Aquacultural Engineering 2 (1983) 27-47

Large-scale Microalgae Production for Nursery Rearing of Marine Bivalves

Niels De Pauw*, Jan Verbovent and Christine Claus*

*Laboratory for Mariculture, State University of Ghent, J. Plateaustraat 22, B-9000 Ghent, Belgium tlnstitute for Marine Scientific Research (IZWO), Prinses Elisabettdaan 69, B-8401 Bredene, Belgium

A BS TRA CT The feasibility o f large-scale bloom induction o f nutritionally suited natural phytoplankton species, to feed a semi-industrial nursery o f edible shellfish, built on the Belgian coast, was tested. The outdoor microalgal production unit consisted o f four tanks o f 100 m 2 surface each (two o f l m depth and two o f O.5 m depth), equipped with different mixing devices. The cultures were run as chemostats in which seawater was enriched with commercial inorganic N, P, and Si fertilizers. Depending on the season, between 5-10 and 80% o f the culture volume could be harvested daily, with algal densities ranging from 50 000 to 500 000 cells per mL By manipulation o f operational parameters such as detention time, nutrient levels and nutrient ratios (N : Si : P), unsuited or less suited species o f algae (e.g. Chlorella and Phaeodactylum) could be replaced by more desirable species (e.g. Skeletonema, Nitzschia, Chaetoceros). Various biological and technological problems encountered during year round operation, including collapsing o f the culture, seawater enrichment, water quality, foufng and water treatment, are commented.

INTRODUCTION Nursery rearing of edible bivalve molluscs means the controlled culturing o f postlarvae (i.e. spat), from a few millimeters in length up to a few 27 Aquacultural Engineering 0144-8609/83/0002-0027/$03.00 - Applied Science Publishers Ltd, England, 1983. Printed in Great Britain

28

N. De Pauw, J. Verboven, C. Claus

centimeters (Bayes, 1979). The high mortality within this size class, when transferred directly from the hatchery to the natural environment, can be seriously reduced when the spat is reared under controlled conditions (COST, 1978; Persoone and Claus, 1980; Claus, 1981 ; Claus et al., 1981). A major bottle-neck in high-density nursery rearing is the need for large quantities of live unicellular algae as food for the molluscs, because no alternative inert feeds are presently available (Persoone and Claus, 1980; De Pauw, 1981). The well-established, but expensive indoor techniques used for monospecific algal production in hatcheries, cannot, for economic reasons, be scaled up to the large volumes needed for feeding nursery bivalves (De Pauw, 1981 ). As a result, nearly all commercial enterprises involved in nursery rearing of bivalve molluscs utilize natural phytoplankton present in the seawater as food. Culturing of the postlarvae is performed either in situ (in the sea), or in onshore constructions with flow through pumping of seawater (Claus, 1981 ; De Pauw, 1981 ). Though satisfactory in some cases, results are often inconsistent due to food shortage or the development of undesirable species. Therefore, the alternative for the nursery rearing of juvenile bivalves may be the large-scale production of microalgae relying on speciescontrolled induced bloomings of natural phytoplankton (De Pauw, 1981). However, as Goldman (1979) very properly observed, the control and manipulation of phytoplankton species in large-scale outdoor cultures, is perhaps the major unresolved problem in mass-culture technology. In order to contribute to the solution of this problem, and after experimenting several years with small-scale (1 m 2) outdoor culture units (De Pauw and De Leenheer, 1979; De Pauw et al., 1980), experience is presently being gained on the Belgian coast with large outdoor production units for the continuous culturing of natural phytoplankton to feed a semi-industrial nursery (cf. Claus et al., 1982). In addition to acquiring an insight in the biological and technological problems resulting from the scale-up of these cultures to the semiindustrial level, our major objective is to gain control over the species composition of these induced blooms of marine phytoplankton.

Mieroalgae for nursery marine bivalves

29

This paper aims at reviewing the results obtained so far with four o u t d o o r algae tanks of 100 m 2 each, after one full year of operation.

M A T E R I A L S AND METHODS

Plant description
A scheme of the algal production plant located at Bredene, Belgium (51 N), is presented in Fig. 1. The plant consists of one seawater stocking tank (32 m 3 capacity, 1 m depth) and four rectangular algae production tanks ( 1 0 0 0 m 2 surface, t w o tanks o f 1 m depth and t w o of 0.5 m depth), built on a concrete floor. The 1 m deep tanks are made o f w o o d e n panels supported by iron frameworks; the 0.5 m deep tanks are made of large hollow cement bricks, reinforced inside with iron bars fixed in the concrete foundation. All tanks are waterproofed with a non-toxic 0.5 mm thick plastic PVC liner, which is protected at the b o t t o m , by a 10 cm thick layer of fine sand. Three of the four production tanks are equipped with different mixing devices in order to evaluate the mixing yield of these systems: (1) an eight-bladed paddle-wheel, connected to a 1.5 kW electric m o t o r provided with reductor permitting rotations between 0 and 26 rpm; (2) 22 air-lift pumps o f 1 m length and 5 cm in diameter, fixed along the side walls o f the tank; and (3) 11 perforated (3 mm) air-bubbling tubes (3 cm in diameter) fixed at the b o t t o m . The air is provided by t w o 1-5 kW cycloblowers (0-1 bar (3-5 m a min-l)). The paddle-wheel m o t o r as well as the cycloblowers, are connected with time-switches, in order to regulate the mixing time. The fourth tank has no mixing device (0.5 m depth) and serves as a control.

Culture procedure
The culture procedure is represented in Fig. 2. Seawater is pumped up from the nearby Sluice Dock lagoon (82 ha) into a reservoir to allow pre-sedimentation of larger particles. By means of a constant head system (3 m high, 31-5 cm in diameter, 0-23 m 3 capacity), seawater is distributed with constant pressure from

,,-===:
............
,
m

/--/

~;~

s_s,,, /7 IL

:..
r~

c~. |I _ _ _ ~ -i ~ - ~ -

~ - ~ - ~ - ~ -~,-i~ -~,-- ~ ~ r T ~ : . - ~ .... . ~ , lit ~ : : : =

seawater.
algae drai, suspension

I
~~-~H

/ I

~ <~ ~ ~ ~ ................................. t * r t ~ ,

............

/I--m~]~--i,

i,

iAll

[,

*~#

~ sca]e

pump = i
m

e~

It_:
I+ ~

i l

lagoon

c~
A4

li I"

AI A3

A2

A m
CROSS SECTION

.... :ALGAE TANKS

lagoon

Fig. 1. Ground plan and cross-section of the algal production plant feeding a semi-industrial nursery for bivalve molluscs. R = seawater reservoir; N = indoor nursery; A --- algae tanks of 100 m 2 each; A1 = 100 m 3 tank with air-bubbling tubes; A2 = 100 m 3 tank with air-lift pumps; A3 = 50 m 3 tank with paddle-wheel and partimentation; A4 = 50 m 3 tank without mixing device; F = fertilizer reservoir; C.H. = constant head.

Microalgae for nursery marine bivalves

31

lagoon

CULTURE PROCEDURE [RAW ] ! SEAWATER ] ] [SEDIMENTATION 3[ [BAG FILTRATION It"

ENRICHMENT

]l

BATCH CULTIVATION

l[ CONTINUOUS
CULTIVATION BIVALVE NURSERY
Fig. 2.

~ _

continuous supply

11

continuous harvest to nursery

Culture procedure steps to induce blooms of natural phytoplankton populations.

the reservoir to the four algal production tanks. Inflow rates can be regulated with plastic valves. In order to set the flow rates rapidly and inexpensively, a simple harvest-measuring device was constructed. This device (Fig. 3) consists o f a plastic bottle (2 liter contenance) with a vertical slit (1 cm wide, 15 cm height) cut out on its side. The bottle can be connected to the seawater inflow pipe of each tank. Its working principle is based on the height of the water level inside the bottle, flowing through the slit, and varying with the inflow rate. Calibrated marks next to the slit, allow direct read-offs of the harvest percentage. An elongated inflow tube inside the bottle results in a flat (air-bubble free) and thus easy-to-read water level. At the opposite ends of the tanks, overflows are constructed at respective heights of 0-5 m or 1 m. Continuous water supply through the inlet creates equal overflow (i.e. harvest rates). This harvest can be drained by gravity to the bivalve nursery, or back to the Sluice Dock lagoon if unsuited or unneeded. Since microalgae need inorganic nutrients to grow, the usually nutrient-limited seawater has to be enriched to a certain level with

32

N. De Pauw, J. Verboven, C. Claus

la~

iNFLOW

OUTFLOW

Fig. 3. Harvest rate measuring device (range flow rate: 0-62.5 liter min-1). 1, inflow piping; 2, disconnectable joint; 3, calibrated scale (% harvest day-1 of the tank); 4, plastic bottle (2 liter contenance); 5, slit (1 cm wide, 15 cm high); 6, elongated inflow tube (submersed inflow opening prevents disturbance of water surface inside the bottle by air-bubbles); 7, water level to be read off (level changing with flow rate of incoming water).

macro-elements such as nitrogen, phosphorus and eventually silicates (the latter to induce diatom species). As N-, Si-, and P-source, the following commercially available fertilizers are used: crystalline a m m o n i u m sulphate (21% N); liquid technical Na20.2SIO2 (14.5% Si), k n o w n as soda silicate solution; and technical phosphoric acid 85% (26% P). The cost price of these fertilizers amounts, respectively, to about US$0-10, 0-18 and 1.6 per kilogram (US$1 ~- 44 Bfr). Of these nutrients, separate stock solutions are made up with tap water and stored in 1 m 3 barrels placed outdoors next to the algal culture tanks, and provided with a drain valve.

Mieroalgae for nursery marine bivalves

33

The continuous cultures are enriched daily with a fixed amount of these nutrient stocks: depending on the season between 2 and 8 liters of nutrient solution 100 m -2. The production of the algae as induced blooms of natural phytoplankton species, consists of two successive steps. During the first step (batch phase) an algal bloom is triggered in the culture tank filled with enriched, only coarsely filtered seawater (filtration through a polypropylene filter bag, mesh size 50/~m), containing a large assortment of algal species. Depending on the solar radiation, the batch phase usually takes between 5 and 10 days to develop a sufficiently thick algal culture. During the second step, the cultures are run as continuous cultures with a permanent inflow of 50#m filtered seawater, daily enriched, resulting in an overflow harvest drained to the nursery (cf. Claus et al., 1983).

Physical, chemical and biological analyses

Every day analyses of the algal cultures included registration of the water temperatures (submerged min.-max, thermometers), the pH and the algal culture density measured with a simple turbidity stick (a modification of the classical Secchi disk, readings in cm) or by measuring the optical density with a spectrophotometer at 678 nm in a 1 cm cuvette (Sorokin, in Stein, 1973). Twice a week, N, P, and Si levels in the cultures were determined and cell numbers of occurring algal species counted in a haematocytometer (Bfirker or Fuchs-Rosenthal counting chamber, depending on the algal concentration). Ammonium-nitrogen was measured with a Kjeltec 1003 Distilling Unit (Kjeldahl without destruction) and later on also with a simple reagent kit (Aquamerck 11117, Ammonium test, Merck). Nitritenitrogen was also determined with a kit (Aquamerck 8036 Nitrit, Merck). Total inorganic nitrogen was measured with the Kjeltec apparatus mentioned above. Nitrate-nitrogen was then calculated as the difference between total inorganic nitrogen and n i t r i t e - a n d ammonium-nitrogen. Orthophosphate was determined according to Standard Methods ( 1975) and silieium with another kit (Aquamerck 8022 Silicat, Merck).

34

N. De Pauw, J. Verboven, C. Claus

Although originally developed for freshwater analyses, these kits proved to be valuable for seawater analyses as well, according to our experience and calibration tests.

RESULTS

Nutrients, culture conditions and algal growth

Daily, the algal culture tanks fed with lagoon seawater, were enriched with 0-5-1.6 g N m -2, 0.6-1-6 g Si m -z and 0--0-1 g P m -z, resulting in an approximate N :Si :P ratio (by weight) of 10 : 10 : 1. Incoming seawater from the lagoon contained between 1-0 and 2.0 mg Si liter -1, and between 0.25 and 1.0 mg P liter -a, but was usually nitrogen limited (level undetectable). Depending on the algal growth and harvest rates, the nutrient levels in the algal cultures fluctuated between 0-3-5 mg N liter -a, 0-4.0 mg Si liter -1 and 0-0.75 mg P liter -1. Zero values should be interpreted as undetectable levels. The salinity of the seawater in the Sluice Dock lagoon, varied between 23%o during winter, and 31%o during summer. The pH o f the cultures during the day varied according to the season and the dilution rates, from 8.0 to 9-0 in winter, and from 8-5 to 10-5 in summer. The pH in the shallow tanks (0.5 m) was usually 0-5 to 1 unit higher than in the deeper tanks (1 m), due to higher algal concentrations. Mixing, even with air-lift pumps or air-bubbling, did not result in a decrease o f the pH due to introduction of CO2 from the air. Without mixing for more than 1 h, a vertical pH gradient was observed in the cultures (up to 0.3 units lower at the bottom). Mean water temperatures ranged from 2C in December and January up to 20C in July and August, with extremes of --1.5C and 25.5C. Temperatures in the mixed cultures differed by only 0.5C of those of the Sluice Dock lagoon. Those o f the two 0.5 m deep culture tanks differed by nearly IC more or less, in comparison with the 1 m deep tanks. When mixing was not carried out for a period of at least 1 h, a thermal gradient occurred with a temperature increase or decrease of 4-5C in the upper 0.5 m layer. The species composition of the induced natural phytoplankton blooms (without artificial but natural inoculation from the incoming seawater) was mainly dominated by the small chain-forming diatom

Microalgae for nursery marine bivalves

35

Skeletonema costatum (Gr6ville) Cleve, occurring almost year-round except in June. The harvest rates were adapted according to the time of the year, and the solar radiation (Fig. 4A). During winter, only 5-10% of the culture volume could be harvested per day (detention time 10 days), while up to 80% per day could be harvested during the summer months (detention time 1-3 days). Too low harvest rates usually resulted in collapsing of the culture followed by coagulation and sedimentation of the microalgae. Too high harvest rates on the contrary, resulted in culture wash out. The average and maximal harvest regimes that possibly could be maintained at our latitude (5 IN) are given in Fig. 4B. Depending on the dilution rate (harvest rate), the cell density for Skeletonema varied between 50 000 and 500 000 cells liter -1. Highest cell densities were always obtained in the 0.5 m deep systems in comparison with the deeper culture systems (2-7 times higher). The maximal production of Skeletonema, measured by the optimal cell density in the 1 m deep culture tanks, was estimated to be between 100 000 and 300 000 cells m1-1, corresponding to turbido stick readings of 30-50 cm. In terms of daily cell production of Skeletonema, average yields during winter were about 20-30 X 109 cells m -2 day -1, increasing towards the summer to about 400 X 109 cells m - 2 day -~ (Fig. 4C). These figures correspond to a calculated biomass production of 1.I-1-8 g and 22.3 g dry weight m -2 day -~, respectively (1012 cells equal 56 g dry weight). During June and August, peak values of 30 g dry weight m -2 day -1 were measured. The culture tanks were only emptied and cleaned if the microalgae failed to bloom although enough nutrients were available. Since this phenomenon seldom occurred, the culture tanks were cleaned only a few times a year (depending on the case, between two and eight months of operation). To what extent non-cleaning of the tanks during a certain period has affected the yields obtained, could not be evaluated. Although regularly observed, the grazing pressure of herbivorous organisms on the algal culture was also difficult to evaluate. Potential micropredators, such as rhizopodes (Pelomyxa sp.), colorless flagellates (Monas sp., Desmarella sp., Stephanocoeca sp.) naked dinoflagellates (Gymnodinium sp.), ciliates (Cyclidium sp.) and rotifers never occurred in ihigh densities in the tanks while in operation.

36
zm_. A.
11 SOLAR

N. De Pauw, J. Verboven, C. Claus

RAmATION , O I T E N D BELGIUM LAT. 51" N ~

1~I,_

.
. . . .

mean v i l u e 1960-80 / / t ~ /
. c , . _

~ /

\
% (dayS) 1- -III ~hlrvest rite - ~\(%hlrvost -III

'lie

/ ~ J / /p.d

"

~ ,
N E _= I_ R. HARVEET

"- ~- - .

-11

REGIME

r/

5I|_

l,
/li~ /

I
I

2- -51 41 I 4 11 S -Z| IJ -10

4ill ", ALGAE CELL PRODUCTION

111_ III] "% , [ - - - ~ ,' X Xl XII 11180

//11 I L;/;I
I I

I I,I
I

;1

I L!.IL.
' . ~ - - - ' ~ --- ~ -r4 ~ ... . . .- . - ' ' I

." " e,. ,ie ~ e i n I - . - . . , i I, - . vI vii ,11)81

<IlL!

iI |n

--~---~ i
I~

II

IV

vnl

IX

xI

Fig. 4. Solar radiation (A), harvest rate (B), and algal yield (C) obtained with induced blooms of natural phytoplankton (mainly Skeletonema) in enriched seawater. 100m 2 culture tanks located at 51N, during one year of operation (1980-81). Benthic macropredators included polychaetes (Nereis sp., Nephtys sp., Polydora sp., Lineus sp.) and bivalves (Cardium edule , Mya arenaria, Mytilus edulis). Over 60 kg of C. edule between 1 and 2.5 cm, were recovered from the two 1 m deep culture tanks after a non-cleaning

Mieroalgae for nursery marine bivalves

37

period of c. eight months. Cirripeds (Balanus sp.) fixed to submerged parts of the tanks (walls, air-lift pumps, etc.) occurred in great numbers after 10 months of operation. Free-swimming macropredators included copepeods (Eurythemora sp.) amphipods (Gammarus sp.), some crabs (Carcinus sp.), a few fishes (Anguilla sp., Gobius sp.) and metazoan larvae (meroplankton, i.e. nauplii of Balanus sp.). Most of these organisms probably entered the system at the larval stage, because the incoming seawater was filtered over 50 tam, although some might have been introduced via the occasionally overflowing filter bags which had to be cleaned up to twice a day during the summer months (highest harvest rates). From June until October macroalgae (Ulva lactuca, Enteromorpha sp., Ulotrichales), developed abundantly on the tank walls or floated at the surface. This macroalgal development was strongest in the airmixed culture tanks, where amounts up to 40 kg per week and per 100 m 2 (wet weight) of Enteromorpha (mainly floating at the surface) had to be removed. A similar development of macroalgae has been observed in the eutrophicated Sluice Dock lagoon from May till August. In June, the standing crop of Ulva was estimated to be at least 80 tons dry weight (0.1 kg dry weight per m-2). During this period, the pH of the incoming seawater increased up to 10, giving rise to difficulties in the bivalve nursery. From mid-August on, Ulva biomass started to decay (whitish color of lagoon water) with a noted increase in N and Si levels (N rose from 0 to 2.2 mg liter-a, Si from 0.8 to 2.0 mg liter-a and P remained constant at about 1.1 mg liter-a). Within a week this increase was followed by a natural bloom of Skeletonema which, however, collapsed after eight days probably due to nutrient exhaustion.

Influence of operational parameters on the species composition

During our study, large-scale experiments were carried out in which the operational parameters were deliberately altered, in order to evaluate the influence of these parameters on the species composition of the induced algal cultures. In practice the effects observed were always compared to those parallel cultures in which the parameters remained unaltered. The

38

N. De Pauw, J. Verboven, C. Claus

factors studied included temperature, nutrients (N, Si, P levels and ratios), detention time and mixing. The experiments resulted in induced blooms of the following algal species: Skeletoma costatum (Gr6ville) Cleve, Nitzschia longissima var. closterium W. Smith, Phaeodactylum tricornutum Bohlin, Chaetoceros sp., Chaetoceros simplex var. calcitrans Paulsen, Thalassiosira nordens]iOMii Cleve, Chlorella cf. vulgaris Beijerinck and Chlorella cf. saccharophila (Kriager) Migula. The number of induced Thalassiosira blooms was too small to draw any significant conclusions and therefore is not further considered. Experiments with Nitzschia were carried out between April and September, and with Phaeodactylum between May and November. The dominance of only biradiate types of Phaeodactylum from September till November was remarkable. Chaetoceros calcitrans occurred from July till November. However, the mass development of this species was restricted to a culture tank destined for experiments with salt nutrientrich well water (12.1 mg N liter -1, 10.4 mg Si liter -1, 1-3 mg P liter-1, 23%o) inoculated with lagoon seawater. The genus Chlorella occurred year-round, with a preference of C. cf. saccharophila to develop during the winter months. The results obtained so far from these manipulation experiments are summarized in Table 1. Table 2 gives two examples of deliberately induced changes in species composition by manipulation of operational parameters. When interpreting these data, it should, however, be kept in mind that, though certain conclusions can already be drawn, they only apply to a restricted number of environmental combinations of parameters tested, and are as such insufficient to perform a comprehensive analysis. Indeed, for most parameters, the ranges within which the different species may potentially thrive have probably been only partially studied. Moreover, several other important factors such as pH and availability of carbon, were not considered. DISCUSSION

Species control
From our experiments, we may conclude that nutrients, detention time and temperature played a major role in determining the species composition of induced blooms of natural phytoplankton.

TABLE 1 Operational Characteristics During B l o o m i n g of Several Natural P h y t o p l a n k t o n Species (Bredene, Belgium, 5 1 N) Average Average % Harvest temppH (per day] erature ( C] NH+-N 0-5 2-3 0.7 0-5 0-1 1-6 3-3 0-1 2-1 1-3 0-1 0-03 8-3 1-8 4-0 0.8 2-1 ~ 0-2 6.8 2-5 0-3 12-3 0.5 0-3 0.1 2-4 0-8 1-2 6.0 12.3 8.8 14-7 17.1 Si P 3-2 4-4 11-2 11-9 19-4 8.5 19.9 16-0 8-5 17-9 16-0 19-2 21-4 20-1 24.0 5.0 8.5 8.6 9-1 0-15 5-10 8.5 50 8.8 50 9.4 25 9-7 80 9.7 75 9.5 50 8.7 5-10 10.1 25--45 10-3 20-50 8-3 20--40 8-9 75 8-8 25 8-9 0-25 8.6 0-10 8.9 5-10 Continuous (3) 15 minh -~ (2) 5 m i n h - ' (t) Continuous (2) Continuous (1) 5 min h -~ (1) No mixing (4) 30 minh -~ (3) 5 m i n h -~ (3) No mixing (2) Continuous (1) Continuous (2) No mixing (3) Continuous (2) No mixing (3) No mixing (4) 5 min h-' (3) 3-9 (1.9/4-9) 2-8 (1-0/3-5) 0-9 (0-3/1-5) 0.8 (0-3/1-0) 0.2 (0-0/0.2) 2-7 (0.0/5.0) 1-3 (0-3/2-5) 0-1 (0.0/0.5) 3-7 (1.0/10.0) 0-4 (0-3/1.0) 0.1 (0.0/0.5) 0-03 (0.0/1-0) 1-7 (0-7/2.6) 0-7 (0-3/1-0) 1-2 (0.4/1.9) 3-0 (0-0/4.7) 3-7 (I-0/10.0) 7-5 (6-9/9.4) 1-2 (0.3/2-5) 1-3 (0-4/2-9) 1-5 (1-5/1-5) 1.4 (1.0/2-3) 1.7 (1-3/10.3) 0.4 (0-3/0-6) 1-3 (l-2/1-4) 1.8 (1-2/3-6) 0.3 (0-2/0-4) 1.0 (0-6/1-1) 1.2 (0.5/2.0) 0-2 (0-1/0-3) 0-4 (0.2/0-5) 0-3 (0-1/0.7) 4-0 (0-5110-1) 1.8 (1-2/3.6) 0.2 (0-0/0.3) 0-2 (0-1/0-3) 0.1 (0-01/0-05) (-/) 1-1 (0-7/1.l) 0.4 (0.0/1-1) 0-5 (0-1/0.9) 0-4 (0-4/0.4) 0.3 (0-0/1-2) 0-8 (0-8/0-8) 0-4 (0-4/0-4) 0-5 (0-3/0.8) 0-7 (0-4/0-9) 0.9 (0-7/1-3) 1-0 (0-9/1-1) 0-5 (0-1/1-3) 0-3 (0-0/1.2) Average nutrient levels, mg liter -~ (min./max.J Average Average N: Si N: P Mixing duration (culture tank number]

Species

Culture period

28/11/8012/1/81 9/2/8127/3/81 23/3/812/5/81 30[3[8118/5/81 17/7/817/8/81 16/10/8130/11/81 Phaeodaetylum 9/5/81tricornutum 22/6181 26/5/8125/6/81 16/1018130/11/81 Nitzschia longissirna 14/5/81vat. elosterium 26/5/81 26/5/8125/6/81 11/6/816/7/81 Chaetoceros sp. 23/7[812/8/81 24/71812/8/81 5/8/8114/8/81 Chlorella sp. 20/10/8012/2/81 16/1018130/11/81

Skeletonema costatum

40

N. De Pauw, J. Verboven, C. Claus

TABLE 2

Two Examples of Simultaneously Induced Blooms with Different Species Composition as a Result of Manipulation of Nutrient Ratios and Detention Time

Culture Nutrient levels tank (mg liter-1) number NI-I~4-N Si P

Aver- Aver- % Harvest Dominant species Average age age (per day) cell coneN :Si N :P entration (per ml)

1 2

Example 1:16 October-30 November 1981; 5-12C 2-7 1-7 0.4 1-8 7-0 20--40 Skeletonema 99 000 Chlorella 3 900 000 3.7 1.8 0-3 2-0 12-0 5-10 Phaeodactyluml't b~t,,a 511000 5 000 Example 2:26 May-25 June 1981; 12-20C Nitzschia
0.1 1.0 0.5 0-1 0.2 75

Phaeodactylum t't' 1'b' Nitzschia

147 000 11 000 1 000

0.1

0.8

0.4

0-1

0.3

25--45 Phaeodactylum ['t' 1'b' ~

81 000 344 75 000 000

a 'b' and 't' are biradiate and triradiate forms, respectively, of Phaeodactylum.

L o w nitrogen levels tended to favor the development of Skeletonema. This reflects the ability of this species to utilize ammonium-nitrogen at limiting levels (De Manche et al., 1979; Harrison and Davis, 1979; Mickelson et al., 1979; Rodhouse et al., 1981 ). In contrast, Skeletonema could not cope with low Si levels (cf. Paasche, 1973; Harrison and Davis, 1979), an inability shared with other diatoms such as Thalassiosira pseudonana (D'Elia et al., 1979). The high demand for silicium probably explains w h y Goldman and Mann (1980) found a predominance of Skeletonema during winter time in sewage-enriched algal cultures (high Si content of sewage), compared to cultures only enriched with nitrogen dominated almost exclusively by Phaeodactylum. The preference o f Skeletonema for a low N : Si ratio appeared also in competition experiments with Phaeodactylum and Skeletonema (De Pauw and Naessens, 1982).

Microalgae for nursery marine bivalves

41

Phaeodactylum on the other hand, though lacking a siliceous envelope (Bourreley and Dragesco, 1955) and having as such no Si requirement for cell division (Lewin et al., 1958), is nevertheless capable of silicate uptake (D'Elia et al., 1979) but its growth is not affected in Si-limiting conditions. The preference of Phaeodactylum for high N : Si ratios was also demonstrated by De Pauw and Naessens (1982) in indoor competition experiments. The same experiments also showed the preference of Skeletonerna for high N :P ratios (> 10) in contrast with Phaeodactylurn preferring low N:P ratios (<5). Nitzschia longissima var. closterium which is a weakly silicified diatom species, resembles Phaeodactylum in its ability to grow in Silimited cultures (Harrison and Davis, 1979). This could explain the dominance of Nitzchia in cultures with low as well as high N : Si ratios. Chlorella, being a chlorococcalean alga and thus not requiring silicium for cell division, shows a preference for higher N and P levels in comparison with diatoms (Dunstan and Tenore, 1974; De Pauw and De Leenheer, 1979, 1980; De Pauw et al., 1980). The influence of the dilution rate on the species composition is linked to the different growth rates of the species concerned, while the growth rate itself is strongly influenced by the light intensity (incident radiation) (Nelson et al., 1979). The dilution rate, related to the algal density (self-shading), also determines the degree of light-limitation within the culture. As a result, high dilution or harvest rates can only be coped with by fast growing (usually small-sized) microalgal species. In competition experiments, Skeletonema could easily outcompete Thalassiosira at dilution rates of 0.03 h -1 (= 0.72 day-l), but was outcompeted itself at the same dilution rate by Chaetoceros (Mickelson et al., 1979). Analogous conclusions could be drawn from experiments by Harrison and Davis (1979). These authors, moreover, stressed the importance of N and Si limitation coupled to different dilution rates in species composition. Another interesting finding by Harrison and Davis (1979) is that decreasing light intensity at the lowest dilution rate selected for an assemblage similar to that observed at the high dilution rate and high light intensity. This explains why during winter when light intensities are low, the dilution rates have to be decreased in order to cope with the reduced growth rate of the species wanted, while in summer the same species can be obtained at much higher dilution rates but which correspond with higher light intensities (cf. Table 1).

42

N. De Pauw, J. Verboven, C. Claus

In this regard, fast growing Skeletonema (/am~ = 0"04-0-09 h-l; cf. Paasche, 1973; Harrison and Davis, 1979; Mickelson et al., 1979) could stand relatively high dilution rates year-round even in winter time when light intensities are low, while Nitzschia (/~max = 0"02-0"04 h -1; Harrison and Davis, 1979) could be cultured at high dilution rates only during spring and summer, probably being light- and temperature-limited during winter. Phaeodactylum (/~max = 0"05 h-i; D'Elia et al., 1979) was observed in our cultures at relatively low dilution rates, even when silicium was present in excess. Phaeodactylum is probably capable of maintaining cell growth even at low light intensity (Nelson et al., 1979), in contrast to Skeletonema which needs more light. This would also explain why during the warmer season (above 10C), Phaeodactylum is able to outcompete Skeletonema at relatively low dilution rates (10-25% harvest day -1) as observed by Goldman and Mann (1980). This continues to be in contrast with our findings and those of Riva and Lelong (1981) and Rodhouse et al. (1981) that Skeletonema can also dominate the plankton assemblage at temperatures exceeding 10C or even 20C but at dilution rates of 50% harvest day -1 and more. These observations confirm the experimental work of Dunstan and Tenore (1974). At high dilution rates, temperature would thus become a determining factor of secondary importance for Skeletonema. The lowest dilution rates seemed to favor the development of Chlorella in our cultures. This confirms the experimental findings of Dunstan and Tenore (1974). It should, however, be emphasized that in our experiments the dilution rate and the nutrient levels were often linked to one another in a way that at low dilution rates nitrogen and phosphorus supplied daily to the cultures tended to accumulate, while silicium (partially present in the incoming seawater) usually tended to decrease as a consequence of the previous blooming of Skeletonema, which died off after silicium exhaustion. As a result, possible beneficial effects of low dilution rates or favorable N : Si : P ratios for Chlorella are difficult to assess. Temperature is also considered by many authors to be an important ecological factor influencing the outcome of competition among phytoplankton species in enriched cultures (Dunstan and Tenore, 1974; Goldman and Ryther, 1976; De Pauw et al., 1980; Goldman and Mann, 1980; De Pauw and De Leenheer, 1980). Skeletonema, Phaeodactylum

Mieroalgae for nursery marine bivalves

43

as well as Chlorella all showed a broad range of temperatures within which dominance of these species could be observed. In outdoor cultures light and temperature conditions are usually coupled to one another, which again makes it difficult to distinguish between light- and temperature-limitation effects on the species composition. With regard to mixing intensity, no significant influence on the species composition was observed (confirmed by the experiences of Mickelson et al., 1979). Indirectly, the absence of mixing results'in a non-homogeneous dispersion of the supplied nutrients (especially silicates) throughout the algae cultures. In large units this could cause a 'patched' development of different algal species, related to 'patched' different nutrient ratios in the culturing tanks. This would explain why in the unmixed algal tank Chlorella was co-dominating with Skeletonema during the winter period, while in the mixed parallel cultures, only Skeletonema developed. Last but not least, attention must be drawn to the fact that when interpreting competition phenomena, one should also consider possible allelopathic effects among species, as demonstrated by Elbr~ichter (1976). Between Phaeodactylum and Thalassiosira no such effects have, however, been detected (D'Elia et al., 1979). In conclusion, we can state that, though it was demonstrated that manipulation of the species composition of induced blooms of natural phytoplankton is feasible in practice even in large-scale systems, further experimentation is needed to elucidate precisely under which combinations of factors the desired species can be triggered to bloom. Technological problems Most technological problems encountered were related to tank construction, operation and maintenance of the installation. The originally designed 1 m deep tanks consisted of cemented brick walls, which collapsed due to water pressure. They were redesigned and rebuilt with wooden panels supported with iron frameworks, giving full satisfaction, except for corrosion of the iron supports which remained a problem. The 0-5 m high tanks built with bricks, on the contrary, proved to be resistant.

44

N. De Pauw, J. Verboven, C. Claus

Serious leakage causing loss of algal harvest and unstable harvest rates occurred due to the perforation of the plastic liners in the culturing tanks. The vulnerability of these liners (0-5 mm) made effective cleaning an almost impossible task. Thicker liners of 1 mm and more could probably solve this problem, but add substantially to the cost of the installation. The highest harvest volumes obtained during summer could not be completely directed to the bivalve nursery as a result of overflow drains with too small dimensions. A similar under-dimensioning of discharge drains resulted in too lengthy waiting periods for emptying the culture tanks (up to 9 h). An insufficient or discontinuous slope on the tank bottom, as well as discharge valves being placed too high, also made it difficult to empty the tanks completely. In order to reduce the risk of corrosion by seawater, pumps and mixing machinery were regularly serviced and greased. To avoid frost damage, pumps and pipings were thoroughly insulated with polystyrene or rockwool. However, when serious freezing (--4C and more) was forecast to last for a certain period, all tanks and accessory pipings were completely emptied. The use of inground algal basins would probably reduce the problem of ice formation. Addition of alkaline sodium metasilicate solutions to enrich seawater with silicates usually resulted in irreversible formation of colloids, causing Si to disperse insufficiently throughout the culture. The experience was, however, that when silicium stock solutions were acidified with hydrochloric acid (up to pH values of 1-0-2-0) (cf. also Riva and Lelong, 1981) no colloid formation was observed when brought into contact with seawater. More elaborate results on this will be published in the future.

Cost price of algal production

The cost price of algal production was calculated, including expenses for fertilizers, pumping and mixing, and labor (cf. De Pauw, 1981). Investments, maintenance and capital return were not considered. Provisional algal production costs throughout the year ranged, according to algal production (mainly S k e l e t o n e m a ) , from US$4 per kg dry weight during summer up to US$23 per kg during winter (smaller yields).

Microalgae for nursery marine bivalves

45

This is relatively cheap when compared to the cost price of monospecific algal cultures produced indoor with artificial light for which figures are greater than US$120 to 200 per kg dry weight for lsochrysis galbana (Walne, 1976), and US$162 per kg for Thalassiosirapseudonana (De Pauw, 1981). However, when interpreting these figures according to cost-benefit analysis the expenditure for algal production has to be weighed against the relatively high commercial value of the end product, in our case the bivalve spat produced. It should also be taken into consideration that only in the case of suited algal species will satisfactory growth of the bivalves be obtained; presently this is not always the case with induced blooms of natural phytoplankton. Future research is therefore focused both on the reduction of the cost price of algal production and on increasing the reliability of inexpensive bloom induction of suited species of natural phytoplankton through species control by manipulation of operational parameters.

ACKNOWLEDGEMENTS We acknowledge the Belgian Ministry of Scientific Programmation, and the Ministry of the Region of Flanders, for sponsoring this research. Our sincere thanks are due to Professor Dr G. Persoone, Director of the Laboratory for Mariculture, for his stimulating advice and support demonstrated during this project, and to Dr Jr. E. Jaspers, Director of the Institute for Marine Scientific Research for reading through the manuscript. For technical assistance, the authors are greatly indebted to Lic. H. De Buck, M. Dochy (Analyst), Ing. G. Jonckheere, and ir. T. Moens de Haze. The drawings were made by F. Persyn to whom we extend our sincere thanks. REFERENCES Bayes, J. C. (1979). How to rear oysters. Proc. lOth Annual Shell1~sh Conference, Shellfish Assoc. of Great Britain, London, pp. 7-13. BourreUy, P. & Dragesco, J. (1955). Contribution /~ la connaissance d'une algue rarissime Phaeodactylum tricornutum Bohlin. Bull. de Microscopie Appl., 2 (5), 41-6.

46

N. De Pauw, J. Verboven, C. Claus

Claus, C. (1981). Trends in nursery rearing of bivalve molluscs. In: Nursery Culturing of Bivalve Molluscs, eds C. Claus, N. De Pauw and E. Jaspers, EMS Special Publication No. 7, European Mariculture Society, Bredene, Belgium, pp. 1-33. Claus, C., De Pauw, N. & Jaspers, E. (eds) (!981). Nursery Culturing of Bivalve Molluscs. EMS Special Publication No. 7, European Mariculture Society, Bredene, Belgium. Claus, C., Maeckelberghe, H. & De Pauw, N. (1983). Onshore nursery rearing of bivalve molluscs in Belgium. Aquacultural Engineering, this issue, p. 13. COST (1978). Proposal for a coordination of research activities in the field of mariculture. Report from the Secretariat to the Cost Senior Officials Committee, COST/58/78, European Cooperation in the Field of Scientific and Technical Research. De Manche, J. M., Curl, H. C., Jr., Lundy, D. W. & Donaghay, P. L. (1979). The rapid response of the marine diatom Skeletonema costatum to changes in external and internal nutrient concentration. Mar. Biol., 53,323-33. De Pauw, N. (1981). Use and production of microalgae as food for nursery bivalves. In: Nursery Culturing of Bivalve Molluscs, eds C. Claus, N. De Pauw and E. Jaspers, EMS Special Publication No. 7, European Mariculture Society, Bredene, Belgium, pp. 35-69. De Pauw, N. & De Leenheer, L., Jr. (1979). Mass culturing of marine and freshwater algae on aerated swine manure. In: Cultivation o f f i s h Fry and its Live Food, eds E. Styczynska-~urewicz, T. Backiel, E. Jaspers and G. Persoone, EMS Special Publication No. 4, European Mariculture Society, Bredene, Belgium, pp. 44173. De Pauw, N. & De Leenheer, L., Jr. (1980). Outdoor mass production of marine microalgae for nursery culturing of bivalve molluscs. Paper presented at the 3rd Intern. Conf. on Production and Use of Microalgae, Trujillo, Peru, 26-31 October, 1980. De Pauw, N., Verlet, H. & De Leenheer, L., Jr. (1980). Heated and unheated outdoor cultures of marine algae with animal manure. In: Algae Biomass, eds G. Shelef and C. J. Soeder, Elsevier/North-Holland Biomedical Press, Amsterdam, pp. 315-41. De Pauw, N. & Naessens, E. (1982). Nutrient induced among the marine diatoms Phaeodactylum tricornutum and Skeletonema costatum. (In preparation.) Dunstan, W. M. & Tenore, K. R. (1974). Control of species composition in enriched mass cultures of natural phytoplankton populations. J. appl. Ecol., 11,529-36. Elbr~ichter, M. (1976). Population dynamic studies on phytoplankton cultures. Mar. Biol., 35,201-9. D'Elia, C. F., GuiUard, R. R. L. & Nelson, D. M. (1979). Growth and competition of the marine diatoms Phaeodactylum tricornutum and Thalassiosira pseudonana. I. Nutrient effects. Mar. Biol., 50, 305-12.

Microalgae for nursery marine bivalves

47

Goldman, J. C. (1979). Outdoor algal mass cultures. I. Applications. Water Res., 13, 1-19. Goldman, J. C. & Ryther, J. H. (1976). Temperature-influenced species competition in mass cultures of marine phytoplankton. Biotechnol. Bioengin., 18, 112544. Goldman, J. C. & Mann, R. (1980). Temperature influenced speciation and chemical composition of marine phytoplankton in outdoor mass cultures. J. exp. mar. Biol. Ecol., 46, 29-39. Harrison, P. J. & Davis, C. O. (1979). The use of outdoor phytoplankton continuous cultures to analyse factors influencing species selection. J. exp. mar. Biol. Ecol., 41,9-23. Lewin, J. C., Lewin, R. A. & Philpott, D. E. (1958). Observations onPhaeodactylum tricornutum. J. gen. Microbiol., 18,418-26. Mickelson, M. J., Maske, H. & Dugdale, R. D. (1979). Nutrient-determined dominance in multispecies chemostat cultures of diatoms. Lirnnol. Oeeanogr., 24 (2), 298-315. Nelson, D. M., D'Elia, C. F. & Guillard, R. R. L. (1979). Growth and competition of the marine diatoms Phaedaetylum tricornutum and Thalassiosira pseudonana. II. Light limitation. Mar. Biol., 50, 313-18. Paasche, E. (1973). Silicon and the ecology of marine plankton diatoms. If. Silicateuptake kinetics in five diatom species. Marine Biology, 19,262-9. Persoone, G. & Claus, C. (1980). Mass culture of algae: a bottle-neck in the nursery culturing of molluscs. In: Algae Biomass, eds G. Shelef and C. J. Soeder, Elsevier/ North-Holland Biomedical Press, Amsterdam, pp. 265-85. Riva, A. & Lelong, P. (1981). Growth of juvenile bivalve molluscs associated with continuous cultures of natural marine phytoplankton (Western Mediterranean, France). In: Nursery Culturing o f Bivalve Molluscs, eds C. Claus, N. De Pauw and E. Jaspers, EMS Special Publication No. 7, European Mariculture Society, Bredene, Belgium, pp. 253-68. Rodhouse, P. G., Ottway, B. & Burnell, G. M. (1981). Bivalve production and food chain efficiency in an experimental nursery system. J. mar. biol. Ass. UK, 61, 243-56. Standard Methods (1975). For the examination of water, sewage and industrial wastes. 14th Edn, Amer. Publ. Health. Assoc., Washington DC. Stein, J. R. (Ed.) (1973). Handbook o f Physological Methods - Culture Methods and Growth Measurements, Cambridge University Press, London. Walne, P. R. (1976). Factors affecting the relation between feeding and growth in bivalves. In: Harvesting Polluted Waters, ed. O. Devik, Plenum Press, New York, pp. 169-76.