1NGLESPE'IFS DOMINANCE IN A SUBSURFACE PHYTOPLANKTrONCONCNTRAIONAT A MEDITERRANEAN SEA FRONT

RIC' 1AiR XV. GOULD, JR. ANr) DENIS A. WIESENBURG

I r mrJ r~n LJMNt)L C ANP, 0O. "N4WJ'AP41Y

CyK4 c 11)00 by'y Ih Atw rio S. cicty of Linmwo tljW $ h

REPORT DOCUMENTATION PAGE Form ApprveI OMB No 0704-0188

Pubsi Wootigbudn Ior this c01nllofnoaon W80 wkimte toaeae1 orpr resos nluding the Ime fo e I ns ntructions. eacigexisting daasouc

iWi collectian ol Informatia4% including suggestions fo euigthis burden. So Waat ngton Hedurtr eri .Directorate for inlornatlon operaos" and AeponeS. 115JefesnDavis Highway, Suite 1204. Adlington. VAI22202-4302. and to the Office ot Management and Budget, . Papeaor tAeducion P aect tg-Om1ias. WeaVngon. DC 20 3.

1. Agency Use Only (Leave blank). 2. Report Date. 3. Report Type and Dates Covered.1 1990 Journal Article

4. Title and Subtitle. S. Funding Numbers.Single-species dominance in a subsurface photoplankton pmOmW Elemnt Na 61153concentration at a Mediterranean seafront

Prol n. 031076. Author(s).

Task No. 330Richard W. Gould and Dennis A. Wiesenburg Asn

_N494470

7. Performing Organization Name(s) and Address(es). 8. Performing OrganizationReport Number.

Naval Oceanographic and Atmospheric Research Laboratory* JA 333:025:89Stennis Space Center, MS 39529-5004

9. SponsodnglMonitoring Agency Name(s) and Address(es). 10. SponsoringlMoniodng AgencyReport Number.

Naval Oceanographic and Atmospheric Research Laboratory* JA 333a025:89Basic Research Management OfficeStennis Space Center, MS 39529-5004

11. Supplementary Notes.*Formerly Ntval Ocean Research and Development Activity

12a. Distribution/Availability Statement. 12b. Distribution Code.Approved for public release; distribution is unlimited.

. Abstract (Maximum 200 words).A narrow band of high Chl a (23.1 ug liter-') was observed at a salinity frontin the western Mediterranean Sea in late November 1987. The biomass peak was founddeep in the photic zone, at 54 m, in a region of low light. A single diatom species,Thalassiosira partheneia Schrader in gelationous colonies, represented 98% of thetotal phytoplankton biomass in the layer and achieved abundance9.8 x 10.,cellsliter- .Although lack of temporal sampling preclbdW -0e-i-s-edetermination of theprocesses responsible, the high biomass accumulation and dominance by this specieswas likely due to its preference for low light coupled with turbulence-induced highnutrient levels. Interleaving patterns in the temperature, salinity, and nutrientprofiles suggest increased horizontal advection at the front. High shear at theboundaries of the interleavings might confine the biomass to a thin band, as wellas generate turbulence to mix nutrients in from surrounding layers. Alternatively,stabilization of the water column following a brief, pulsed upwelling event couldhave reduced dispersion of the biomass, thereby confining it to the thin layer weobserveg0.)

14 fTrmdr. .. '- 15. Nu er of Pages.

Ocean Environmots; (IJY Phytoplankton: ( Chemical 16. Pi Code.Oceanography; U) Ocean Currents, I) C

17. Security Classification 18. Security Classification 19. Security Classification 20. Limitation of Abstract.of R f Ibi Abstract.of R assli f ied n cilssif ied Unclassified SAR

NSN 7840-.1-280-50 Standard Form 298 (Rev. 2-89P.scnbe4 by ANS St ZV 't

Notes 211

Lmno. Oceanor., 35(l), 1990, 211-2200 1990, by the American Society of Limnoloy and Oceanography, Inc.

Single-species dominance in a subsurface phytoplanktonconcentration at a Mediterranean Sea front

Abstract-A narrow band of high Chl a (23.1 ber 1987, we observed a Chl a maximumjig liter-') was observed at a salinity front in the concentration of 23.1 ug liter-' in a narrowwestern Mediterranean Sea in late November band at 54 m in the Almeria-Oran Front in1987. The biomass peak was found deep in the

photic zone, at 54 m, in a region of low light. A the western Mediterranean Sea. Subsequentsingle diatom species, Thalassiosira partheneia microscopic analysis revealed a gelatinousSchrader in gelatinous colonies, represented 98% colony-forming diatom (Thalassiosira par-of the total phytoplankton biomass in the layer theneia Schrader) as the overwhelminglyand achieved abundances > 9.8 x 106 cells liter-'.Although lack of temporal sampling precludes dominant species with an abundance >9.8precise determination of the processes responsi- X 106 cells liter- '. What factors enabled thisble, the high biomass accumulation and domi- single species to attain such a concentra-nance by this species was likely due to its tion?preference for low light coupled with turbulence- Biological, chemical, and physical factorsinduced high nutrient levels. Interleaving pat-terns in the temperature, salinity, and nutrient must be examined in any attempt to answerprofiles suggest increased horizontal advection at the above question. Although it is likely thatthe front. High shear at the boundaries of the several factors were responsible, it is im-interleavings might confine the biomass to a thin portant to assess the relative importance ofband, as well as generate turbulence to mix nu-trients in from surrounding layers. Alternatively, each. In terms of the biology, species inter-stabilization of the water column following a brief, actions and the affect of competition mustpulsed upwelling event could have reduced dis- be considered, as must grazer control andpersion of the biomass, thereby confining it to nutrient concentrations and distributions.the thin layer we observed. The nature of the physical environment with

regard to turbulence, water mass interleav-In an environment where multispecies as- ing, and convergence/divergence regimes

semblages are common and the "paradox must also be determined to assess their rolesof the plankton" is the norm (Hutchinson in controlling the observed distribution.1961), it is unusual to find an open ocean This biomass accumulation was unusualregion dominated by a single species. Al- for several reasons. First, it was totally dom-though phytoplankton "blooms" are wide- inated by a single species. Second, such highspread and have been reported in coastal cell numbers are rarely observed in the openareas, upwellings, enclosed bays, and during ocean-even in a bloom-and have not, tospring stratification, a combination of our knowledge, been reported for the Med-somewhat atypical conditions is required for iterranean away from the coast. Third, thebloom formation (Paerl 1988). In Novem- biomass-rich layer was very narrow-only

about 6 m thick. Fourth, the bloom wasAcknowledgments found late in the year (this station was oc-

We thank Irene P. DePalma for providing the Chi a cupied on 29 November) when there wasdata and Mark Spears and R. Glenn Casey for the little thermal stratification. Finally, the bio-nutrient analyses. Numerous discussions with Greta mass was concentrated deep in the waterFryxell concerning the ecology ofdiatom colonies were column and in very low light, which is dis-very helpful. cussed later.

This research was conducted as part of the ChemicalDynamics in Ocean Frontal Areas program ofNORDA A strong salinity gradient stretches nearlyand funded by the Office of Naval Research, Program from Almeria, Spain, to Oran, Algeria, inElement 61153N, through the NORDA Defense Re- the Alboran Sea in the western Mediterra-search Sciences Program. During this work R.W.G. nean, creating the Almeria-Oran Front (A-was supported by a National Research Council Post- gdoctoral Associateship. This document has been re- 0 Front). This front results from less salineviewed and approved for public release. NORDA Con- Atlantic water flowing in through the Straittribution 333:025:89. of Gibraltar and converging with more sa-

212 Notes

~SPAIN

3/3'

, ..JO FRONT J

36o

Fig. 1. Station locations in the western Mediterranean during November 1987 in relation to the Almeria-Oran Front. The Chl a value of 23.1 gg liter-' was recorded at 54 m at station 32. The position of the frontwas determined from CTD transects during the cruise.

line Mediterranean water. The Alboran Ba- gion with a high-density sampling strategy.sin contains twin anticyclonic gyres and the Component investigations included phy-A-O Front is the eastern boundary of the toplankton species composition, abun-eastern gyre. The front extends to a depth dance, biomass, and productivity; activityof 200 m, but the upper 60-75 m are un- of the respiratory electron transport systemstable and meanders are common. The front (ETS); bacterial abundance and production;is present year-round and surface currents Chl a distribution; nutrients; temperature;there average 40 cm s- . The thermal con- salinity; oxygen; and optical and meteoro-trast across the front is not large, only 1V- logical data. Satellite imagery collected at20C, but the two-salinity-unit change over the Naval Ocean Research and Develop-a distance of 2-15 km creates a sharp den- ment Activity (NORDA) and telemeteredsity boundary (Arnone et al. 1990). Pre- to the Lynch provided both real-time datavailing winds in the vicinity of the front are for locating the frontal boundary and a syn-generally light but variable in both direction optic view of the temperature field in theand intensity. Atmospheric fronts move area. The NORDA towed underwaternorthwest to southeast. The A-O Front does pumping system (TUPS) was used to exactlynot represent a direct wind response, but locate the front and to map surface featuresthe large-scale wind stress of the western (Rein et al. 1985). A complementary cruiseMediterranean probably does affect frontal to the same region in May 1986 examinedstrength and location (G. Heburn pers. the same properties in spring (Lohrenz etcomm.). al. 1988b).

Samples were collected in the Alboran Sea The TUPS consists of a 2.4-m-long towfrom 14 November to 3 December 1987 body containing a SeaMarTech, Inc., fluo-from the USNS Lynch. Station locations are rometer and Sea Tech, Inc., beam trans-indicated in Fig. I relative to the A-O Front, missometer (670-nm operating wavelength,but only station 32 is labeled-the frontal 25-cm pathlength); Sea Bird, Inc., CTD sen-station where the large biomass accumula- sors; and Biospherical Instruments upwell-tion was observed. The goal of the study ing and downwelling irradiance sensorswas to assess the biological, chemical, and (model QSP 200). The system was towed 2physical properties in a dynamic frontal re- m below the surface at 400 cm s-I and data

Notes 213

16-

DNCE (1(m)Fig. 2. Temperature (.) and salinity (-) at 2 m along west-to-east TUPS transect perpendicular to

Almeria-Oran Front (transect indicated in Fig. 1). The location of station 32 is noted.

were collected every 12 s, which equated to et al. (1988b). For phytoplankton samples,one measurement about every 50 m hor- glutaraldehyde was added to a final concen-zontally. tration of 1% for whole-water samples and

The initial protocol for locating and map- 2% for net tows. Samples were kept refrig-ping the front was to tow the TUPS for 6- erated until analysis back in the laboratory8 h along a straight transect perpendicular at NORDA. The discrete water samples wereto the front. XBTs were launched every 30 settled and counted as by Gould and Fryxellmin or less along the transect, as well. Then, (1988) with a Zeiss IM-35 inverted lightafter plotting temperature, salinity, and flu- microscope. In addition, cell sizes and areasorescence along the transect, features of in- were measured for subsequent biovolumeterest at the front and on both the Atlantic and biomass estimates with an attachedand Mediterranean sides were selected for drawing tube in conjunction with a Sum-immediate, detailed vertical stations. Tem- magraphics model MM 1201 digitizing tab-perature and salinity data from a TUPS let and Zenith 248 microcomputer. Phyto-transect through the A-O Front are shown plankton biomass was calculated from cellin Fig. 2. The location of station 32 in close biovolume with the modified Strathmannproximity to the sharp salinity gradient is equations (equations 7 and 8, Smayda 1978).noted. After arriving back at a site, there Incident surface solar flux was measuredwere two CTD casts (Neil Brown Instr., Inc., with an Eppley pyrheliometer and convert-Mark IIIB CTD)-a deep one (to 500 m) edfrom jWcm

- 2 toEinstm- 2 s-I followingand a shallow one (to 100 m) with a fluo- Parsons et al. (1977). The midday flux onrometer (or transmissometer) and submers- 30 November (the day after station 32) wasible pump (Berkeley model 4BM 12) at- 65,000 ;LW cm - 2. The light level at 54 mtached. A 35-ism-mesh phytoplankton net was estimated from 1, = I0exp(-kz), withwas attached to the hydrocable during the an attenuation length (I/k) of 15 m for Med-deep cast for a vertical net tow in the upper iterranean water provided by Lohrenz et al.125 m. (1988a). The mean cell division rate (jsF,

The temperature, salinity, and fluores- doublings d- 1) of T. partheneia was calcu-cence profiles for the down-cast of the shal- lated in the 54-m sample from the propor- 0-low CTD were examined to select 12 sam- tion of dividing cells (Weiler and Chisholm [3pling depths for the up-cast. Phytoplankton, 1976; Rivkin and Voytek 1986):bacterial, Chl a, and nutrient samples were 0.5A + Btaken from the pump/hose system. Chl a F +B+was measured with a Turner fluorometer 0.5A + B + Cfollowing slightly modified methods of whereA is the numberof single cells recentlySmith et al. (1981), as described by Lohrenz divided, B the number of paired cells, and 3d"

Dist Specal

c.PVoI - .

"lap--

214 Notes

% TRANSMISSMO CONCENTRArION60 68 76 84 92 100 NO, Si(OH)4 QM)I I I I I I

0 1 2 3 4 5TB§UMRE (OC)

13 14 15 16 17 0

20 i 40

40 ~60-4080

~60-

80 **, . 120-

0 10 20 30

M3 7 O (,AM)IFig. 4. Silicate, nitrate, and phosphate concentra-

100- tions vs. depth at station 32.

I LMTY ing mitosis. Also, paired cells might have% TPNSMISIONdislodged during sample preparation (shak-

120'' ing), and they are difficult to distinguish inS?.2 38 valve view. This method cannot detect di-

Fig. 3. Temperature, salinity, and percent transmis- vision rates >1I because there is no way tosion vs. depth at station 32. know if a cell has divided already that day.

The estimate must therefore be consideredC the number of single cells that have not a "minimum division rate."~recently divided. However, for diatoms this Temperature, salinity, and percent trans-microscopic method cannot distinguish A mission at station 32 are shown in Fig. 3.and C, so the equation reduced to The water column is well mixed to 45 m,

but interleaving of water masses is apparentF = B from 50 to 65 m. The surface salinity of

B + D 37.49 indicates that the station fell just onwhere D is the total number of single cells, the Mediterranean side of the front. Theand temperature minimum at 61 m was coin-

cident with a transmission minimum of 65%OF= ln(1 + F~) and marked the location of the Chl a peak.

MF In 2 x t The data plotted are from the down-cast ofthe CTD; by the time we sampled from the

where t is I d. hose during the up-cast, the minimum inThe estimate provided is very likely an percent transmission had shifted to 54 m,

underestimate of the true division rate be- probably as a result of internal wave activitycause several problems and assumptions are or spatial variability associated with shipinvolved with the technique. For example, drift near the frontal boundary. We adjustedbecause we did not have a time series of our sampling accordingly to sample in andobservations, we must assume that our ob- around the Chl a peak by using both the inserved F is equal to the maximum daily F situ transmissometer and a flow-through

and that it is an estimate of all cells undergo- fluorometer in the lab connected to the pump

Ii

Notes 215

BIOMASS (QA C 1)

101 1 10 102 103

0-

-20-

-60-

o DrosM-100 ,a Coccos

Fig. 6. Phytoplankton biomass vs. depth by groupat station 32. Note log scale for biomass.

Vertical phytoplankton net tows were ex-amined from the 32 stations shown in Fig.1 (several stations sampled on different dayswere nearly coincident in space). Gelatinouscolonies of T. partheneia (Fig. 5) were ob-served at every station, but their abundancevaried widely. The colonies were mostabundant at stations along the front. De-tailed discussion of the distribution of thisspecies and others will be presented else-where in conjunction with additional phys-ical and chemical data (Gould et al. in prep).

Fig. 5. Thalassiosira partheneia colony. A and B. Figure 5A gives an impression of the sizePhase contrast illumination. C. Nomarski differential of the colonies; they were several millime-interference-contrast illumination. A. Orientation view ters long and clearly visible with the nakedof a colony with several pennate diatoms interspersed eye. Several bicapitate pennate diatomson the threads. B. Higher magnification of the colonyshowing numerous interconnecting threads. C. Detail (Nitzschia capitata Heiden and Kolbe?) wereof cells indicating delicate frustule structure and vac- interspersed on the threads of the colonies.uous nature. The numerous interconnecting threads ap-

pear as dark lines in Fig. 5B. The high mag-line. This allowed us to place the CTD/pump nification of Fig. 5C shows some of the de-directly in the very narrow Chl a maximum. tail of the cells and their vacuous nature.

Interleaving of water masses in the vicin- The diameters were generally 8-10 jim.ity of the front is also apparent from the The biomass vs. depth profiles for dia-nutrient profiles in Fig. 4. Silicate, nitrate, toms, dinoflagellates, coccolithophorids, andand phosphate all show a similar "saw- other algae at station 32 are shown in Fig.tooth" pattern from 40 to 60 m. It is of 6. There are major differences in speciesinterest that the Chl a maximum at 54 m composition between the six sample depths.corresponded to a relative peak in nutrients, All four groups exhibited increased biomasssuggesting that the biomass there might not around 50-55 m, but particularly the dia-have peaked at the time of sampling but was toms. The diatom maximum also was aboutstill increasing. 5 m deeper than the maxima for the other

216 Notes

Table 1. Depth distribution of Thalassiosira par- very similar to the ranges of 2-6 and 1-100theneia abundance (cells liter-') and biomass (Ag C fg Chl a /m -3 reported by Jimenez et al.liter-') at staion 32. (1987) and Nicholis and Dillon (1978), re-

T. pa,,heneta % of to,-, a spectively. Although total cell numbers (and

Wm) Abundance Biomus Abundance Bioma subsequently biomass, which was derived10 31,000 2.00 11.2 14.2 from cell numbers and biovolumes) and Chl

30 90,700 5.85 37.5 36.6 a varied widely over the depth range ex-

50 1,690,000 109.12 70.4 85.2 a;.iined, the Chl a /m - 3 and C Am- 3 ratios54 9,860,000 635.98 92.1 98.3 remained remarkably stable in the upper 5465 1,630 0.10 1.3 4.2 m, but both increased in the deeper two80 787 0.05 1.6 4.9 samples (65 and 80 m).

The increased C Mm- 3 ratio reflected agroups and decreased sharply within the next taxonomic shift. The proportion of diatoms10 m. in the surface samples was considerably

Table I provides the depth distribution greater than at 65 and 80 m, where smallof T. partheneia abundance and biomass at monads dominated (coccoid cyanobacteriastation 32. Note the dramatic increases in and eucaryotes). The larger proportion ofthose parameters from 30 to 54 m and the nondiatom biomass at depth caused the Ceven more dramatic decreases below 54 m. Am- 3 ratio to increase because the modifiedThat single diatom species represented an Strathmann equations estimate less carbonincreasing proportion of the cell numbers per cell for diatoms compared to other phy-and biomass from the surface to 54 m, where toplankton cells of the same size, due to theit accounted for 92% of the total phyto- larger vacuole volume of diatoms. Size-fre-plankton cell numbers and 98% of the total quency analysis confirmed the shift tobiomass. The light level at the maximum smaller cells at depth. In the upper fourwas low, about 81 #Einst m- 2 s - 1, 2.7% of samples, the percent of the total biovolumethe surface flux. Yet, the division rate of T. that had an equivalent spherical diameterpartheneia estimated from the proportion of -<5 Mm ranged from 0.3 to 3.0%, whileof dividing cells was 0.20 doublings per day. in the 65- and 80-m samples the percentagesThis does not indicate tremendous produc- were 23.4 and 29.9%. The increase in Chltion, but does indicate, along with the a Jm - 3 suggests that the cells might havehealthy appearance of the colonies and cells, been exhibiting a photoadaptive responsethat the population was by no means se- by increasing the amount of Chl a in thenescent. Total integrated phytoplankton cell cell at lower light levels. The C: Chl a valuesnumbers and biomass at station 32 were above 54 m are close to an empirically de-1.21 x 10"1 cells m - 2 and 7.03 g C m- 2 rived estimate of 23.4 for another Thalas-(integrated to 80 m, 0.5% light level). siosira species at a light level of 70 jsEinst

Total phytoplankton abundance, bio- m - 2 s- I (from data of Falkowski et al. 1985),mass, and Chl a vs. depth followed much suggesting that the Uterm6hl settling meth-the same pattern as T. partheneia (Table 2). od could detect most of the biomass in theseChl a per unit biovolume, C per unit bio- samples, at least in the upper 54 m.volume and C: Chl a ratios are also given. Although both C and Chl a per unit bio-The Chl a per unit biovolume values are volume increased in the 65- and 80-m sam-

Table 2. Total algal abundance (cells liter'), biomass (ag C liter-'), Chi a (ag liter-'), and ratios of Chl aand C jm - of biovolume by depth at station 32.

Depth Chl asm- C $M- 5m) Abundance ioman Chi a (FA) W : Chl a

10 276,000 14.11 0.60 3.3 76.5 23.530 242,000 16.04 0.67 3.9 93.6 23.950 2,400,000 128.40 2.78 2.1 95.2 46.154 10,700,000 647.02 23.12 3.3 92.9 28.065 124,000 2.48 0.22 11.8 133.4 11.380 49,400 1.04 0.09 13.3 153.9 11.6

Notes 217

pies, there was a proportionately larger in- briichter and Boje (1978) also showed lightcrease in Chi a, resulting in lower C: Chi a inhibition.ratios. Alternatively, the lower C: Chl a ra- The importance of a colony habitat to thetios at depth might have been due in part diatoms and to the water column trophicto some "missing biovolume" (and subse- structure is not clear. Colony formation ev-quently carbon) associated with the higher idently occurs with healthy cells, as the pho-proportion of picoplanktonic cells in those tosynthetic activity of single cells is less thansamples. Perhaps some of the small cells did colonies, and the colonies disintegrate asnot settle in the chambers before counting, they age or if they are light damaged (El-were too small to be seen, or were destroyed brfichter and Boje 1978). In the Northwestby preservation, but were retained on the Africa upwelling area, T. partheneia colo-GF/F filters during the Chi a determina- nies represented from 6 to 48% of the pri-tions. However, even if enough C was added mary production at different stations andto the 65- and 80-m samples to adjust the light depths and from 6 to 33% of the totalC: Chi a ratios to 25, the increased bio- water-column production (Elbrichter andvolume in the samples would decrease the Boje 1978).Chi a Um

- 3 ratio to just 4.4 fg. This rela- Several hypotheses have been advancedtively high ratio still suggests that photoad- suggesting possible advantages and disad-aptation, not loss of small cells, was re- vantages of a colony lifestyle. For example,sponsible for the lower C: Chi a ratio. nutrient availability to individual cells may

About a dozen species of the diatom ge- be increased or decreased, depending onnus Thalassiosira have been reported to whether flow through the matrix is in-form gelatinous colonies (Fryxell et al. 1984), creased or decreased. If the colonies are anand such colonies have been reported from adaptation to increase sinking speed (Sme-a number of locations, including upwelling tacek 1985), which according to Smaydaareas, warm-core rings, near the Antarctic (1970) is a means of contacting more nu-ice edge, and now at a Mediterranean salin- trient-rich water and enhancing nutrient up-ity front (Elbriichter and Boje 1978; Fryxell take, increased water column turbulenceet al. 1984; Youngbluth and Paffenh6fer may be needed to resuspend the colonies.1987; Fryxell and Kendrick 1988). Al- The gelatinous matrix might also act as athough a variety of environments is repre- chelator to complex toxic substances or trap-sented, all seem to have two factors in com- required trace metals. Grazers fed moremon, at least at the time when the colonies readily on single cells of T. partheneia thanwere observed: turbulence and high nu- on colonies, so there may be some protec-trients. In the warm-core rings, which are tion against filter-feeding predation (Schnackgenerally considered oligotrophic, colonies 1983). Elbrichter and Boje (1978) observedwere frequently observed after storms that numerous heterotrophic dinoflagellates, cil-mixed the water column and brought nu- iates, and amoebae within healthy colonies,trients into the euphotic zone (Fryxell and but we did not observe such associations inGould 1983). Also, at least for T. parthe- this study nor were they observed in theneia, the colonies have been reported to be warm-core ring colonies (Fryxell et al. 1984).light inhibited and exhibited greater pri- Other than cells of T. partheneia, only bi-mary production at the 50% light depth capitate pennate diatoms were observed incompared to the 100% light depth (Elbrich- the colonies. Bacteria present in the coloniester and Boje 1978). In warm-core ring 82- may be using dissolved organic carbon re-E, sampled in August 1982, T. partheneia leased by the diatoms, or they may be pro-colonies were observed at ring center with viding remineralized nutrients to the dia-maximum cell numbers ofabout 40,000 cells toms. Bacteria have been shown to beliter-' at 59 m and decreasing abundances important colonizers of other types of ma-shallower (T. P. Watkins unpubl. obs.). rine snow and aggregates (Alldredge andWhen isolated colonies were placed in a Youngbluth 1985). Large colonies also ab-constant-light incubator on board ship, the sorb light less efficiently, which may protectcells bleached and died within -2 d(R.W.G. against photoinhibition (Kirk 1983). Mar-pers. obs.), and culture experiments by El- galef (1978) views the mucous secretion as

218 Notes

a self-regulating control on nutrient absorb- blings per day, and the interleaving patterntion to prevent population crashes. in the temperature, salinity, and nutrient

In terms of water-column processes, the profiles support this possibility. However,colonies act as scavengers that rapidly re- the abundance maxima of T. partheneia atmove materials through sedimentation, even nearby stations were generally around 40 mwithin a few days (Billett et al. 1983: Lampitt (Gould et al. in prep.), suggesting that the1985; Youngbluth and Paffenh6fer 1987). biomass accumulation still formed in fairlyAs mentioned, the matix may be unpalat- low light levels, even if it was advected toable or too large for filter feeders and this station 32.may affect abundances and species com- We believe most evidence supports theposition in higher trophic levels. Clearly, following scenario. First, the preference ofmuch experimental work with cultures is this species for low light levels enabled it tostill required to elucidate the advantages of increase in abundance late in the year anda colony growth habit. at depth in the water column. Second, the

Except for reports of blooms related to growth was probably stimulated by en-river input and manmade effluent near the hanced turbulence and associated nutrientMediterranean coast and in semienclosed flux from below the thermocline. High shearbays (UNESCO 1988), other studies in the zones at the boundaries of the interleavingseastern and western Mediterranean during might confine the biomass to a thin band,different seasons have reported low Chi a as well as generate turbulence to mix nu-values (Furnestin 1973; Dowidar 1984; Es- trients in from surrounding layers. Or, sta-trada 1982, 1985). In fact, Chi a values of bilization after a brief, pulsed nutrient-in-23.1 /Ag liter-' in the open ocean are rare jection event could have reduced dispersionand typically from highly productive coastal of the biomass, thereby confining it to a thinupwelling regions such as those off Peru or layer, as we observed. The intermittent na-Africa (Andrews and Hutchings 1980; Es- ture of enhanced production at fronts hastrada and Marras6 1987; Brown and Hutch- been pointed out by others (Pingree et al.ings 1987). 1978; Richardson et al. 1986; Lohrenz et

With the lack of temporal sampling coy- al. 1988a) and suggests that some type oferage, it is difficult to determine precisely pulsed upwelling may indeed be occurring.which factors enabled this single species to At the Almeria-Oran Front the low light,attain such a concentration. The biomass high turbulence, high nutrient regime gaveaccumulation was due to either a hydro- T. partheneia a competitive advantage thatdynamic concentration of cells or growth. was exploited to attain very high biomassIf growth, the accumulation developed in at depth in an open ocean region. Thalas-situ or was advected. siosira colonies are good indicators of tur-

A simple trapping of cells through flow bulent, nutrient-rich areas. The inferencesconvergence evidently was not the cause be- that can be made from their presence incause all species would have been equally dynamic, marine environments under-concentrated, which was not the case. Such scores the importance of microscopic ex-a hydrodynamic control cannot be ruled out amination of samples in conjunction withentirely, however, if the convergence could pigment analyses.have somehow selectively concentrated the Richard W. Gould, Jr.larger colonies. Microscopic examination Denis A. Wiesenburg2

and determination of the growth rate of T.partheneia indicated that the population was Oceanography Division, Code 333healthy and not just an accumulation of se- Naval Ocean Research and Developmentnescent cells. Activity

Advection of a dense population in a nar- Stennis Space Center, Mississippi 39529row layer parallel to the front could accountfor the observed distribution; high cell num- P a MAR, Inc., 101 NW Interchange,bers near the front at stations to the north Bay st. Louis, Mississippi 39520.and south of station 32 (Gould et al. in prep.), 2 Present address: Department of Oceanography,the relatively slow division rate of 0.2 dou- Texas A&M University, College Station 77843.

Notes 219

References HUTCHiNSON, G. E. 1961. The paradox of the plank-ton. Am. Nat. 95: 137-145.

ALLDREDOE, A. L., AND M. J. YouNGaLUTH. 1985. JImENEz, F., J. RoDRIG UEZ, B. BAUTISTA, AND V.The significance of macroscopic aggregates (ma- RODRoUZ. 1987. Relations between chloro-rine snow) as sites for heterotrophic bacterial pro- phyll, phytoplankton cell abundance and biovol-duction in the mesopelagic zone of the subtropical ume during a winter bloom in MediterraneanAtlantic. Deep-Sea Res. 32: 1445-1456. coastal waters. J. Exp. Mar. Biol. Ecol. 105: 161-

ANDREWS, W. R. H., AND L. HutrcHrNos. 1980. Up- 173.welling in the southern Benguela Current. Prog. KIRK, J. T. 0. 1983. Light and photosynthesis inOceanogr. 9:1-81. aquatic ecosystems. Cambridge.

ARNoNE, R. A., D. A. WIESENBURO, AND K. D. LAMprrr, R. S. 1985. Evidence for the seasonal de-SAUNDERS. 1990. The origin and characteristics position of detritus to the deep-sea floor and itsof the Algerian Current. J. Geophys. Res. 95:1587- subsequent resuspension. Deep-Sea Res. 12: 885-1598. 897.

BILLTr, D. S. M., R. S. LAMpiT', A. L. RicE, AND R. LOHRENZ, S. E., R. A. ARNONE, D. A. WIESENBURG,F. C. MAwrouRA. 1983. Seasonal sedimentation AND I. P. DEPALMA. 1988a. Satellite detectionof phytoplankton to the deep-sea benthos. Nature of transient enhanced primary production in the302: 520-522. western Mediterranean Sea. Nature 335: 245-247.

BROWN, P. C., AND L. HtrrcHINGS. 1987. The de- - , D. A. WIESENBURG, I. P. DEPALMA, K. S.velopment and decline of phytoplankton blooms JOHNSON, AND D. GUSTAFSON. 1988b. Interre-in the southern Benguela upwelling system. 1. lationships among primary production, chloro-Drogue movements, hydrography and bloom de- phyll, and environmental conditions in frontal re-velopment. S. Afr. J. Mar. Sci. 5: 357-391. gions of the western Mediterranean Sea. Deep-Sea

DOWIDAR, N. M. 1984. Phytoplankton biomass and Res. 35: 793-810.primary productivity of the southeastern Medi- MARGALEF, R. 1978. Life-forms of phytoplankton asterranean. Deep-Sea Res. 31: 983-1000. survival alternatives in an unstable environment.

ELBRACHTER, M., AND R. BOJE. 1978. On the eco- Oceanol. Acta 1: 493-509.logical significance of Thalassiosira partheneia in NICHOLIs, K. H., AND P. V. DILLON. 1978. An eval-the northwest African upwelling area, p. 24-31. uation of phosphorus-chlorophyll-phytoplanktonIn R. Boje and M. Tomczak [eds.], Upwelling eco- relationships for lakes. Int. Rev. Gesamten Hy-systems. Springer. drobiol. 63: 141-154.

EsTRADA, M. 1982. Phytoplankton of the western PAERL, H. W. 1988. Nuisance phytoplankton bloomsMediterranean at the beginning of autumn. Int. in coastal, estuarine, and inland waters. Limnol.Rev. Gesamten Hydrobiol. 67: 517-532. Oceanogr. 33: 823-847.

1985. Deep phytoplankton and chlorophyll PARSONS, T. R.. M. TAKAHASHI, AND B. HARGRAVE.maxima in the western Mediterranean, p. 247- 1977. Biological oceanographic processes, 2nd ed.277. In M. Moraitou-Apostolopoulou and V. Pergamon.Kiortsis [eds.], Marine Mediterranean ecosystems. PINOREE, R. D., P. M. HOLLIGAN, AND G. T. MARDELL.Plenum. 1978. The effects of vertical stability on phyto-, AND C. MARRASt. 1987. Phytoplankton bio- plankton distributions in the summer on the

mass and productivity off the Nambian coast. S. northwest European Shelf. Deep-Sea Res. 25:1011 -Afr. J. Mar. Sci. 5: 347-356. 102g.

FALxowsiu, P. G., Z. DutiNsKY, AND K. WyMAN. 1985. REIN, C. D., D. A. WIESENBURG, AND D. M. LAvoIE.Growth-irradiance relationships in phytoplank- 1985. A towed instrument vehicle for deep oceanton. Limmnol. Oceanogr. 30: 311-321. sampling. Naval Ocean Res. Develop. Activity,

FRYXELL, G. A., AND R. W. GouLD, JR. 1983. Field NORDA Rep. 90. Stennis Space Center.observations on gelatinous colonies of diatoms RicHARDSON, K., M. R. HEATH, AND S. M. PEDERSEN.from six cruises to Gulf Stream warm core rings. 1986. Studies of a larval herring (Clupea harengus[abstract.] J. Phycol. 19(suppl.): 8. L.) patch in the Buchan area. 3. Phytoplankton, , AND T. P. WATKINS. 1984. Gelati- distribution and primary productivity in relation

nous colonies of Thalassiosira in GulfStream warm to hydrographic features. Dana 6: 25-36.core rings, including T. fragilis, sp. nov. Br. Phy- RIVKIN, R. B., AND M. A. VoYTEK. 1986. Cell divi-col. J. 19:141-156. sion rates ofeucaryotic algae measured by tritiated

AND G. A. KENDRicK. 1988. Austral spring thymidine incorporated into DNA: Coincidentmicroalgae across the Weddell Sea ice edge: Spatial measurements of photosynthesis and cell divisionrelationships found along a northward transect of individual species of phytoplankton isolatedduring AMERIEZ 83. Deep-Sea Res. 35: 1-20. from natural populations. J. Phycol. 22:199-205.

FuRNEsN, M.-L. 1973. Photoplancton et produc- SCHNACK, S. B. 1983. On the feeding ofcopepods ontion primaire dans le secteur sud-occidental de la Thalassiosira partheneia from the Northwest Af-Mediterranee. Rev. Tray. Inst. Peches Marit. 37: rican upwelling area. Mar. Ecol. Prog. Ser. 14: 49-19-68. 53.

GOULD, R. W., JR., AND G. A. FRYxEL.. 1988. Phy- SMAYDA, T. J. 1970. The suspension and sinking oftoplankton species composition and abundance in phytoplankton in the sea. Oceanogr. Mar. Biol.a Gulf Stream warm core ring. 1. Changes over a Annu. Rev. 8: 353-414.five month period. J. Mar. Res. 46: 367-398. 1978. From phytoplankters to biomass, p.

220 Notes

273-279. In A. Sournia [ed.], Phytoplankton man- cell division in natural populations of marine di-ual. Monogr. Oceanogr. Methodol. 6. UNESCO. noflagellates from shipboard cultures. J. Exp. Mar.

SMErACEK, V. S. 1985. The role of sinking in diatom Biol. Ecol. 25: 239-247.life history cycles: Ecological, evolutionary, and YOUNGBLurH, M. J., AND G.-A. PAFFENH6FER. 1987.geological significance. Mar. Biol. 84: 239-251. Massive bloom of the colonial, gelatinous diatom

SMITH, R. C., K. S. BAKER, AND P. DusTAN. 1981. Thalassiosira subtilis in upwelled waters: Distri-Fluorometric techniques for the measurement of bution, abundance, productivity and fate. Eos 68:oceanic chlorophyll in the support of remote sens- 1724.ing. Scripps Inst. Oceanogr. Ref. SIO 81-17. 14 p.

UNESCO. 1988. Eutrophication in the Mediterra-nean Sea: Receiving capacity and monitoring of Submitted: 16 March 1989long-term effects. UNESCO Rep. Mar. Si. 49. Acceted: 1 July 1989195 p. Accepted: 11 July 1989

WEl.ER, C. S., A# S. W. CHISHOLM. 1976. Phased Revised: 28 September 1989