpod specie`, grottp`, demonstrated thc`,c pattern`,. Gmt, p 2, compo`,cd of vertically migrating Slope Water `,pccic`,, had highc`,t abundance', on the northern `,idc of the Gull Stream, and a reduced depth range from north to `,outh (Fig,, 7-9). These patterns arc con`,i`,tcnt with the hypothc,~i,; that these `,pccJcs arc ot Slope Watcr origin and have bccn cntr:tincd into the Gulf Stream sy`,tcm tn Slope Water whtch i , bcmg mixed below the thcrmoclmc. Group 3, conlposcd of vertically migrating Sargasso Sea spccic`,, had highest abtuldanccs in the Sargas`,o Sea in May and tn I~oth the Sargasso Sea and the Central Stream in September but lowest abttndanccs at the North Wall tk)r both times (Figs 8 and 9). Thcsc trends arc consistent with a S;.trgas`,o Sea `,()ttrcc fl)r Group 3 (and an endemic Gulf Stream popttlatiot0. "l'hc dccrea,;e in abundance ol both copcpod group`, across the Stream may reflect an imtbility to survive where changes in the phy,,ical cnvtronmcnt result m a decline in the quality or qttantity of food or alter bchztvior, such as restricting depth range or rcdttcing mtgration (e.g. BovI) et al . , 1978).
The abundances of Groups 2 and 3 wcrc al.,,o con`,i.`,tcnt with the observed cross-stream velocity fields, especially in May (Fig. 3). At that time. cro`,`,-.~trcam transport occurred above the tllcrmoclinc from east (Sargasso Sea) to west. High abundances ot Group 3 animals were observed at the Sargasso Sea station both day and night, reflecting po,,siblc entrainment of Sarga,~so Sea water into the Stream. Cross-stream transport from west (North Wall) to east occurred below the matn thcrmocline, and htgh abundances of Group 2 anim,ds at the North Wall of the Stream may be duc to entrainment ot Slope Water into the Stream. These results suggest that the September sampling was condttctcd during a period of enhanced entrainment of Slope Water (and Slope Water species), while the May sampling was conducted during enhanced entrainment of Sargasso Sea water (and species). Both Group`, 2 and 3 showed concordance of component species" abtmdancc
512 c J A~ml,,sandK F li~i~fl~.ER
trends (KendaU's W. Table 4). suplx, rting the idea that these trends are due to physical transport, v, hich ~ould act equally on all species within a group. Since these differences were confined to species found below the mixed layer, the cross-stream mixing probably occurred along isop}cnais of the main thermocline (Bower et a l . . 1985: BOWER and Rossav. 1989).
The two groups found m the upper 200 m of the Gulf Stream (Groups 1 and 4) sho~ed little consistent pattern of increase or decrease across the Stream. Differences in the integrated abundances of Group 1 between the two sampling times may be due to seasonal differences or cross-stream transport During May. abundances of shallow-living copepod species at the Sargasso Sea station may have been elevated in response to enhanced food conditions 1ollo~ ing the spring bloom in the Northern Sargasso Sea (MENZEL and RYrHER. 1961a.b: DEEVE'~. 19717. The h,gher abundances observed in September at the Central Stream and North Wall stations ma~ be due to convergence of flow at that location (~,'[bHNER and ALLISO,',,. 19867.
The existence of cross-stream exchange of phmkton populations has important impli- cations for regional biogeography. The con•equences of cross-stream mixing include greater dispersal o1 species with enhanced opportunity lor speciatlon, increased compe- tition and predation between Slope Water aild Sargasso Sea species, and the possibility of genetic exch,mge bct~ cell population•. Estimate• of the cro•s-stream tranH~ort associated with continuous m|xing procc•se• suggest that the contribution ol these processes to the mixing of pl,mkton poptil,ltion•, especially when cnhailccd In tl~ namic teatures of the Gull Stream such as mc;mdeis, may be sigmtical|t and of the san|e or greater magnitude as the tr,m•port assooatcd ~.stth ring I'ornhition (Bow t-r ;,llld ROSSll'~. 1989: T. Rossav, personal COlu|nttnic,llion). ILised on RAFOS Iloat tl,it:l, llowl R and Rossil~ (1989) suggested that up to 611"/;. ol the watcl m the c o r e ol the curre| | t at n|am tlleimt~chnc depths between 6S and 75°~V may bc exchanged with ~satcr from outside the (;ult Stream. 11 the average Gull Stre,uii tran•pt~rt at 73"W i,, 87.8 x lip n| ~ • i (i IAt KIN ;.llld ROShR~t, 19S57. then 1.66 x l{} I s m y OI (.;ull Sttc,ml core water may be Io•t through thi• prtlce••. In contra•t, cros•-•treanl tran•port ,l•soclaled with cold-core ring fornl;.ition h;.l• been cstmlated to be 9.11 x lllt~ nl ~ y t (l.'ut, t iSl! R, 1972; RIC iIARDSON, 19S37 ('l'hi• term doe• not include el leers ol •ub•cquclit rc•orpt |on ol the ring into the Gull Sit'cam. ) A similar cdlculatton tot ~atm-corc ruig,, is not av,idable; however, tran•port is probably les• for warni-core ring•, bccau•c of their •m,dler dianicter and shallower depth (WII Ill,, 1982; RICtIARIJSON. 19S3; JoYcl c t a l . , 19S47 Therelore , tran•port ol water parcels alld zoopl,lnkton popttlation• by COlltilltiotis cro•.~-~trcant i l l l \ lng procc••es may bc ;.i MgnlliCalll •ource ot expatriate zooplanktol i pot)ul,itlllll.~ hi the northwe•tern At lant ic,
CONCLUSIONS
(!) Although the temperature gradient between tile Gulf Stream and the Slope Water Wds greater m May 1983 than in September 1982. tile physical environment of tile Gull Stream itself was similar at tile two time•. DIlferences that ~,~ ere observed can be attributed to seasonality or to a warm-stirge e v e n | during the May study. The downstream velocity li¢ld was also similar at both times, but tile cross-stream velocity l ield showed differences m nu,gnittide and directionality.
(2) Adult copepod Sl+CCleS distribution patterns were similar for tile two sampling times. (3) The same copepod Sl~ecies groups were derived independciitly for tile two sampling
T e m p o r a l pt:r',t,,tence ot c o p e p o d typt:s m the Gult S tream 513
times, suggesting that a persistent copepod community structure may be a general feature of the Gulf Stream.
(4) Copepod species group distributions ~ere associated ~ith specific environmental conditions and regions of the Gulf Stream. Group distributions changed across the Gulf Stream in conjunction ~ ith the changing distributions of ph~ steal characteristics.
(5) Copepod species groups may be rehable indicators of regions of the Gulf Stream and could be used as tracers of water mass mixing.
(6) The integrated abundances of two of the four copepod species groups supported the h~pothesis that cross-stream mixing of water parcels and zooplankton populations by continuous small-scale processes is operating in the Gulf Stream.
A~ kl,,~t h, dk, clm.~rt~--Thanks tt~ H T Rt~s_,,by for gcncrou,,l~ pro~ ~dtng s, tmphng tm~e on hi,, crutse~, during the Pega,su,, progr,tm lor the M O C N E S S tows Spect,d thanks to Stuart AIh,,on lor all ot his work on the September tr,m,,c1.t and for hi,, ¢ttorts ,it ',ca Thanks to J W o r m u t h Ior the loan of the M O C N E S S and |o r help at ,,ca D Dor,,on J Ft+nt,tttlC. ~,~. Hahn ,rod S Rt+tt~.seatt prtv,,ttled t¢chnlc,d a,,st.,tancc ,md try,my ~tudents, e,,peclall,, L Bcatty and K Kelly. helped ++tth the s , tmphng Thanks to the c,tptam ,ind crew ot the R V Endeator for their d,,SlStdllC,.2 dl ~.¢,l C ()~, t,ttt. C R Sht~op. S T'~,olnbl.~ dnd ,in dll(lll~ motts rc'.tcv, cr redd :_Ill carher ~,cr',ton ol this tl|,mu,,¢rtpt and h,td nlan.v h¢lplttl ,,,uggeslton',.
Tht,, ~ork ts p,trt ol d D~ctor of Phllosoph~ tlt,,sert,ttt~m b~ C Ashltdn ,it the Unt~crs~t~ ot Rhode I,,,land The Gr,ldttal¢ Sehot~l ~I ()t¢,tn~gr,tph~ ,it the l,_rnt~,cr,,it~ ot Rhode I',ldnd pr t~tdcd hind,, to K Wt.,hner |t~r the re,,¢,trt.h C A,,hll,tn ,.~.ds ,,,upportcd b~ a Unt~cr,,it~ ~}1 Rhode Isl,md gr,tdtt,ll¢ a,.Slst,mt...htp lt~r part i~l the ~ork ,%IHp tmlc ,.~.,t,, ',Upl~t~rtcd I'~'~ ( ) N R gr,mt N(111~114-81 .('-lll)l~2 ,Ind N,%f' gr.tnt O ( [-83-111833 lt~ H T Ro',',h 3
R E F E R E N C E S
At I i.,,.i)n. ,% K (IUS(~) %pdthll dlttl I¢lllpt~r,iI ~drldbthlv ill ,,(~.opl,lnkttlll th~,|rll'~tlll(lns ,intl dblltld,lllC¢S dtro"¢,, the (;tilt %trc,tm ~,1 .% rhc,,t,.. I. lmvcr,,ttv t~l Rhode I,,I;md, IS7 pp
At I IN(IN ,~ K ,ttltl K I' Wl,ll%l H (It)8(~) Sl'~,ttl,tl ,111tl t¢nlpt~r,d p,tttcrtl,, ~1 toopl;tnktl}n I'llt}llld,S ,tl.'I'(}S'~ the (;ull Sire.m1 .~hlrltle I'( ohlgv I'rogre~ .~'rlt'~, 31. 23 t-24-1
,\still I R I ~.~, ,ttttl ( ' I1 Xlll I I r (It~lS7) Vertlc.d h,tl~tt.tl-p,trttttt)ntng I*~ col~cpt~thtc,. ,tntl ,ttlltll,, i}l sttbtr(~ptt,d I.*L't~.lltlC C~pcp(~ds ,~httttw Il loh~v. 94, 5(~1-577
A".IIJI ~"~ ( ' J ( itlt.II ) I lie Iltlluctlu¢ ol the (~tIII %tie,It1| ~111 the regltln,d I~lt~gctlgr,tl'lh ~ ol/t~t~l'Jl,tttkton (cl~pcp(~tl,.) l 'h I) Dtsscrl.tttoil, tlnt~.cr,,tt~ ~t I.~ht~tlc Isl,tntl. 2711 pp
A'MIIIA~,. ( ' I (It)t~3) [ IClltl". Ill topt. 'p~d '~pck'tC". :ll'~I.IIltl.IIIk¢, :teross .tlltl ,tlong ,t (;itll Sire.till 111¢,lntlel. eVltlU'llu¢ h~t cntr, lmntcnt ,tnd dctr~unmcnt (~1 Ihtid p.tr¢cl,, Ir~m tile (;till Str¢,tm lh'lT~-Sl'a ICe,cirri h I. 4|1. 461-482
11 ,~( kt • II, I I . I I. ('R '~t)l)llt k. 1{ I | I Xl I)RI( II .ll|d 1) L Slit}R1 (It)7t)) The dr, trlbttttt~n ~1 tue,.tlpcl,tgtc hshc,, m the ¢tltl,ttorlal .ttld ~t:stcrn North .Atl.lntl,.: t~t¢;,tn Journal o] Mitrtn(" Rc~'ltr~ h. 28, 171,k-2111
lil)~.~. 1 r A S 1 lt}~ll ) A ,.itltl'~l¢ klncln,lttC ii1¢1.11,1111..111 Ior hil l ing I]uttl part.el,. ,It.If, INN d nl¢;. l l ldcrlng 1¢t Jotlrnal o] I'h ~'st( ill ( h eattogrttllh I'. 2 I. 1 7 ~ 1811
|1<)~.~.1 r A S and I R(is'~ll'~ (ltlMtJ) |~.Vld¢lICC t~l trtls,,ilr<~nt,tl c~,ch,mge pr~,cc,,,,c,, tn the Chill Streanl b;.l,,¢d Oil Is~p~cn.d RA|'[).% l l . .d d.,t.~ Jtnlrmll o/Pltv~l~al O(e~mogr.phv, 19, 1177-1 It)(I
I I ~ . r A ~, . tl I" I~,.~',slt ,, .ind J I LIIIIIIRID(.I (IO85) I'hc (hill Strc , lm--h, l rr lcr or blcnder ' Jen t rna lo I Pitt ~l~ al ()(~'lltlOgral~hv, 15.2..I-32
B~',~. ~,1 ~,~ "I | . ( 1~)71 ) The dt,,lrtbutton t)! u,d,tn~ld ct~pcpl)d,, oli the st)uthc,l,,tcrn United St,ttcs between ( ' ape II,~ttcra,, ,tntl Southern t-lorld,~ .~Oltth~oltttttl £'otttrthlttlOtl~ to Zooh~,r. 96, 1-58
II(~'~l) % I I . I ~ tl WH Ill ,tnd J [ ( ' ( ) \ (It)7S) l.ttlllt'~ ol ,%'t,lllatOs(t,h~ im'~alop~ m the Nt~rth,.~.t:..,tern Atl,tntl~. Ill r¢l,ttt~m to Gull Str¢.lrn ('tlltI-C(~r¢ ring', II Phy',.it~h)gtc,d ,rod btt)chenll¢,|l tricot,, ol ¢xp,ttrt.tttt~n Journal o] .~htrt~t~" Re~'~tr~ Ii, 36. 14"~-I 5 t)
('HI mt '~ J (It~SS) Sp;Itml :tlld tcmp(}r,d ,'tbtmdancc pattern,, t~| eh,l¢togn,lth~, m the western North Atl,tnttu I tl~tlrt~gr,q~ht¢ and ~.¢,lst~n.tl ,d~ttlltl,ltluc p,tttcrn,, l)eq~-Sea Resear(ll...I.2. 11141-11151~
C'()",,'~ttt J tl and W P S~)1 '..A (It)83) On the ¢~.tdunc¢ needed to jttdg¢ ec(~logtc,d st.lbtht~, Anlert(an ~att~rah~l. 121. 781)-$24
( ' l ) \ J ,Intl I ~ II ',~,'tl I~t (lt)7g) ()tlgtn... ill t~1.¢,ltll1, pl,lllktt~ll m the Middle Atl,tntl¢ I:hght l;stttartne and ~ oastal ~larttl~" .~( It'lt{ t'. 9. 51)11--527
514 C. J ASHJIA~ and K F WISHNER
CumminGs J. A (1983) Habitat dlmenmons of calano,d copepods m the western Gu l f of Mexico. Journal off Marine Research. 4 I. 163-1 $8
D xvts C S (19t'~l.) Interaction of a copcpod populatton v.lth the mean ctrculatlon on Gcorgcs Bank Journal o] Marme Research. 42.57.~---59U
DLxvls C. S and P H Wtvse (19851 Macrozoopl, ,nkton btomass zn a ~arm-corc Gulf Stream nng" time sertcs changes in size ~tructure. taxonomic compo,.mon, and vertical dlsmbutzon Journal of Geophbstcal Research. 90.8871--8884
DEE', EY G B (1971) The annual c>cle m quantlt~ and composttlon of the zooplankton of the Sargasso Sea off Bermuda i The upper 511t1 m Ltrnnologb and Oceanography. 16.21tl-.-2411
Deeve~ G B and A L Broo~,s (1971) The annual cycle m quantlt~ and composition of the zooplankton of the Sargasso Sea off Bermuda 11 The surface to 2IH)O m Ltmnolog) and Oceanography. 16. 927-943
Dee~e~ G B and A L. Broo~.s (19771 Copepod~ ot the Sargasso Sea off Bermuda species composmon, and '. crtlcal and seasonal dl'.tnbutlon between the surtace and 211t141 m Btdh'ttn of Martne S~ tent e. 27. 256.--291
FAOEr E W ( 19571 Determmauon and anal)sis of recurrent group~ Ecoh~gv. 38. 586--595 Fa(~er E W and J A McGo~, x~ (1963) Zcu~plankton ~peclesgroups m the North Pacztic Sctence. 140. 453--460 Fxr r x,~ G P (1947) Vertical dl,.tnbutlon of the plankton (Sagttta. Cahmu* and Metrtdta) olf the south coa•t of
Ireland Procee~hng~ of the Royal Irtsh Academy. 51. 121-134 Ft'o! ISTER F C ( I%0) Athmtt~ Ocean Athl~ o] Tonperatun" and Sahnttv Proflh'~ and Data fro.t the I G Y of
1957-1958 Woods Hole Oceanographic [n•tffntton Atlas Series No 1. p 21P~ F toLt s t t r F C (ItJ72) C~clonlc ring• h+rmcd b~ the Gulf Stream It~f15-1966 In Sttuhe~ tn phvMtal ocean-
ographv a trttnltc to George $~tt~t otl ht~ SIIth btrththt~. A L GORD,X~,. ethtor. Gordan and Breach. New York. Vol 1. pp 137-H+,S
G XRFIFI I> N II! and D L E~, ~N'~ ( 19871 Shell water entr,unmcnt b,, Gtll! Stream warm-clare ring,. Jour.al o? Gcoph x ~tt al Re~eam It. 92. 13 .I 1(1l- 13.1 ) 12
Gtlll ARIIV m ~l (19S4) The par,ltlox ot the plankton rect~n..Idcrctl, l}r. U, llx rio specl¢~ t2OC~lSl ' ()IK(IS. 43.4f~ 52
Gll M,X.'~ C, and L I~.l~ltl•ll I% (Stlblnllted) (lull Stream xl,,trlll surges theNn,d ,nltl p~tentl,d ~,t~rtlfll) dllomdhc', in the Gulf Stream Journal o] (;¢oph ~ ~+t al Rt'~car~ h
(h~t 11~ R. W Jr. t2 R I1 ~l MtlRI anti G A t:r'~ \1 i I (I~IS¢~) Multivariate st.ttostit• ,IppIncd to phytoplankton data hOnl two (~ulI SIl'c,lnl ~,arnl-l.orc I lll[.l• l.tmnoh~gv and (h eanogral~hv. 31. 951-9<~8
(~Rl( I ( i 1) and A 1) I I xRr (I~Jb2) "l hc al~tludalltC. '.,ca•~m.iI tltcul rcnfc alltl dl•trll'~tltloll ol the Cpl/t~oplankton betv, ecn Nc~,. York and I|cllllutl,i I.~ oh~t~ ¢ll lllonotlral~ll~..1"-. 287-311~J
l lxl K~n 11 and 1, R~r.,Mz'~ (ltlNS) Ihc struttut¢ and IhmSpt~rt ~1 tilt: (lull S, trcan! at 73"W Journal ol I'hz ~t¢a[ O<¢anocr~q~hv, 15. 1439-1452,
I I ~! Mn D. T A I,Ia~,o and 1 I~.t~••n'~ (Itl85) l),tta I,[cp~rt ol the I'cgastl'~ I'rogranl at 73°W Graduate School ol Oteanography. U,uvcrslty ol I,[ht~tlc [',,land I echnl~.al I,[cport Rclcrcncc No, 85-2. 199 pp
t lat rY l R ( 19841 All ~dlshor¢ eddy in the ('dhh+rnht (_'tt~rent s~stcln Part IV pl.u|kttm thqrtbutton• I'rogre~ m (heanography. 13. t)5-1 l I
I I,u,rv k R . J J SIMI,%IIN, J PI l al~, ' , C J Kl)nl IN•kY and 11 Wll SI+NIIAtlN (1986) Biological ctm~,cqttcncc., ol .t recurrent eddy oll I't)mt ( 'onccptlon, C'A Jo.rnal o] (;¢ophwt(al Resear(h. ?1. 12.937-12.95b
I Ix~ ,a r l ) T L and J A ,Mt(;()w,.~ (1979) I'attcrn ,llltl structt~rc m an t)cC.llllC zot)plankttm conunt, nzty Amerltan Zooh~gtst. 1979, 11145-11155.
tlllcll(()tK G . E LI VlNI. D CoN'~ors. E Lls,~aaD and D |:r,~/~ L (19S6) I'h~slcal and bloh~glc.tl anatt~nly of a Gull Stream shmgl¢ Eo~, 07. 11128 (Ab•tr.)
t hm(, N G . editor (191~6) SYNOP Synoptic ()Lean I'rethctton Woods 1101¢ Oceanograpluc ln•tltutmn. Wt)t)ds t lole , MA. 26 pp
tlopM'ss "1 L (1982) The vertical d~stnbutltm ol zot+plankton m the ¢:tstcrn Gulfol Mexico Deq~-Sea Reu'ar~h. 29. libel-- llIg3
Flut.sEmasN K ( 191~61 A re+. i•ltm ol the gcnu.• Lttc tt utta (Ct+pcpoda Cab'molda) with a key to its •peti t• B.Ih.ttn o] Marine S~ ten~ eo] the (;uI] ttlld (. arthheall. 16. 7112-747
! lur t nuar E. M ( 19(~,1 Success,on and dwcr.,~ty m the plankton llor:t ol the ~estcrn North Atlantic Btdh,ttn of Marine St +en~e of the Gull and Cartbbem~. 14.33-.44
['It rtlu~.~o,,,, G. E ( 19641 "l'hc paradox of the plankton Amertctm Naturah~t. 95. 137-145 Jail,'., A [: and R !1 BA¢ ~.t • (1976) On the mcsopcl.tg~c fish faun.t• of Shape Water. Gull Stream. and northern
Sarg:t~,•o Sta DetT-Sea Re~eam h, 23. 223-234 Jt~'~tl T ,%1 . I,l Bacgts , K Ii,~Klll. I' IJIA(KV+IIIIIR. () BR(I~,N. "1 C'()~,II%. R [~VA~,. G ["Rt \ r l l . D
Tem[x~ral persistence of copepod t~p~,, m the Gull Stream 5 [5
Mot HTAI~. D. OLSON. R. SCHLITZ. R. SCHMITI'. P S~lm~. R SMrrH and P ~'IEBE (1984) Rapid evolution of a Gulf Stream ~arm-core rmg ,~amre. 308. 837-840
KREBS C J (19~9) Ecologwalmethodology Harper and R o b . Ncv+ ~ork 654 pp Lu LZBRItmE J and H T Rossa~ (1986) Shelf and slope ~ ater entrammcnt ,n the near surface of the Gulf Stream
EOS. 67. 1028 ( A b s t r ) LO',,<~Ht. RST A R. (1985) Relationship betv, een dl,,erslt~ and the ~ertlcal structure of the upper ocean Deep-Sea
Research. 32. 1535-157() MACKAS D L.. L WASHBt_RH and S L S'~ITH (1991) Zooplankton community pattern associated V, lth a
Cahforma Current cold filament. Journal o] Geoph+stcal Research. 96. 14.781-14.797 McGov*A.',+ J A (1977) What regulates pelagic communi ty structure m the Pacific" In Ocean sound scattermg
prediction. N R. A'qDERSEH and B J ZAHL RA~;EC. editors. Plenum Press. New York. pp 423-444 McGov, xh J A and P W WALKER (1979) Structure m the copcpod community of the North Pacific Central
G~re Eeologtcal Monographs. 1979. 195-22++ ME',ZFL D W and J H RrrHER (ITS60) The annual c+~cle o! primary production In the Sarga~so Sea off
Bermuda DelTa-Sea ResearOi. 6.351-367 ME'.ZeL D W and J H R~nER ( It)Sial Annual ~arlatlon~ in primary production ot the Sargas~o Sea off
Bermuda DetT~-Sea Rewarch. 7.2,~2128S ME".ZEL D W and J H R'crHEIt (l'461b) Zooplankton in the Sarga.+so Sea oil Bermuda and its relatmn to
~wgamc productLon Jmtrnal du Consed. Const'll Permanent hlternatlonaI ptmr L'E~phJratttm de la Mer, 26. 251~-25+~
MooRE [+| B and D L O'BI:RR'+ (1'457) Plankton o[ the Florida Currcnl IV Factor', influencmg the vertical dl~,t rl['Rit ion !.)| ~onle Lore nit )n ¢,llant'qd~ Btll/ellnO[,~larme$ttetltet~[gheGId/atldCtlrlbbt,an.7.2t}7-315
NOF O (I11g8) The propag,lIllm t+t "'.,trcamer,.'" along the periphery, ol v+arm.core rmg,, Deign-Sea Rewarth . 35. 1483- 141~
ORs ar FI B and M Fo'~o (1967) Copcpod~ ol the Florida Currcnl Fauna Caribaea No I Cru,mlcea. Part l C o p c p o d a hr , h lu te o| M , m n c 5~.lcn~.e. U l l l vc rs l l x ot ,'~h,intL 137 pp
l't~ I i~t~ L C (I~,~77) Madwmam al et oh,x,v Wdcy . Nc~ ~ o r k . 3,%~ pp I 'wl R K and S I I (.'<x~,~m, (I~,~l)) Vcr I Ica l d l , ,mbut~on ot zoop I ,a l k ton ove r Ill~.' I Io r lhern dupe., o! the Wy~,dlc
"I ht.n,.~m Ridge: J .urna l . ! I'[ankt+m Rt'~eart It. 2. 223-234 [~.l( II xa(),,tr~ I' [. (19S3) (;till Slle,nn Rmg,, Ill l.ddtcs m marine ~t lcm e. A R I~,lmz',.s<)x. ethh~r. Springer.
llcrhn, pp I I-I-14-I
l~,z,,t, ( i~in P ( I~IXI ) (;ull Sllc,un u~hl+c~rc riilg~ Ihclr pll~si~.,,, chcnlz~lly and bzoh~y .% wm c. 212. II)91-111)(I
l#,~l II ~i J (1972) "life vcIllCa] dl~Irlbulzons and dlurn,d llugr,Illon~ ol c,dam~id ~opcp~ds collc~ted on lhc
'~()ND ~ rmsc. I%'=, II S~,,l;.'nhiIlC atA:ounl l,nndlc~ (',d,nud,lc up Io and including AcZld¢Ida¢ .hn~rna! o/ the Alarme Ihoh,~v .- |~. t mt t .n . f Ihe Umted Kingdom, 52. 315-343
I{t~, I I ~ I (I~h";4) I hc dlcl nl igr. i l lOl ls , ind d lMr l l ) t i l lon ~, ~ l l h l n a mcsop,.2l,lglc ct)n ln l tn l l ty Ill Ihe nor th east Atl,liltlg---4 l ll~: u+pCl+t~d,i l'++,¢ress m ()team~graplt+. 13. 353-11,~
R,,l llr M (; ,uld II B Mo~Xl (191+5)"[he ~,crti~.,ll dr,,Irll'~uIlon ol St~llle Lt~llllltOlI cal,molds In the Sir:ills ol t'h+rlda Ihdlelm ++] Marine .St tt'tlt ¢'. 15.51+5-5711
l,tt~sz Xl (I~J33) l 'aunc tl¢ I r,lnc¢ 26 ('Opel+~otls l+¢l,igltltlCS Olh~.c ~+cnlr,ll sic l',lunl,,llquc. P,lil%, 374 pp S xs~l t)itl D D (19,";4) l'+n,,irtmmcnlal |,tutors mlltlcn~.lng thurlhd dlslrlbnIltiil ol zo~q+l,mkttm and ichthyoplank-
Ion JrJnrnal o] I 'hmklon I,~t'seart h. 6. 767-79!
Sxxl~t~lt~ D D (ItJ,~t0 [nlhlcncc ol the biological and physical cnvmlnlncnl on lhe ,.crtl~.,d distribution of nlcso,,o<~f.l,mkltm and m~tronckton m the ¢,lstcrn tropical P,t~.lhc ~t.lartnc Ihoh+gv. 93, 2(~.L-27tJ
";A'~+t r., II L (19¢~,~) M,lrmc benthic, diversity a ctmlpar,m,,e slutly AlHerlt till Naturahst. 102.24~2,~2 Sa'ql)l r.h tl L (1%9) Marine hcnlhlc diversity ;tnd the shd'qht,,,-tnn¢ hyputhcslS Brookhaven SvlnpOWlltn tit
Ihologv. 22.7 l-,gl) SAS [',,stilt I~ [hi ([tj.~,) A,-i.S uscrs gtt.h', stattsttt s. Vt:rSltlll 5 cdn SAS Institute Inc . Cary. NC. 956 pp St ~tall~ D~ m 1: . I I Sl<~,l'.tz t . D ML../I t and W SL t t l Ill t JR (1959) Clml,lllc stablhty ol eighteen water at
Bcrnltida Jtmrnul o[ ( ;ctqJht'wt aI Researt h. 64.36.~.366 ";ll xv. I' T and II "l Rt+h'.,l¢'~ (lt,~S4) 1"ux~.lrd,. a [..lgr,ingz.m th:s~.rlptlon ol the (;till Str¢.ln| Jot+rnt#l of Ph.x stttzl
()teitnogr~tpht,, 14.52,~540 Sllt , l l D A .[++, IlL'RRIXI,A. R R BIDIt,ARI. R C .~+,nlll. [1 1'xm.'I D [)ItkIV. J MxRRAandK S BAKER
(lg~Jll) Mcrldlunal varl,itit+llS of the sprlngtinte ph) tophutkton conl inunlI ) Ill [he ~;,irg,lsst~ Sea Journal o f Marme Researt h. 48. 379--412
SOKX[ R l~, and | '+ J I?+o1111 (ltJgl) Btomctrt', 2nd edn. W 11 [ ' foeman. Nee+ ~ork. ,~5~/pp
5 1 6 C. J ASHJIA~, and K F V~t:aH~.ER
SrELR A (1932) Copepoda (6): Pleuromamma Gle,,br 181;8 der D e u t ~ h e n Tlefsee-Expedmon Deut,.chc T ,e l sec-Expcdmon "'Valdv,,a-. 181;~--1899. Vol 24. pp !-111;
Sro'.r, tet H ( 1'4~,61 The Gul) Stream. 2nd edn. Unv, cr,,It~ ol Cahforma Press. Berkeley. 243 pp VE~.RI(_K E. L (111.71) Recurrent group,, of diatom specie., In the North Pacific Ecolog,. 52.614.--625 VE~.RI(.K E L (1~82) Ph.~toplankton in an ohgotrophlc ocean ob, .c~dtlons and quc,.tlons Ecological
Monographs. 52. 129-154. WiEaE P H (1982) Rings ot the Gull Stream S{letttt[?c American. 246 61~70 ~AIEaE P. H . K H. B t r r . S H Bo',o and A W Morton, (1'476a) A multiple opening/dosing net and
en,.Ironmental ,,en,qng s~stem for sampling zooplankton Journal o) Marine Researth. 34.3 LL-326 ~,~,IEBEP H . E M HULaErT. E J CARPE~TER. A E J~H~,.G P K ' ,xPPi l I .S H Bo ' to .P B Oat 'qERandJ C
Cox ( 1976bl Gulf Stream cold core ring,, large-, talc interaction sites tor open ocean plankton communi- ties Decp-St'a Rewarth. 23.6t;5--7111
~,l, tl_tt x'.ls R (luSS) Sp,itl,iI hetcrogcnelt~ and niche differentiation m oceanic zooplankton H~drobtologut. 167/ 168. 151-15~
~3, i,,lc,.rr K F and S K At [ t,,,o~ (ItJ86) The thqrlbutlon and abundance ot copepod,, in relation to the ph.~,,ital ,,tructurc of the Gull Stream Deep-Sea Rescam h. 33, 7115-731
WORTI-II"~GTO% L ~. (It~5~J) The IS ~ v, atcr in the Sarg,r,,,o Sc,i De~y,-Sca Research. 5, 2t;7-305 ~.~orrHl~,ro'~ L ~. (It;76) (')n the ,'~orth Athmttt Ctrtuhttton. John,, Hopkln,~ Unr,'cr,,lt~ Pros,,. 1 III pp Z xr J H (It;S-l) Btostattstttal analysts. 2nd edn. Prentice-Hall. lnc , Englev, ood Cldls. NJ. 718 pp
