convergence within the Mississippi River plume front. JOHN J ...... MVRTH~ C R (1976) Horizontal diffusion characteristics in Lake Ontario Journal of Ph~ steal ...
( onttnentalShelf Research ~¢ol 12 No 11 pp 1265 1276 1992 t'rmted m Great Bntmn
1~27S~l,~43/92$q 0ql+ I}Ill) © 1992 PergamonPressLtd
The surface accumulation of larval fishes by hydrodynamic convergence within the Mississippi River plume front JOHN J
GOVONI* a n d CHURCHILL B
GRIMESr
(Recet~ ed 19 December 1990, m rm tsed f o r m l(/October 1991 accepted 16 Decembel 1991)
A b s t r a c t - - l n September 1987 and May 1988 densities ol larval fishes at the surface ~ere greater within the frontal zone of the M~ss~sslpp~ Rwer plume than they were in plume or shelf waters Sharp turbMltv d~scontmultms of 10-50 m w~dth accompamed bv temperature and sahmt~ dlscontmmtles ~ere embedded w~thm this large-scale frontal zone whmh ~tself spanned 2-20 km Apparent convergent m o h o n normal to turbM~ty d~scontmumes had veloc~tms m the range ol 0-0 8 m s -~ lateral shear was also evMent Exceptmnally high densmes of larval fishes ~ere not always observed at turbld~ty dlscontlnmtles TurbMlty d~scontmumes otten sinuous in conhguratlon formed and dissipated over time scales ol 2-6 h Expected densities of larval fishes within frontal convergence zones, calculated with a slmphfied ad~eetlon-dlllUslon model approximated mean and median densmes observed within the large-scale frontal zone
INTRODUCTION
HYDRODYNAMIC convergence has been used to explain the accumulatton of larval fishes within the frontal zone of the Mlsslsstppl River plume in the northern Gulf of Mexico (GovoNI et a l , 1989) and the Rh6ne Rtver plume in the western Mediterranean (SABAT~S, 1990) In the Gulf of Mexico, larval gulf m e n h a d e n , BrevoorUa patronus, occurred in inordinately htgh denslttes at the surface along turbtdlty d~scontmuttles during winter when the Mississippi Rtver plume was well defined as a thin lens of cold, turbid, lowsalmlty water projecting over the warmer, saltier shelf waters m the northern Gulf (GovoNI et a l , 1989) In thts study, densmes of larvae were h~ghly vartable, ~e exceptmnally high densmes were not evtdent everywhere along these turbidity d~scont~nulttes The m o v e m e n t of flotsam toward, and ~ts accumulation at, the frontal mterface tndmated convergence of surface water, but no direct observatmns of convergent veloctttes attendant wtth esttmates of larval fish density were avadable The strength of convergence was indexed instead wtth a simple formulatmn that was based upon the density difference between the plume and shelf waters, but derwed values were not correlated with densmes of larval fishes Frontal convergence, as a m e c h a m s m that accumulates larval fishes was consequently vtewed as a localized and complex c o n d m o n Thts p e r c e p n o n , lacking in observattonal detad on the vertmal and horizontal structure of the frontal zone and the k m e m a t m s of frontal convergence, ts vague SABAT~S (1990) s~mdarly used frontal
"National Manne Fisheries Servme, N O A A , Beautort Laboratory Beautort NC 28516 U S A +National Manne Fisheries Servme N O A A , Panama C~ty Laboratory/ Panama City FL 32407 U S A 1265
1266
J J GOVONIand C B GRIMES
convergence to explain the aggregated distribution of larval fishes at the rlverlne plume front without details of the physics of the frontal zone While nverlne and estuarlne plume fronts have been well studied from the physical perspective (see reviews in BOWMAN and lVERSON, 1978) recent work has shown that the physical behavior of plumes and their associated frontal zones can differ among plumes of differing spatial scales (CURTIN and LEGECKIS, 1986, CURTIN, 1986a, 1986b, GARVINE, 1986 LUKETINAand IMBERGER, 1989, O'DONNEL, 1990) In c o m m o n with other rlverme plumes, the Mississippi River plume exhibits small-scale turbidity discontinuities, but unhke many smaller plumes (GARVINE, 1987), 11 IS acted upon by Corlohs (WRI6Hr and CO[EM&N, 1971) Here we examine the surface distribution of larval fishes about the Mississippi River plume m autumn and spring, when rlverlne discharge is seasonally minimal and maximal (DINNELL and WISEMAN, 1986) We then compare the density of surface dwelling larval fishes within the plume's frontal zone with concomitant estimates of the velocity of frontal convergence Finally, we compare observed densities of larval fishes within the frontal zone with expected densities calculated with a simplified version of the model of OLSON and BACKUS (1985), a model that describes the concentrating of depth-keeping fishes at fronts in a simplified convergent flow field The purpose of this later objective is to determine ff convergence can account for observed densities of fish larvae METHODS Because frontal convergence (GARVINE, 1977) and the accumulanon ot larval fishes (GovONI et a l , 1989) is a surface p h e n o m e n o n we examined the distribution of surtace dwelling fish larvae as observed in neuston collections We collected lchthyoneuston In and about the Mississippi plume front with a 1 by 2 m neuston net fitted with a 947/tm mesh net towed for 10 mln at 1 8 km h -1 and measured the x eloclty of frontal convergence on two cruises in S e p t e m b e r 1987 and in May 1988 The neuston net sampled the upper 0 5 m of the water column Stations allocated at approximate 5 km intervals along transects approximately normal to the Mississippi Delta were occupmd once on each crmse (Fig 1) Hydrographic profiles of the water column (temperature and salinity) were obtained electromcally at each station along with collections of lchthyoneuston Stations were occupied in serial fashion regardless of tidal phase or time of day Each transect required approximately 12 h to complete We estimated convergent velocities at positions where the plume's turbidity discontinuities, evidenced by sharp color and textural discontinuities of the sea-surface, coincided with t e m p e r a t u r e and salinity discontinuities One of these dlscontlnmtles was encountered within a transect, while six others were encountered between transects (Fig 1) At turbidity discontinuities, colored surface drifters (computer cards) were released along short transects that perpendicularly traversed the frontal interface from 185 to 500 m on either side (e g red drifters in plume water, white at the frontal Interlace, and blue m shelf waters) Preliminary work with both surface drifters and surface-water drogues in July 1987 indicated that surface drifters were more responsive to the m o v e m e n t of shallow plume water and that neither drifters nor drogues responded beyond 500 m perpendicular to the frontal interface Positions that m a r k e d the beginning and end of the line o1 drifters, as well as the position of the frontal interface, were recorded on a Loran-C plotter The elapsed time from the release of drifters farthest away from the turbidity dlscontlnmty
Surface accumulation ot larval fishes
1267
A *4 o6
010 09 o8
,go 00'
.21
064 @46 0 7 0 Q . - 0 5 g48. 0 4 5 *37 0 6 9 066~ (7~8 49" C~t@75 ,38 •68 -67 057 eSO 043 056 *St • 55 .52 042 • 54 *53 041
.35
026
,34
039 040
~L. 023
029
• 33 "32
,22
030 031
I 89 ° 00'
I 90 ° 00'
B 01 q2
o3 o4 *5
• 37 ~ e51 e38 C~45
e33
~-022
024
• 5 2 0 3 9 044 "~" e31 • tK~ O40 0 4 3 042 Oao 029 041 028
012 o6 011 e7 013 QIO O9 Dr4 "8 CtS 016 OI7
025 C~IL
121 ,20
027
20° 00'
,18
i19
I
I
90 ° 00'
89 ° O0 °
Fig 1 Stations occupied m and about the Mississippi Rl,,er plume {open symbols denote stations occupmd during the da} closed symbols denote stations occupied during tht. mght and encircled symbols denote locations ot turb~dltv dlscontlnUltXes where conxergencc wa, mea,ured) A--September 1987, B--Ma'~ 1988
until their ahgnment along the lrontal mterface was used with distance, as measured b} the difference between positrons, to estimate the velocity ot apparent frontal convergence (t e observanons were Lagranglan) We employed the advectlon-dlttUslon model ot OLSON and BACkUS (1985) to calculate expected densities of larval fishes at the plume front This model assumes that larval fishes disperse and swim randomly m the horizontal but sense and counter vemcal motion As a result, vertical advectlon and diffusion Is ~gnored It also assumes that &ffuSlVltmS along (the y-coordinate) and normal to the frontal interface (the t-coordinate) are equivalent If one assumes further that observations were made mhen the distribution of larval fishes w as stable, 1 e at steady-state, then the Olson and Backus model takes the form
1
C = Co [ ( 2 k / ( _ V / L ) c q ~ ] l / 2
exp
Q - ? ~ )aoZX v- -
1268
J
J
GOVON1
and
C
B
GRIMES
STATION NUMBER 6I
5
4i
3i
2i
1I
2,2 213 24 25 26
a2 a,a a4 a~ a6
54 5,5 5,6 57 s,8 59 6o,6~
0
A
181B
- -
~ 3 6 ~
~ 3 4 %
,,, ..............
tlllt,
Z o_
m[
B
12 ..................
iiiiii,t,,,
,
................. llll
.....
~,,t,
............
Iq
i ,,,
CL_%-U
6
,Ic
2 e ~
I
10
-J
12
DISTANCE OFFSHORE TO INSHORE (krn)
Fig 2
Selected hydrographic sections across the Mlssisslpp~ River plume m September 198,7 A--temperature, B--sahmt~ C~Jenslty (Or)
where Co is the observed median density of larval fishes in plume and shelf waters, i e
outside of the lrontal zone (number of larvae x 103 m -'), K is the horizontal diffustvlty coefficient in the r-coordinate (m 2 s-~), Vls o b s e r v e d c o n v e r g e n t velocity (m s t), L is the length scale of convergence (500 m), a~, is the inverse square of the length scale, X, associated with horizontal distance away lrom the front ( m - l ) , and Z is the depth at the frontal interface (m) If interest is only m the maximum density of fishes at the surface near the trontal interlace, then
( VL I1/2
C = Co \ ~ l
All parameter estimates were empirical except K, the dlffUSlVlty coetficient DlffUSlvlty is spatially and temporally scale dependent (MuRTm, 1976, OKUBO, 1976) LAPICQUE and BREITTMEYER (1973, in LEWIS, 1984) estimated the dlffUSlVity of passive particles perpendicular to the frontal interface of rlverlne plumes to be in the range of 2-10 m 2 s- i Because of the apparently narrow width of convergence zones, diffusivity was allowed to vary from 1 to 10 m 2 s 1 in our calculations of expected densmes
RESULTS
In September 1987 the plume was a thin ( 1 - 4 m ) surface lens of turbid, warm, low-salinity water that overlay the clear, cool high-salinity shell waters (Fig 2) In May
1269
Surface accumulation of larval fishes STATION NUMBER
28 29 30 31 32 33 0
I
I
I
I
I
I
40 a9 a,8 a7 a6 a,s a,4
41,12 ,ia 4,,1 ,l,~ ,1,6 'V
\
24
\\
IIl[llllll[[[l[[]l[
~illlllll[lllll[illllllll[I
53 5? 5,1 50 49 48
25
[11 [ll[lll[[I
i
H,ihlH,, ''hNIH''I''HH
//t
'I"
o
121"I\
° ,,
\\ ~11[1111
;I\\ %
,l[[l]
II[llllllllllllllllllll
i ill
litllilll
IL ......
'' *''lb'll
' H 'lllllllllll
I IJ
,,/
e #-'--.
'I\
....
], Hih
Ilil[[IL[lililillllil[l
[H
ill[I
ii ill
ill
.............
II
DISTANCE OFFSHORE TO INSHORE (km)
Fig 3
Selected hydrographic sections across the Mississippi Ri~er plume m May 1988 A - - t e m p e r a t u r e , B--s,lhmtv, C ~ e n s i t ~ (at)
1988, when discharge from the Mississippi River is greatest, warm and treshened plume water occupied the water column down to 12 m or more (Fig 3) Temperature and salinity differences between plume and shelf waters were greater in spring than in autumn In either season, plume water was horizontally separated from ambient shelt waters by a frontal zone evidenced by the horizontal compression and surface-intersection ot the same lsopycnals that constitute the depth-discontinuity layer beneath the plume (sensu GARVINE, 1987) We defined the large-scale frontal zone as encompassing those adjacent stations within a transect that differed in surface denslt~ by >1 ot value (Fig 4) So defined, this large-scale frontal zone spanned from 2 to 20 km (Fig 5) E m b e d d e d within this large-scale frontal zone, were small-scale, ephemeral turbidity discontinuities, characterized by marked contrasts in sea-surface color and texture with accompanying foam and flotsam rows These discontinuities were encountered on five dates in September 1987, two in May 1988 (Table 1) Turbidity discontinuities usually lay near the inshore margin of the frontal zone. but they were not apparent everywhere within this zone (Fig 4) The spatial scale of turbidity discontinuities ranged from 10 to 50 m (normal to the interface) These fronts had temperature, salinity, and ot discontinuities that ranged from 0 107 to 1 293 A°C, 0 643 to 4 237 A%o, and 0 433 to 3 247 Aut for the upper 0 5 m of water (Table 1) The interface of turbidity discontinuities was often sinuous, and on occasion, secondary lines of color and sea-surtace texture were observed parallel to, or intersecting, the primary front Turbidity discontinuities appeared to lorm and dissipate over a period of 2-6 h Surface drifters typically moved from both sides of turbidity discontinuities toward, then
1270
J J GOVONIand C B GRIMES
29 ° 00'
90o00 '
89 ° 00'
29 ° 0 0 '
00 ° 00'
8 9 v 00"
Fig 4 Plan view ot surlace lsopvcn,tls m and about the Ml~lsslppl Rl~er plume (open svmbols denote stations occupied during the da,, closed s)mbols denote stations occupied during the night and encircled s~mbols denote locations of turbidity discontinuities) A--September 1987 B--May 1988
aligned laterally along the Interface The alignment ol surface drifters along the interface indicated the presence of lateral shear Apparent convergent velocities differed between plume water and shelf waters, and at some fronts, drifters released in either plume or shell waters did not move toward the interface at all (Table 1) Apparent convergent ~elocitles on both sides ot the interface ranged from 0 to 0 8 m s -1, but showed no consistent pattern with the dlllerence m water density across turbidity discontinuities Overall, movement of surface drllters toward the interface was faster in shelf waters than it was m plume water, the fronts observed on 31 May 1988 were exceptions (Table 1) Averaged for both sides of the frontal rater/ace, velocities ranged from 0 05 to 0 25 m s 1in September 1987 and from 0 15 to 0 40 m s- 1 in May 1988 Frequency distributions ol the larval fish densities were highly skewed and kurtotm (skewness and kurtosls = 3 20 and 13 65 In September 1987, 1 78 and 2 (19 in May 1988)
1271
Surface accumulation of larval fishes
I)44e
~,
*let
• s s s~b /ells 1=04 , .e le3.
.e74 ,---"
....
o,o
t::e
L
*4
//
4
d
6
* '
-)9* 00'
(,106 726e~
,..0 ~ 0 l eeu~e , - ~ . o - ' " 6080/t82,121~8 t ~ t ~ S S e192 oily
-..Q2~
e*'r ,*Jt'1~t /0*4 -~-~-~ ~-2¢ .lr~ 02co
\,
,st.
" 'J 4
0
~4'' *23
03~ '~ "1 "'-"
I 89 ° 00'
I 90 ° 00'
=1~3" /elG8
teoO
/
721e ~
.aoe ~ 11177363 ¢
1142* 1~" t *61,U _ _ \ 0 ~1C~ .952 ,46. ~ 'J=s ." 0 ~0123
K..,o o,'~) o )o,:]
I
90 ° I X )
.
i140
~D38 ~ ~ el6 °28 ~ ~ Oll *26 ~)28 e13 • \\~ ~-*3(~
~8
063 ~=o4 0168
*76 .31 *162
29° ~ '
.71
I 89 ° O0
Fig 5 The spatial distribution of larval fishes in and about the Mississippi River plume frontal zone (stippled areas indicate the plume s frontal zone open symbols denote stations occupied during the day closed symbols denote stations occupied during the night and encircled symbols indicate locations ot turbidity discontinuities) Numbers are densities of larxal hshes (number x 1(I~ m 2) A - - S e p t e m b e r lq87 B - - M a y 1988
and did not contorm with a random normal distribution (P < 0 05 analysis of variance for normality (SHAHRO and WILK, 1965)) Means were greater than median densities (Fig 6) If lrontal convergence operates to accumulate larval fishes, densities should be greater within the frontal zone than in shelf or plume waters Frontal convergence is a surface phenomenon, however, and larval fishes typically disperse from the surface layer by day ( N E I L S O N and PERRY, 1990), possibly producing dlel differences in the action of surface convergence on their distribution Mean and median densmes of larval fishes at the surtace were greatest at stations within the large-scale frontal zone when those stations were occupied during the night (Figs 5 and 6) In autumn, 11 of 16 densities that were above the 75th percentile for time of day (i e >204 x 103 m -2 for stations occupied during the day and >325 x 103 m -2 for stations occupied during the night) occurred within the frontal zone (Fig 5A) In spring, 16 of 26 densities that were above the 75th percentile for time of
1272
J J GovoNI and C B GRIMES
Table 1
Phvstcal charactenstlcs of the small-scale front o f the Mlsstsstppz Rzver plume as obsen ed at turbtdttv dlscontmmttes
D~scontmulty
Date 5Sept 87 6Sept 87 7Sept 87 8Sept 87 9Sept 87 31 May 88 31 May 88
Station number
AT (°C)
AS (%0)
Actt
46 60, 61 75 77 81 54 62
0 113 0408 0112 0369 0283 0 107 1 293
064~ 1087 2779 1057 4237 0 532 4 234
0443 0680 2036 0676 3247 0 433 3 017
Convergent velocity of shelf water (ms I)
Convergent velootv ot plume water (m~ I)
02 01 05 01 02 00 00
01 (10 ()0
(11 ()0 o 8 0 3
day (1 e
> 1 3 7 x 103 m - 2 for s t a t i o n s o c c u p i e d d u r i n g the d a y and > 5 1 x 103 m : for s t a t i o n s o c c u p i e d d u r i n g the night) d i d so ( F i g 5B) E x c e p t i o n a l d e n s i t i e s did n o t o c c u r r e g u l a r l y at t u r b i d i t y d i s c o n t i n u i t i e s O n e o f five d e n s i t i e s o b s e r v e d at t u r b i d i t y discontinuities was a b o v e the 75th p e r c e n t i l e in S e p t e m b e r 1987, o n e of two in M a y 1988 (Fig 6) I n S e p t e m b e r , this e x c e p t i o n a l d e n s i t y (Sta 46) o c c u r r e d w h e r e c o n v e r g e n t v e l o c i t i e s w e r e high, b u t n o t m a x i m a l , a n d w e r e r o u g h l y s i m i l a r o n b o t h sides o f the d i s c o n t i n u i t y In M a y , the e x c e p t i o n a l d e n s i t y (Sta 54) c o r r e s p o n d e d with the m a x i m a l c o n v e r g e n t v e l o c i t y o b s e r v e d ( T a b l e 1, Figs 1 a n d 5) T h e m a x i m u m d e n s i t y o b s e r v e d In this s t u d y , 1837 l a r v a e x 103 m -2 (Sta 56, M a y 1987) o c c u r r e d o u t s i d e of the f r o n t a l z o n e as we d e f i n e d it, b u t this d e n s i t y , n o n e t h e l e s s , was close b y in an a r e a o f c o m p l e x h o r i z o n t a l h y d r o g r a p h i c
1000 A
A
+
900
1000
1837
N"
B
900
i
E
E 800
g
% 700
700
600
LU 6 0 0
500
Z
I
400 z w o
800
300 20O
100 0 DAY
NIGHT
FRONTAL ZONE
500
>.- 4 0 0 pz
uJ o
300 200 100 0
DAY
NIGHT
PLUME AND SHELF
DAY
N IG H T
F R O N T A L ZONE
D AY
N IG H T
PLUME AND SHELF
Fig 6 Descriptive statistics of the density of fish larvae in and about the Mlssisslpp= River plume front Boxes enclose mterquamle ranges with the number of observations w~thln crosses depict means, horizontal slashes depict medians, and vertical hnes ranges A--September 1987 B--May 1988
1273
Surface accumulanon of larval fishes
1 837
A
130¢
E
lSOO}
1200
1200 r
1100
1100 r
1000
tO00 r
900
900 t
B
800 I"
~ ~oo
700 t
°°°r
,oo
~ soo
500]-
4oor
400 300 200 100
:
-
I
2
3
4
5
7
8
9
200~
,o:[......;,
10
I-
0
1
2
3
4
6
I 7
i 8
I 9
i 10
DIFFUSlVITY (m= sec-~) Fig 7 The accumulation of larval fishes at the surface at the Mississippi River plume front Solid curves depict expected calculated densities o1 larval fishes for the range of apparent convergent velocities (v) and median densities of larvae outside ot the frontal zone (Cv), boxes enclose observed lnterquartile ranges in densities o1 larval fishes within the frontal zone by day (left) and night (right), numbers within boxes indicate the number of observations crosses indicate means horizontal slashes indicate medians, and vertical lines indicate ranges of densities A--September 1987 ( C 0 = 8 2 x 103m 2),B__May1988(C0~63x 103m 2)
structure (Figs 1, 2 and 4) Given the dynamic nature of the plume front, this exceptional density may be a remnant aggregation of larvae accumulated by frontal convergence, then advected away from the frontal zone by stlrrmg Densities of larval fishes within the frontal zone are probably the result of their accumulation along ephemeral convergence zones and subsequent dispersal and mixing durmg relaxation of convergence Expected densmes of larval fishes calculated with the advectlon-dlffus]on model were within the range of observed densities within the large-scale frontal zone, given a length scale (L) of 500 m, background densities (C0) of 82 x 103 m -2 in September 1987 and 62 x 103 m -2 m May 1988 (median densities of larval fishes observed m plume and shelf waters irrespective of time of day), and velocities (V) of 0 05-0 25 m s - t in autumn and 0 15-0 40 m s -1 in spring (the end points of the range of convergent velocmes averaged for both sides of the frontal interface) Ranges of expected densities overlapped the lnterquartlle range of observed densities and lay close to mean and median densmes, especially at dlffuslvIty coefficient values above 5 m 2 s - t (Fig 7) DISCUSSION
The MISSISSIppi River plume shares some slmllarlttes w]th other rtverme plumes It is a buoyant plume with a large-scale frontal zone of complex horizontal hydrographic
1274
J J GovoNi and C B GRIMES
structure where lsopycnals characteristic of the depth discontinuity layer Intersect the surface E m b e d d e d within this large-scale frontal zone are small-scale turbidity discontinuities where convergence is often present The large-scale front resembles the "degenerate fronts" of the Connecticut River plume, small scale fronts resemble " b o u n d a r y fronts" (GARVINE. 1987) Current thought suggests that convergent motion of plume water results from spreading of plume water over ambient shelf waters as well as trom the action ot horizontal pressure gradient forces As such, observed convergent motion of shelf waters is apparent, not real, i e shelf waters are overrun by and submerge under plume water (GARVlNE, 1987) The apparent velocities of convergence observed here, although somewhat higher, are of the same order of magnitude as those calculated by GovoNI et al (1989) under the assumption that convergence was driven m both plume and shelt waters by pressure gradient force Observed convergent velocities also agree with those observed for the smaller Connecticut River (GARVINE, 1977) The Mississippi River plume differs trom other rlverlne plumes in other ways Its larger scale subjects it to the influence ol Corlolls force (WRmHT and COLEMAN~ 1971) unlike the Connecticut River plume (GARVINE, 1987) Its turbidity dlscontlnmtles are more often associated with the inshore margin of the plume rather than the offshore margin, as Is typical ot the Connecticut River plume (GARVlNE, 1987) In profile, it is deeper in the spring than other, smaller plumes Also, it is deeper in spring than it is in autumn or winter (GovoNi et a l , 1989), owing to the vernal freshet (GuNXER. 1979) Our application oi the advectlon-dltfuslon model is a highly simplified mechanistic representation that is based upon several assumptions Densities of fish larvae simulated by the model depend most heavily on the assumed value ot diflUSlVlty, but there is some question concerning appropriate values for dltfUSlVlty within frontal zones While the range of values assumed in our application seem appropriate tor the spatial scale ot observed convergence zones, LEwis (1984) suggested values ot the order of 1(}-100 m 2 s Further, the model assumes that dlffuslvity is symmetrical in the ~- and ~-coordlnate~, but LAM et al (1984) indicate that diffusion along the front may be much greater than it is across the tront With the simplified model, densities dechne precipitously with dfltUSlV~tv values >1 m 2 s ~, and approach an asymptote with values above >5 m 2 s - I (F~g 7) A n o t h e r assumption ot the model is that the population of larval fishes in and about the Mississippi River plume front is stable, l e that spawning of fishes within the trontal zone does not account for greater abundances of larvae there Whereas there is evidence that some fishes, including m e m b e r s ot the Engraulldae c o m m o n to the Mississippi plume (GRIMES and FINUCANE, 1991) spawn within trontal zones (GovoNI, in press), the neuston net, with its 947 ktm mesh, sampled only large larvae, not yolk-sac or small larvae, the products of recent spawning Yet another assumption Is that larval fishes have the capability of countering downward motion at the frontal interface W~thln the context ot OLSON and BACKUS (1985), m a x i m u m downward motion is ~ 3 V / L Given the observed horizontal convergent velocities, downward motion should have ranged trom 0 005 to 0 20 m s -1 at a depth of 5 m or less along the trontal Interface, whereas sustained (as opposed to burst) swimming speeds of larval fishes range from 0 003 to 0 0i m s- 1 (BAILEY and HOUDE, 1989) To the extent that these ranges overlap, fish larvae should be capable of countering downward motion at the frontal interface some of the time With our assumptions, expected densities of surface-dwelling lar~ al fishes at the lront,ll interface, calculated with the advectlon-dlffuslon model, agree in the main with observed densities Frontal convergence can explain why densities of larval fishes were greater within the
Surface accumulation of larval fishes
1275
l r o n t a l z o n e t h a n they were in the a d j a c e n t p l u m e or shelf waters, b u t tt does n o t explain why e x c e p t i o n a l l y high densities were n o t e v i d e n t e v e r y w h e r e within the frontal zone T h e d t s c o n t i n u o u s d i s t r i b u t i o n of larval fishes within the f r o n t a l z o n e is best e x p l a i n e d by the e p h e m e r a l n a t u r e a n d the k i n e m a t i c c o m p l e x t t y of c o n v e r g e n c e z o n e s assoctated with t u r b i d i t y d i s c o n t i n u i t i e s I n w i n t e r , e x c e p t i o n a l densities of larval fishes o c c u r r e d m p r o x i m i t y ( < 100 m) of t u r b i d i t y d i s c o n t i n u i t i e s (GovoNI et al , 1989), b u t varIatton a m o n g serial, replicate surface s a m p l e s t a k e n a l o n g t u r b i d i t y d t s c o n t l n u l t l e s in w i n t e r was htgh b o t h at m e t e r a n d k i l o m e t e r scales This v a r l a t t o n ts p r o b a b l y typical a l o n g these features, a n d m a y e x p l a m why e x c e p t t o n a l densities were n o t o b s e r v e d m the tew, single collections t a k e n in a u t u m n a n d spring (only two of seven single c o l l e c n o n s t a k e n n e a r these features m a u t u m n a n d spring p r o d u c e d e x c e p t i o n a l d e n s m e s ) A c c u m u l a t t o n of larval fishes at the t r o n t a l m t e r f a c e is h m l t e d by the time scale within whtch c o n v e r g e n c e o p e r a t e s (the f o r m a t i o n a n d dlsstpatlon of sharp, small-scale t u r b t d l t y dtscontmuttles) a n d is o p p o s e d by the d i s p e r s i o n of larvae away from the front T h e s e physical features form d u r i n g e b b tide a n d dissipate d u r i n g flood tide a 12 h d i u r n a l ttdal cycle prevails in the n o r t h e r n G u l f ot Mextco (SEIM et a l , 1987) D l s p e r s t o n within the large-scale t r o n t is eftected by t u r b u l e n c e associated with the c o m p l e x i t y of the f r o n t F o r e x a m p l e , m e a n d e r s and eddtes that result t r o m h y d r o d y n a m i c l n s t a b t h t y are c o m m o n a t t r t b u t e s of r i v e r m e p l u m e fronts (S~MPSON a n d JAMES, 1986) F r o m a ship's l a u n c h , o n e of us (J J G ) has also o b s e r v e d m e t e r scale vertical mixing a n d h o r i z o n t a l eddies a l o n g t u r b i d i t y d l s c o n t l n u I t m s T h e spatial d~strlb u t l o n ot larvae w~thln the large-scale f r o n t a l zone is the aggregate result ol the r e p e a t e d / o r m a t I o n a n d d e g e n e r a t i o n of c o n v e r g e n c e zones assocmted with small-scale turb~dlty d~scontmumes A~hnmvledgements--We acknowledge with appreciation the na,dgatlonal skills and practical advice ol CMDR ( S Nelson (U S NOAA Corps) and the essential counsel ot Drs A Bratko,~lch (NOAA Great Lakes Environmental Research Laboratory), D B Olson (Rosenstml School ot Marine and Atmospheric Scmnces Um~ers~tv ot Miami) and W J Wlseman (Louisiana State Um~ersitv)
REFERENCES BAIIE'* K M and E D HOUDE(1989) Predatum on eggs and larvae ot marine fishes and the rccrmtment problem 4dvances m Marzne Biology, 25 1-83 BO,aMANM J and R L IVERSON(1978) Estuarlne and plume fronts In O~eantt/ront;mcoastalpro~e;~es M J BOWM~,Nand W E ESAlaS editors Sprlnger-Verlag Nev~York, pp 87-104 ( URTINT B (1986a) Physical observations m the plume region of the Amazon Rlx er during peak dlschargc--ll Water masses Continental fhelfResearch, 6 56-71 CURSINT B (1986b) Physical observations m the plume region o1 the Amazon River during peak dlscharge--IlI Currents (.ontmental Shelf Research 6, 73-86 CURnNT B and R V LEOECKIS(1986) Ph)sical observations m the plume region ol the Amazon Rixer during peak dlscharge--I Surtace varlabdltV Continental fhelfiResearch 6 31-51 DINNEII S P and W J WtSEM~N(1986) Fresh ,aater on the Lousxana and Texas shell Continental ghel/ Re~earch 6 765-784 G~,RVlNER W (1977) Obser,,atlons of the morion tmld ot the Connectmut Rp~er plume doutmd o: Geoph~ ~ltal Research 82 441-4-54
G~RXlt~ER W (1986) The role of brackish plumes m open shelf waters In The role of fre;h.ater outfloi, tn coastalmarmeecosystems, S SI.REStET,editor Sprmger-Vt.rlag pp 47-65 GARWNE R W (1987) Estuary plumes and tronts In shell waters a laver model Journal o: Physical Oceanography 17 1877-1896 Govosi J J D E Hoss and D R COLB~(1989) The spatial distribution ol larval fishes about the. Mississippi Rl~er plume Lmlnologv and Oceanographx, 34 178-187
1276
J J GOVONI and C B GRIMES
GOVONLJ J (In press) Flux oflarval fishes across frontal boundarles examples from the MlsSlsslppl Rlver plume front and the western Gulf Stream front In winter Contrtbutlons tn Sctence Natural Htstor~ Museum of Lo~
Angeles GRIMESC B a n d J H FiNUCANE(1991)Spatlaldlstrlbutlonandabundanceofichthyoplankton, chlorophylland macrozooplankton around the Mississippi River d~scharge plume and the role of the plume in fish recrmtment Marme Ecology Progress Serws, 75,109-119 GUNTER G (1979) The annual flows of the Mississippi River Gulf Research Reports 6 283-290 LAM D C L C R MURTHY and R B SIMPSON (1984) Effluent transport and dl]]Uslon model~]or the coastal zone Sprlnger-Verlag, New York, 168 pp LAP1CQUE G and J PH BREI~ME~ER (1973) A t t e m p t to evaluate the extension in the area of pollution of n v e r origin and application to some Mediterranean rivers Revue lnternattonale d'O¢eanographte Medlcale 30 61-87 LEWIS R E (1984) Circulation and mixing in estuary outflows Continental Shelf Research 3 201-214 LUKETIN~. D A and J IMBERGER(1989) Turbulence and entrainment in a buoyant surtace plume Journal o] Geophwwal Research, 94, 12619-12636 MVRTH~ C R (1976) Horizontal diffusion characteristics in Lake Ontario Journal of Ph~ steal O~eanograph~, 6 76-84 NEILSON J D and R I PERRY(1990) Dlel vertical migrations of marine hshes an obligate or tacultatlve process Advances en Marme Biology, 26, 115-168 O'DONNEL J (1990) The formation and fate of a river plume a numerical model Journal o¢ Phwlcal Oceanography, 20, 551-569 O K w o A (1976) R e m a r k s on the use of "diffusion diagrams ' m modeling scale-dependent dlttUslOn Deep-,Sea Research, 23, 1213-1214 O1 SON D B and R H BACKUS (1985) T h e concentrating of organisms at lronts A cold-water fish and a warmcore Gulf stream ring lournal of Marine Research 43, 113-137 SABATESA (1990) Changes m the heterogeneity of mesoscale &strlbutlon patterns o1 lar~ al fish associated with a shallow coastal h a h n e tront E~tuarme Coastal and Shelf Sctence 30 131-14(/ SHAPIRO S ~ and M B WILK (1965) Analysis of variance test for n o r m a h t ) (complete samples) Btometttka 52 59l~11 SLIM H E B KJERFVE and J E SNEED (1987) Tides of Mississippi Sound and the adjacent continental shelf Evtuarme Coastal and Shelf Sctence, 25, 143-156 SIMPSONJ H and I D JAMES (1986) Coastal and estuarlne fronts In BaroHmtcproces~es on continental ~hel~es C N K MOOERS editor, A m e r i c a n Geophysical U n i o n Washington, pp 63-93 WRIGHT L D and J M COLEMAN (1971) Effluent expansion and lnterfaclal mixing m the presence oi a salt wedge Mississippi River Delta Journal of GeophvstcalRevearch 76 8649-8661 ,
