An experimental simulation of changes in diatom and flagellate blooms

An experimental simulation of changes in diatom and flagellate blooms

J. up. tnw. Biol. Ecol., 1978, Vol. 32, pp. 285-294 0 Elsevier!North-Holland Biomedical Press AN EXPERIMENTAL DIATOM T. R. It7stitutc SIMULATION O...

797KB Sizes 13 Downloads 163 Views

J. up. tnw. Biol. Ecol., 1978, Vol. 32, pp. 285-294 0 Elsevier!North-Holland Biomedical Press

AN EXPERIMENTAL DIATOM

T. R. It7stitutc

SIMULATION

OF CHANGES

AND FLAGELLATE

PARSONS,

P. J.

HARRISON

IN

BLOOMS

and R.

WATERS

of’Oceanogrrrpl~y, lJniwr.~it~~of’Briiish Cohouhia. Vuncouvrr, B.C., Canada

Abstract: The sequence of events from a diatom to a flagellate bloom water enclosure. The two factors used to control the sequence were indicate that diatom growth can be manipulated to occur under specific nutrient concentrations. Once diatom growth has begun it appears to flagellates. The importance of this experiment to the study of food chain

were simulated in a large sealight and nutrients. The results conditions of light intensity and be more rapid than that of the ecology in the sea is discussed.

INTRODUCTION

Differences in the ability of phytoplankton species to reach their maximum growth rates under various conditions of light and nutrient limitation have been suggested by Dugdale (1967) and Eppley, Rogers & McCarthy (1969) as important factors in causing species succession in phytoplankton blooms. In addition to the observed succession of species within one environment, it has also been found that certain oceanic areas tend to be dominated by different sized organisms and this may also be attributed to oceanographic differences in light and nutrient regimes as well as to differences in water mass structure (Semina, 1972; Parsons & Takahashi, 1973). From this it is apparent that observations leading to conclusions on species succession have been derived mostly from laboratory studies of phytoplankton physiology while those on size distributions in the sea are by definition based on ecological data. There is at present no agreement on the essential mechanisms governing phytoplankton cell size and species succession, and discussions of this subject contain various ideas, none of which are mutually exclusive (see r.g., Hecky & Kilham, 1974; Parsons & Takahashi, 1974; Laws, 1975 ; Malone, 1975; Semina, Tarkhova & Truong Ngoc An, 1976). In order further to investigate this problem it has been possible to use a large controlled ecosystem enclosure (CEE) in which species succession is known to occur (e.g., Takahashi et al., 1975). but which can also be manipulated with respect to some of the environmental factors such as light, nutrients, and mixing. In such an experimental facility a number of analytical techniques have been used in order to follow changes in populations and environmental parameters. Over a period of 60 days the results show substantial changes in a phytoplankton population resulting from the manipulations of the external environment. 285

286

T. R. PARSONS,P. J. HARRISONAND R. WATERS METHODS

The controlled ecosystem enclosure (CEE) used in this experiment has been described by Menzel & Case (1977). It consists of a plastic contained water column, 9.5 m in diameter and 23 m deep containing GZ1300 m3 of sea water. The unit was floated in the sea and internal mixing of the surface waters was provided by an automatic bubbling unit which injected ~0.02 rn’ of air into the water column at IO-min intervals. The device was designed by Dr P. Thibault and the location of the outlet at 5 m gave an isothermal mixed layer of about 4 m with a temperature gradient between 4 and 8 m of about 3 C. The temperature in the upper mixed layer increased from 11’ C to more than 15 ‘C during the experiment. Salinity remained constant at about 29.5 ‘& throughout the water column, following initial destruction of the ‘halocline by mixing on the first day of the experiment. Upwelling of water from 22 m to the surface was provided by a 3 h.p. Jabsco pump and the quantity of water upwelled over different time periods varied from none at the beginning of the experiment, up to 15 m3/day as indicated in Fig 2B. Nutrient enrichment of the water column was made by adding KH,PO,, KNO,, and Na,Si0,.9H,O in sufficient quantities to raise the level of nitrate from zero after five days at the beginning of the experiment to more than 2 pg-at. NO,-N/l after 30 days. Nutrients were added in an atomic ratio of N : P : Si of 10 : 1 : 10. The amount of nutrients in the water column O-4 m was monitored and additional nutrients were added as required to maintain the desired concentrations: thus, for the first 30 days of the experiment, following nutrient exhaustion on Day 5, the amount of upwelling and nutrient addition was kept low (Fig. 2B) in order to produce a nitrate concentration of no.5 pg-at. NO,-N/l. On Day 30, both upwelling and nutrient addition were increased and the resultant nitrate concentration was brought to > 3 pg-at. NO,-N/l by Day 35. On Day 38 of the experiment a light shield was installed over the top of the CEE. The light shield consisted of opaque polyethylene strips z 2 m wide, the effect of which was to reduce the average radiation in the water column (&4 m) to zone-fifth of its former value (Fig. 2A). Nutrients and particulate material were determined by methods given by Strickland & Parsons (1971). Phytoplankton counts were made using an inverted biological microscope and size spectra of particulate material were obtained using a Model TAII Coulter Counter. Total radiation was determined with a 2~ pyroheliometer and extinction of light was measured within the CEE using a quantum scalar irradiance meter (Booth, 1976). Primary productivity was measured by the 14Ctechnique. For the purpose of obtaining data on the primary productivity of different size fractions, the 14C uptake experiments were performed on the total phytoplankton taken from 2 m in the CEE and also on a smaller size fraction obtained after removing larger particles with a nitex net, 35 pm diameter mesh. In practice, this net was found to remove most of the particulate material >20 pm diameter as measured on a Coulter Counter. Light-productivity curves were deter-

SIMULATIONOFCHANGESINBLOOMS

287

mined for the total and < 35 pm size fractions using screens having relative transmittances of 52, 34, 13 and 77:. Incubations were carried out for 4 h over the approximate period of 10.00 to 14.00 h, and at ambient temperature, 0.5 m below the sea surface.

RESULTS AND DISCUSSION

The sequence of events in the phytoplankton population is shown by the concentration of chlorophyll a, the relative number of small and large particles, and the numbers of diatoms and flagellates (Fig. 1). It is apparent that three blooms occurred during the 60-day experiment (Fig. 1C). These were at the beginning of the experiment (chl CImaximum z 10 mg/mj), after about 32 days (chl a maximum x3 mg/m3), and after 60 days (chl a maximum x 5 mg/m3). The initial phytoplankton bloom was due to diatoms captured at the beginning of the experiment and their numbers declined through to Day 30 (Fig. 1B) as the second maximum in chlorophyll developed. The second bloom (xDay 32) was in phase with a flagellate bloom (Fig. lB), while the final bloom was in phase with a large increase in the diatom population, followed by a smaller increase in flagellates towards the end of the experiment. By using two size categories from the Coulter Counter data (viz., mean particle diameters of 6 pm and 32 pm) it is possible to give (Fig. 1A) further information regarding the 3 maxima in the concentration of chlorophyll. The small particle sequence generally follows the count of flagellates but is out of phase by about 5 days with respect to the first flagellate maximum. Similarly, the large particle sequence shows a decrease at the beginning and an increase at the end of the experiment in phase with the diatom counts: however, a maximum in large particles at =Day 34 corresponds to the flagellate maximum (Fig. lB), rather than a diatom maximum count. These differences in data obtained with the Coulter Counter and from direct cell counts are important because both methods are being used for ecological and physiological studies. From further examination of samples it was apparent that the Coulter Counter data showed a maximum in large particles around Day 34 because of the presence of large dinoflagellates, particularly the red tide organism of Gonyaulux catenella (Whed. et Kof.), as well as a number of colourless dinoflagellates. Thus the content of total flagellates (Fig. 1B) reflects the sum of small flagellates and large flagellates which in total become most numerous at about Day 32. In the Coulter Counter data, cells with a very small diameter ( x 6.34 pm) reach a maximum earlier than the large cells ( z 32 pm diameter) and this accounts for the 5-day shift in maximum numbers of cells between the microscopic counts of total flagellates (Day 32, Fig. 1B) and the separate maxima of small particles (Day 25) and large particles (Day 34); both of the latter may be attributed to closely associated flagellate blooms. From these results it is important to note that differences in observations made with a particle size analyzer

T.R.PARSONS, P.J.HARRlSON ANDR.WATERS

288

and with speciescountsmay be broadlyresolvedby dependingless on size differencesbutmoreon therelativeabundance of flagellates anddiatoms. The question of what are the forcing components in determining the sequence of events, diatom + flagellate -+ diatom blooms, is to some extent answered by an examination of the nutrient and light regimes which were controlled during the course of the experiment. From the nitrate and ammonia data it may be seen A



. 4 C 201

i 10

. 15c

.

* .

l /? ,h i

L

10

20

30

40

50

. 60

DAYS

Fig. 1. Changes in phytoplankton 04 m: A, measurements made with a Coulter Counter; (0) particles of mean diameter 6.34 pm. (m) particles of mean diameter 32 pm: B. changes in the number of diatoms (0) and flagellates ( n ): C, changes in chlorophyll U.

Fig. 2. Environmental changes within the CEE. O-4 m: A. changes in light intensity: B. changes in nitrate and ammonia: nutrient additions shown as 1; upwelling of deep water shown as q . indicating 5 m3 of upwelled water added to the surface layer: C, changes in the silicate concentration.

SIMULATION

OF CHANGES

IN BLOOMS

289

(Fig. 2B) that after nitrate exhaustion by Day 5, the small amount of upwe~ling and sporadic nutrient additions up to Day 30, kept the nitrate and ammonia concentrations, in general, below 2 pg-at. NO, or NH,-N/l with a sporadic ‘overrun’ on both nutrients due to the combined effect of upwelling and nutrient additions on Days 15 to 20. The average nitrate and ammonia concentrations during the period between Day 5 and Day 30 were 0.34 and 0.7 pg-at NO, or NH,-N/l respectively. Since the measured K, of a mixture of Skeletonemu and Clzcretocwos spp. obtained from water outside the CEE was 0.6 pg-at. NO,-N/I (Harrison & Davis, pers. comm.), it may be presumed that diatoms were nitrogen limited at the ambient inorganic nitrogen levels within the CEE. It was found impossible to measure what was probably the lower K, value of the flagellate bloom which occurred by Day 30; from previous laboratory experience (Dugdale. 1976), however, the K, is often similar to the ambient nutrient concentration at steady state, i.e., 20.35 pg-at. NO1-N/l or, about one-half of that determined for the diatom bloom. Thus, based on estimated values of K, it is possible that flagellates can ‘out-compete’ diatoms when inor~dnic nitrogen concentrations are very Tow. Alternatively, it may be argued that the flagellate bloom may have been utilizing dissolved organic nitrogen derived from the breakdown of the earlier diatom bloom. Certainly the presence of colourless flagellates among the flagellates causing the chlorophyll maximum at this time would be indicative of organic substrates being present for various kinds of heterotrophic growth. The general intention during this initial 30-day period was to keep the background nitrogen concentration at about 0.5 pg-at. NO,-N/I while supplying a continual input of nitrogen sufftcient to sustain a phytoplankton bloom. In this respect it was also necessary to maintain a level of silicate which would not prevent the potential growth of diatoms. This level was estimated to be in excess of 3 pg-at. SiO,-Si/l based on half-saturation constants for silicate uptake by diatoms of about 0.3 ,ug-at. SiO,-Sijl (Goering, Nelson & Carter, 1973; Paasche, 1973; Conway & Harrison, 1977). The average silicate concentration during the first 30 days was 4.6 &g-at. SiO,-Sijl (Fig. 2C). Differences in the light available to the phytop~ankton contained in the CEE during May and June are shown in Fig. 3. The average iight intensity, O-4 m, for the first 35 days, was 0.088 gcal cm-’ min-l and after the light screen was introduced 0.018 gcal cm-? min-I. The K, and I’,,,, values of the phytoplankton populations in the CEE are shown in Fig. 3. From these data it may be seen that the K, of the total phytoplankton fraction was nearly always lower than that of the fraction < 35 pm diameter (mean total value of 0.031 compared with mean small particle value of 0.047 gcal cm-’ min-‘f. This indicates that the smaller fraction, which was composed mostly of flagellates, in general had a higher light response than the total fraction which included all the larger diatoms. Thus, by lowering the light intensity in the water column &4 m, an advantage would have been given to the diatoms, which had virtually disappeared (Fig. lB, less than lo/ml) on Day 33.

290

T. R. PARSONS, P. J. ffARRISON

AlVD R. WATERS

In these experiments the flagellates which grew well at light intensities of 0.088 gcal

cm -2 min -I would be considered ‘high light’ organisms; species which were present among this group included Gony~u~~~c~t~~ell~,as well as many small flagellates of the genus Ch~yso~h~o~ll~jnu and some colourless flagellates. The diatoms which grew in response to the lower light regime (Days 36 to 60) included the species Le~t~cy~j~dru~ d~i~ic~~ Cleve, ~~~~~t~ce~o~~ debits (Ehrenberg) Cleve and C. suci~l~ (Ehrenberg) Lauder, Navicula sp., Skeletonma costatm (Greville) Cleve, Thalussiosira and Coscinodiscusspp., most of which are generally found under temperate spring bloom conditions. The final small bloom of flagellates (Fig. IS, ~Day 50) was composed almost entirely of colourless flagellates.

.

1

I

I

to

20

1

I

I

,

30

40

50

60

OAYS Fig. 3. Values of P,,,

(A) and k; (B) for total phytoplankton fm) and p~ytop~ankton passing through a 35 pm diameter net (a).

From the data in Fig. 3B it may also be seen that the P,,,,, for the total and ~35 pm size fractions do not show any consistent difference over most of the period of the experiment. These results may, however, be misleading in respect to the 24-h growth potential of the diatom and flagellate fraction. The reason for this would have to be attributed to the 4-h incubation period on which the photosynthes~slight curves were based. This 4-h incubation is a good instantaneous measure of

SIMULATION

OF CHANGES

IN BLOOMS

291

the in situ light response (i.e., the K,) of phytoplankton, but may not be a good time period over which to measure the true P,,,,X.For example, Paerl & Mackenzie (1977) have shown that while small nanoplanktonic organisms have a high initial photosynthetic rate, this rate is not sustained over the whole day, and furthermore, there are apparently high losses of 14C during darkness among nanoplankton as compared with the net plankton. Ecological evidence of a similar phenomenon is given by Durbin, Krawiec & Smayda (1975); in the seasonal succession in a coastal phytop~ankton community, they showed that net plankton (consisting predominantIy of diatoms) had higher assimilation numbers than nanoplankton (containing the flagellate community). Furthermore, in the specific case of diatom as compared with dinoflagellate growth rates, Chan (in press) has shown that the light saturated growth rates of diatoms was two to three times greater than dinoflagellates over a wide range of cell size. Thus, it appears that under favourdble conditions the diatom community has an ability to divide either more efficiently (i.r., with less cell loss), or more rapidly (i.e., with a higher sustained P,,, over daylight hours) than the flagellate commLlnity. Three effects which could also have influenced the above size distribution of phytoplankton in the CEEs are temperature, upwelling, and zooplankton grazing. Temperature showed a gradual increase of x,4 C over 60 days. Since a diatom bloom was evident when the water temperature was z 15 C, and since a second bloom occurred later at x20 C, the effect of temperature was not considered to be a controlling factor in these experiments although, under other circumstances, there could be an effect. Upwelling was controlled by the bubble chamber at 5 m, and so this was constant throughout the experiment. Total zooplankton decreased throughout the experiment from an average maximum of x8000/m’ to a minimum on Day 50 of =2000/m’, followed by a slight increase to x 3000/m’ by Day 60. These changes are out of phase with the diatom -+ flagellate + diatom bloom indicated in Fig. 1 and so it may be assumed that the zooplankton were not, in these experiments, exerting a controlling influence on the main features of phytoplankton succession; it appears that the zooplankton may have been responding to the availability of diatoms with a phase lag of about 20 days. In summary, therefore, the sequence of events in the CEE may be explained as follows. Diatoms decreased in the first 30 days due to Iow nitrogen concentrations, while flagellates were able to grow well at these same concentrations, or by utilizing organic substrates left by the dying diatoms. The flagellates also tolerated high light intensities during this period. After Day 30, when nitrate was increased and light was reduced to about 20% of its previous value, diatoms increased from Day 35 to the end of the experiment. A second small flagellate bloom which started to grow at the end of the experiment (xDay 50) was composed almost entirely of colourless flagellates. In an earlier paper (Parsons & Takahashi, 1973) we suggested that there would be a dominance of large-celled phytoplankton over smaller-celled phytoplankton

292

P. R. PARSONS,

P. J. HARRISON

AND

R. WATERS

in waters having nitrate concentrations > w 2.2 pg-at. NO,-N/l and light intensities > x 0.01 gcal cm-?min -I and, conversely, that small phytoplankton would dominate in low light and low nutrient environments. The present experiments require modification of these suggestions since the environment studied was by comparison a ‘high light’ regime which, at the same time. produced small and large flagellates.

Fig. 4. Hypothetical

explanation

for maximum growth of diatoms and nutrients.

under

certain

conditions

of light

The most general conclusion which may now be drawn on the basis of previous data and the present experiments is represented in Fig. 4; there may be a diatom ‘corridor’ of maximum growth leading to diatom blooms in the ocean and surrounding this corridor is an area in which flagellate ecology will predominate, including both small and large cells. Since diatoms are only represented by one class of algae (the Bacillariophyceae) it is not too surprising to find a relatively narrow region in which they might predominate as compared with a large region of flagellate ecology since the latter are represented by several classes of marine algae including the Dinophyceae, Haptophyceae, Chrysophyceae, and Xanthophyceae. The key to diatom dominance appears not only to be a favourable region to growth of light and nutrients, but also the efficient and rapid multiplication of diatoms compared with flagellates when such growth conditions are satisfied. This has already been discussed and appears to be related to cell losses by flagellates either during cell division, to a high respiration rate, or to a higher intrinsic growth rate of diatoms over flagellates. While some classes of flagellates may grow at all combinations of light and nutrient regimes, there appears to be a definite region

SIMULATIONOFCHANGESINBLOOMS

293

where this growth is non-competitive with the much more efficient growth of diatoms. The restrictions based on diatom growth as shown in Fig. 4 have not been accurately quantified in this experiment with respect to the absolute values for nitrate uptake or the ideal light conditions: Fig. 4 is primarily intended as a guide to how two essentially different food chains in the ocean may originate. Thus, a diatom-based food chain is found in high nitrate environments except when at very high nitrates (such as due to agricultural fertilizers in river estuaries) there is a shortage of silicate, in which event, flagellates may predominate. In terms of light regimes, diatoms cannot compete with those flagellates capable of growing at very low light intensities. particularly if the latter are facultative heterotrophs or are colourless (obligate heterotrophs), as they may often be in the plankton. At higher light intensities (~0.01 gcal cm-? min-‘) it appears that diatoms have a distinct growth advantage over flagellates in terms of their 24-h growth rate. As the light intensity is increased still further, diatom growth appears to be generally inhibited and ‘high light’ flagellates again take over. This argument is, in part, supported by an early suggestion on the light response of different classes of algae (Ryther, 1956) in which diatoms were said to have a light response intermediate between low-light green flagellates and high-light dinoflagellates; however, more recent investigators (e.g., Dunstan, 1973; Chan, in press) have not been able to confirm this generality and the question must remain open to further irzsitu study. In this discussion the effects of temperature and differences in upwelling intensity have not been included because they were not part of the experimental manipulations; this does not mean that these processes are unimportant in the ocean as further factors governing the ratio of flagellates to diatoms in different environments.

ACKNOWLEDGEMENTS

The authors wish to acknowledge the support of the CEPEX staff in collecting samples and carrying out certain analyses. The National Research Council and the National Science Foundation are also gratefully acknowledged for their support of this project.

REFERENCES BOOTH, C. R.. 1976. The design and evaluation of a measurement system for photosynthetically active quantum scalar irradiance. Limnol. Oceanogr.. Vol.21, pp. 3266336. CHAN. A. T.,(in press). A comparative physiological study of diatoms and dinoflagellates in relation to light intensity and cell size - I. Continuous culture. Mar. Bioi. CONWAY. H. L. & P.J.HARRISON, 1977. Marine diatoms grown in chemostats under silicate or ammonium limitation. IV. Transient response of Chaetoceros dibilis, Skeletonemu costatum and Thalassiosira gravida to a single addition of the limiting nutrient. Mar. Biol., Vol. 43, pp. 3241. DUGDALE, R. C.. 1967. Nutrient limitation in the sea : dynamics, identification and significance. Limnol. Oceanogr.. Vol. 12, pp. 6855695.

294

T. R. PARSONS. P. J. HARRISON AND R. WATERS

DUGDALE,R.C., 1976. Nutrientmodelling.In, The Sea, Vol.6 edited by E.D. Goldberget al. J. Wiley, New York, pp. 789-806. DUNSTAN,W. M., 1973. A comparison of the photosynthesis - light intensity relationship in phylogenetically different marine microalgae. J. exp. mar. Biol. Ecol., Vol. 13, pp. 181-187. DURBIN,E. G., R. W. KRAWIEC& T. J. SMAYDA,1975. Seasonal studies on the relative importance of different size fractions of phytoplankton in Narragansett Bay (USA). Mar. Biol., Vol. 32, pp. 271-287. EPPLEY,R. W., J. M. ROGERS& J. J. MCCARTHY,1969. Half saturation constants for uptake of nitrate and ammonium by marine phytoplankton. Litnnol. Oceanogr.. Vol. 14. pp. 912-920. GOERING, J. J., D. M. NELSON & J.A. CARTER_1973. Silicic acid uptake by natural populations of marine phytoplankton. Deep-Sea Res.. Vol. 20, pp. 777-789. HECKY,R. E. & P. KILHAM.1974. Environmental control of phytoplankton cell size. Limnol. Oceanogr., Vol. 19, pp. 361-366. LAWS, E.A., 1975. The importance of respiration loss in controlling the size distribution of marine phytoplankton. Ecology, Vol. 56, pp. 419426. MALONE,T.C., 1975. Environmental control of phytoplankton cell size. Limnol. Oceanogr., Vol. 16, pp. 633-639. MENZEL, D. W. & J. CASE, 1977. Concept and design: controlled ecosystem pollution experiment. Bull. mar. Sci., Vol. 27, pp. 1-7. PAASCHE.E., 1973. Silicon and the ecology of marine plankton diatoms. II. Silicate-uptake kinetics in five diatom species. Mar. Biol., Vol. 19. pp. 262-269. PARSONS,T. R. & M. TAKAHASHI,1973. Environmental control of phytoplankton cell size. Limnol. Oceanogr., Vol. 18, pp. 511-515. PARSONS,T. R. & M. TAKAHASHI,1974. A rebuttal to the comment by Hecky and Kilham. Limnol. Oceunogr., Vol. 19, pp. 366-368. PAERI_H. W. & L. A. MACKENZIE,1977. A comparative study of the diurnal carbon fixation patterns of nanoplankton and net plankton. Limnol. Oceanogr., Vol. 22, pp. 732-738. RYTHER,J.H., 1956. Photosynthesis in the ocean as a function of light intensity. Limnol. Oceanogr., Vol. 1, pp. 61-70. SEMINA,H. J., 1972. The size of phytoplankton cells in the Pacific Ocean. In/. Revue ge.7. Hydrohiol., Bd 57, S. 177-205. SEMINA, H. J., LA. TARKHOVA& TRUONG NGOC AN, 1976. Different patterns of phytoplankton distribution, cell size, species composition and abundance. Mar. Biol., Vol. 37, pp. 389-395. STRICKLAND, J. D.H. & T. R. PARSONS,1971. A practical handbook of seawater analysis. Bull. Fish. Res. Bd Can.. No. 167, 310 pp. TAKAHASHI,M., W.H. THOMAS,D. L. R. SEIBERT,J. BEERS,P. KOELLER& T. R. PARSONS,1975. The replication of biological events in enclosed water columns. Arch. Hydrohiol., Bd 76, S. 5-23.